Comment Engine

Lenny Rachitsky shared in his recent job report that design jobs growth is down relative to PM + eng. I'll tell you my theory why: Design culture is broken in lots of companies. Often design teams & designers are the most resistant to change org in the EPD triad, with highly vocal AI opponents, and little skill or interest in the art of campaigning for influence or resources. Won’t hold a number like a PM, not yelled at about timelines like engineering. While I have brought design topics to the board convo, not a single board has pressed me our design talent, strategy, or velocity. Most teams treat design like a tax they don’t want to pay, and those that *do* take a deep interest and want to invest in design get back big “get out of my figma” energy. And if you’re too precious about craft to dirty your hands with the dark art of corporate politics, good luck getting more headcount. If a PM or engineer can get 85% there with tailwind and a dream, you better come to the table with more than “I represent the user.” Great designers are worth more than almost anyone on the team, and I’ve worked with lots of gems, but this is 0% surprising to me.

If AI can already write specs, summarize research, and prioritize backlogs… 😫 Then the real question for Product Managers is uncomfortable: What do you do that can’t be automated? Because the answer isn’t “more tools.” It’s better human capability. The PMs who will stand out aren’t the most efficient. They’re the ones who double down on what AI can’t replicate. Here’s how to actually build those traits: 1. Train your judgment (not just your analysis) Stop waiting for perfect data. Make small bets, reflect on outcomes, and sharpen your decision instincts. 2. Develop real customer empathy Get out of dashboards. Talk to users. Sit in their environment. Notice what they don’t say. 3. Practice narrative thinking Don’t just present facts—craft stories. If people don’t believe in the direction, the strategy doesn’t matter. 4. Learn to navigate people, not just problems Map incentives. Understand resistance. Great PMs move people, not just roadmaps. 5. Get comfortable with ambiguity If everything is clear, you’re operating too late in the process. Step into messy, undefined spaces—that’s where value is created. 6. Build conviction (and know when to drop it) Have a point of view. Defend it. But stay flexible enough to change when reality proves you wrong. 7. Master constraint-setting You don’t win by doing more. You win by choosing what not to do—and protecting that focus. 8. Invest in taste Care about quality. Details. Experience. The difference between “works” and “delights” is still deeply human. 9. Strengthen your coaching ability Your impact scales through others. Develop people, don’t just manage output. 10. Own outcomes, not outputs AI can recommend. You decide—and live with the consequences. AI will make product managers faster. But speed isn’t the advantage anymore. If you want to stay relevant, don’t compete with AI on execution. Compete where it can’t play 🥷🏻 What's your take on this?

After some reflection, I have isolated one of the reasons I am not a billionaire yet. I am sure there are many others. But the main one is probably my time management. I do not buy back enough of my time. I simply do not think like a billionaire. I mean, when was the last time Jeff Bezos went to the Post Office? Do you think Elon Musk takes his kids to school every morning? Does Sam Altman bring out the compost? Time for some deep reflection heading into Q2. I need to decide which type of man I want to be. Will report back. Feel free to make any suggestions in the comments...

So this week alone 5 leading AI Agent vendors asked us to try their apps. From public companies to new start-ups. Time is getting tight, and we already have 30+ AI Agents, so I tell everyone: "Can we loop back after SaaStr AI Annual in May?" We honestly have no time to deploy another AI Agent until then. All but one said, "Sure! Talk to you then!" But one said ... "You know what, give us 5 minutes to check in and we'll just deploy it entirely for you and get you up and running." And we did, and it's up and running, and it's great. More on that AI agent vendor soon. They are already up on our SaaStr AI Agents page. The rest? Well, see ya in June!! If we have time then. Very seriously: this is the AI Age. You have to help your customer deploy the AI Agents, and if possible, just deploy them for your customers. If you are selling like it's 2023, I mean ... good luck winning with AI Agents. Good luck.

We just publicly launched Zillow AI mode. I got to walk investors through our AI strategy yesterday at Zillow's AI Investor Day, and then my team demo'd the products live. It was a great day. Everyone sees the surface layer of AI in real estate: ask a question about a home or a neighborhood, get a smart answer. That part matters. But any company with access to public content and an API key can build that. The iceberg is everything underneath. In the same conversation on Zillow, you can check what you can actually afford, compare homes against your commute, schedule a tour, connect with an agent, start the pre-approval process, get deep advice on what trade-offs you need to make from our rich media. One continuous flow, built on years of platform infrastructure connecting search, touring, financing, media, and transactions. Getting surface-level AI right is necessary. But coordinating end-to-end actions across a regulated, high-stakes process where someone is making the biggest financial decision of their life? That's the genuinely hard part. And it's what keeps showing up in how people actually use this. They don't just want easy answers. They want to move forward. Same principle on the professional side. We showed updates to Follow Up Boss and our agent tools that use AI to surface buyer intent signals earlier, draft messages, and prioritize their time so agents connect with the right person at the right time. Consumers show up more prepared, agents engage at the right moment. That's the end to end transaction I've been at Zillow 9 years building AI. Yesterday felt like a milestone for the team. Proud of what they built and how they showed it.

Narrative violation: PM openings are at the highest levels we’ve seen in over 3 years There are over 7,300 open PM roles at tech companies globally, and trending up. This is 75% above the low we saw in early 2023, and already up nearly 20% since the start of this year. Today we have the most open PM roles we’ve seen since 2022. You can see all of these open roles here: https://trueup.io/product Full report here: https://lnkd.in/gHPzuDJa

A career high point followed by months of toil!🎢 It's been 4 months since the highlight of my working career and our incredible win at the Prolific North Tech Awards.🏆 Since then, it has arguably been the hardest few months of my career. Why? Growth📈 As a boot-strapped entrepreneur, scaling is a double-edged sword. In 18 months we'd quadrupled monthly revenue and our team, however the growing pains it brought with it, hit hard: ▪️ Processes - What worked once before, wasn't as reliable with scale. ▪️ Standardisation - We couldn't rely on heroics from individuals, we needed a consistent methodology. ▪️ Unwinding Decisions - What we thought was the right thing to do, months later had to be undone What did I learn? You cannot grow purely on momentum. If you don't consolidate and stabilise your foundations, growth can become crippling. 🧱 So, what did we do? ▪️Narrowed the focus - fewer target markets, fewer products, more expertise ▪️ Streamlined everything - consistency in our delivery has binned unnecessary operational cost ▪️ A new mantra - Our "Get it right first time" framework delivers projects to customer need, on time and to budget. We have stopped swaying from our proven process for the pursuit of our next deal. As we approach our Anumana's 5th birthday next week, I have never been prouder. 🎂 As a team, we have navigated a difficult period and built a framework "The Anumana Way" that actually gives our team the freedom to excel in spite of the structure. I've more excited than ever for the future, even if I'm a little more guarded about growth! #WhatATeam! #EntrepreneurLife #Growth

Your revenue is in Stripe. Your pipeline is in Salesforce. Your product data is in Postgres. And the only person who knew why “expansion_mrr” ≠ “expansion_mrr_final_final” left six months ago. Meanwhile, your team keeps shipping AI features. RevOps is still reconciling three versions of the same metric. Leadership is still asking why dashboards don’t line up. Data still gets pinged for “quick” questions. No one agrees on the numbers. Managing the data an organization actually runs on has quietly gotten harder, not easier. Structify just launched Data Maps. It's a system-level map of how your business actually works: • How revenue really flows (not how it’s labeled) • How product usage actually drives expansion • How your GTM motion behaves in practice No more "which number is right?" debates - go check it out: https://www.structify.ai/

Dashboards that talk back. Literally. 💬 We just launched a conversational AI layer for Power BI — ask your data a question, get an instant answer. We've made a short video showing exactly how it works. Watch it, try to poke holes in it, and tell us what you think — we mean it. 👉 Kindly find the link to the product page and video in the comments. Your feedback is the roadmap. 🙏 #PowerBI #ConversationalAI #DataAnalytics #Innovation

This one skill built my entire $7M company. Two years later, that same skill almost killed it. For two years, I was StreetTalk. I filmed every interview, took every sales call, and ran every product to the park myself. 30,000 steps a day, nine hours straight, then uploading footage at the Apple Store at midnight because I didn't have WiFi. It worked. $50K months by myself at 21. I almost just kept going. But I started noticing something. If I got sick, revenue dropped. A day off meant nothing got filmed. Clients waited when I was shooting. I wasn't building a company. I was the company. The hustle that got me here was the same thing keeping me stuck. So I had to learn a completely different skill set. Not outworking people. Out-thinking them. I brought on partners who knew how to build what I didn't. Luke built the systems I never could. Jesse brought 17 years of scaling agencies from $2M to $250M. Then I started delegating everything that wasn't me at my best. I don't touch day-to-day production anymore. The business runs without me on the street. That was the hardest part. Letting go of the thing I was best at because it was no longer the most valuable use of my time. Now I spend my days on the problems that actually move the needle. New product lines, new cities, the stuff that gets us from $7M to $70M. Getting to $7M was about showing up every single day and outworking everyone in the room. Getting to $60M is about building something that works when I'm not in the room. If you're a founder doing everything yourself and it's working, pay attention. That's the most dangerous place to be. At some point you have to stop being the engine and start building one. We're looking for brands that want to work with the engine.

I've posted maybe five times (I have reposted aplenty) in the last two years on LinkedIn. Not because I didn't have anything to say. Because I was heads down working and convinced that talking about it was somehow less important than doing it (Caribbean work ethic, thanks, Dad!). That was the wrong call. In the last two years, I've been CPO and a fractional product leader across three continents simultaneously. I've helped a company in Italy transform from a services business into an AI product company with Fortune 500 distribution deals. I've been CPO of a platform in Brazil, trying to give first-generation entrepreneurs access to the capital that the system was designed to keep away from them. I've placed a senior fractional PM into a business in under 72 hours when a founder needed someone who could hit the ground running immediately. I've made good decisions. I've made expensive mistakes. I've learned things I didn't expect to learn. I'm going to start sharing what I know. Some of it will be useful. Some of it will be wrong. All of it will be honest — no ghostwritten takes, no content strategy. Just what I've actually learned. If you're a founder or CPO trying to figure out how to build a product function that actually works — I think we'll have a lot to talk about. www.produx.tech

I've sat in rooms where creator marketing was an afterthought. A line item. A "nice to have." The companies winning aren't in those rooms anymore. At YouTube, I was a creator partner manager — the bridge between what creators needed and what the platform built. At Meta, I was on product marketing, translating creator insights into roadmap priorities. Same goal from different seats: make creators core to the business, not an afterthought. Owned distribution is getting harder. Creators already have it. The difference between "experimenting with creators" and "building creator infrastructure" comes down to where creators sit in the org — and who owns them. This applies whether you're selling sneakers or software. The playbook scales. 𝗕𝗿𝗮𝗻𝗱: 𝗖𝗿𝗲𝗮𝘁𝗼𝗿𝘀 𝗯𝗲𝗰𝗼𝗺𝗲 𝘁𝗵𝗲 𝘃𝗼𝗶𝗰𝗲, 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 𝗮𝗺𝗽𝗹𝗶𝗳𝗶𝗲𝗿𝘀 Starbucks embedded two creators into their core social strategy for 12-month partnerships. Not campaigns. Infrastructure. Ulta turned employees into creators when most companies keep their workforce silent on social. → The shift: creators shape messaging from day one, not just distribute it. 𝗚𝗿𝗼𝘄𝘁𝗵: 𝗖𝗿𝗲𝗮𝘁𝗼𝗿 𝗰𝗼𝗻𝘁𝗲𝗻𝘁 𝗳𝗲𝗲𝗱𝘀 𝘁𝗵𝗲 𝗳𝘂𝗻𝗻𝗲𝗹 Lowe's built an affiliate model rewarding creators for conversions over time. HubSpot signs creators to 3-placement minimums, then extends to 12 months based on performance. Why? Multiple touchpoints build familiarity. One post doesn't build trust. Consistency does. → The shift: creator content fuels paid media. Whitelisted ads. UGC in performance campaigns. Distribution you can measure. 𝗣𝗿𝗼𝗱𝘂𝗰𝘁: 𝗖𝗿𝗲𝗮𝘁𝗼𝗿 𝗶𝗻𝘀𝗶𝗴𝗵𝘁𝘀 𝗶𝗻𝗳𝗼𝗿𝗺 𝘁𝗵𝗲 𝗿𝗼𝗮𝗱𝗺𝗮𝗽 Steve Blank's customer development framework: build products by staying close to customers. Creators are the ultimate proxy. They're in constant conversation with your target audience — they know what resonates before your research team does. At YouTube, creator feedback directly shaped feature priorities. Creators weren't just users — they were the product's most important signal. → The shift: creators aren't just marketing. They're R&D. 𝗪𝗵𝗮𝘁 𝗺𝗼𝘀𝘁 𝗰𝗼𝗺𝗽𝗮𝗻𝗶𝗲𝘀 𝗺𝗶𝘀𝘀: They think creator partnerships are about distribution. They're actually about co-creation. To benefit fully, companies need to tie creator actions to business results — revenue, retention, acquisition. Something I've been wondering: should I also be talking to the CRO? If creators drive conversions, the revenue leader needs a seat at the table. Short-term: Write checks, get content. Long-term: Build infrastructure, own the category. This isn't a social media function. It's a creator-led media and growth performance engine. I don't have all the answers here — but I know the companies treating this like a side project are already behind. Who owns this at your company? #CreatorEconomy #ContentStrategy

Anthropic shipped a feature called /dream yesterday. Claude Code can now periodically stop, reflect on everything it's learned about your project, and synthesize it into higher-level insights. the community response was... medicinal. "OK well now we need /acid to handle all of it's hallucinations" - 681 upvotes then came /xanax for when it panics mid-refactor /therapy for when it gaslights you about a bug it introduced /rehab for people who can't stop typing --dangerously-skip-permissions and /shit to clean up AI-generated code anthropic's product roadmap is apparently a pharmacy now. 1,675 upvotes. 287 comments. the thread turned into the best comedy writing room on the internet for about four hours. meanwhile. the same day /dream shipped, half the subreddit was in full meltdown over usage limits. people on the $200 Max plan burning through their entire quota in two prompts. six separate complaint posts hit the front page simultaneously. "here's an incredible new feature" while your existing features are on fire. timing is everything. but buried under the jokes and the outrage was the actual story of the day. a deaf developer built a terminal flash notification plugin for Claude Code. pulses your terminal background when Claude finishes a turn or waits for input. because nobody else was building accessibility tooling for AI dev tools. a doctor built a website. a 73-year-old cardiac patient built a health app. someone built a 122,000-line trading simulator. I track 180+ posts across 5 subreddits every night and write up what actually matters. the daily digest auto-publishes at midnight. zero editing. fully automated pipeline. this is issue 8. full writeup in cb Shawn Tenam ⚡️....building with Ai that can now report on AI and sound better then your favorite creator updates 😉

I used to be proud calling myself a “growth hacker”. Yesterday my post went viral. I shared how we turned obsession into a playbook that grew 2 companies from $0 → $10M ARR. A lot of people asked about one thing: why i insisted on doing customer support myself. I used to think growth was about hacks. There are hacks. sure. some really work. but if you only see growth as hacks, you lose the game. 1️⃣ 🛞 Growth is not just output. It’s a loop. Founders try to “hire growth.” They bring in a content creator, a social media manager, etc. Looks legit, right? There’s a dangerous trap. Take influencer marketing. a typical flow: find 50 creators → send 45 outreaches → get 20 replies → close 15 deal → campaign runs → next cycle but look closer. one creator asks: “what’s your actual USP?” another says: “i tried your new video model. looks better, but consistency still breaks.” most teams move on. we don’t. we ask: - where did it break? - what prompt did you use? - what did you want instead? Suddenly, your marketing angle changes. Your product roadmap changes. Same signals, completely different outcome. That’s the loop. growth is not just output. Break the loop, you lose. 2️⃣ 🪐 Growth is a long-term game The highest ROI growth doesn’t show up immediately. - branding - SEO - product these compound quietly. Take branding. Branding isn’t just big moments. It’s so many small, even tiny, things. Your bug notifications, your weekly marketing emails, your... support ticket replies. Every touchpoint teaches users what to believe about you. You won’t see it in a dashboard tomorrow. but give it months, it stacks. Then one day, it looks like “sudden growth.” I don’t have a clean way to quantify this. a lot of it is intuition. But if you stay in the loop long enough, you start to feel it. None of this is hacky, none of this is fast. But this is how you actually win. 🎙️Curious how others think about this. especially where you disagree.

Strong lineup for our next Boston Generative AI Meetup on April 8. This one (#21) is focused on Coding and Engineering with AI — and the conversation is shifting. Not “Can AI write code?” But: What happens to the engineering org when it does? Over the past three years, this meetup has grown to nearly 6,000 members and become the largest continuously operating in-person AI community in the world. What matters most, though, is the room. Builders. Engineers. Founders. Operators. People actually working with these systems every day. For this session, we’ve pulled together a group that reflects that: • Alden Do Rosario, CEO and Founder, CustomGPT.ai — building production AI agents on proprietary data • Scott Weller, CTO and Cofounder, EnFi, Inc — founder/operator scaling agent-based systems and companies • Matt Flaherty, Director of Engineering Maven AGI — engineering leadership across AI, security, and product systems These are people building real systems — with real constraints — inside real companies. The format stays the same: moderated, audience-driven, and followed by strong networking. 📍 Microsoft AI NERD, Cambridge 📅 Wednesday, April 8 🕕 6:00 PM doors | ~7:00 PM panel 🍽 Food + drinks RSVP in the comments, and thanks to Squark AI and Microsoft for sponsoring. If you’re building with AI — this is one of the best rooms in Boston to be in. #BostonGenerativeAI #CodingWithAI #AIEngineering #BostonTech

For most of my life, I told myself I was a backstage director. The one who builds, not the one who presents. I wore it like a badge: "I let the work speak for itself" It took me years to realize that was not a philosophy. It was a hiding spot. #TheComfortableLie I'm creative. I have ideas that keep me up at night. I have proven what I'm capable of, and have the resources and people setting me up for success. But there is always this last mile. One more feature, one more tweak, and one more reason to convince myself and everyone else that it is "not ready yet." That's not perfectionism. That's fear wearing a product roadmap as a disguise. #TheCoFounderMirror One of the best things about my co-founder is that she sees right through it. Her question is never "what if it fails?" It is "what if it doesn't, and you never tried?" We're approaching something big right now. Something that matters for a company at our stage. And I had the usual urge to add just one more thing. She looked at me and said, "I don't care if you're ready. I don't care if the product's ready. We go out, we face it, we learn, and we come back and build again." She's been right about this more times than I'd like to admit. And she's not the only one. Past mentors and peers have told me the same thing. It just lands differently when it comes from someone building alongside you every day. #TheRealAudience Here's what I've learned. The people who believe in you will believe in your rough draft. The ones who don't trust you at your earliest won't trust you at your most polished either. And critics show up at every stage regardless. Both add growth. Supporters give you energy. Critics give you edges. But you have to be in front of both to get either. If you're a builder who keeps hiding behind "one more feature," I see you. I am you. The stage is not as scary as the story we tell ourselves about it. Ship it. Say the thing. Launch the page. Get on the call. Whatever your version of "front stage" is, the audience is already waiting. What's one thing you've been holding back because it doesn't feel "ready enough"? #ReveriextRhythm #FounderJourney #BuildInPublic #Startups #AI #Leadership #LearningInPublic #ShipIt #Entrepreneurship

Introducing Next Weeks Founders Coffee Speaker Klaudia McDonald Founder and COO of Bobo a platform that gives families and clinicians a shared, real-time view of health between visits. This talk explores how to build defensible, non-obvious GTM strategies tailored to your product, user psychology, and distribution specifically in complex spaces like healthcare and multi-stakeholder ecosystems. Drawing from experience scaling a pediatric digital health platform, Klaudia will cover how to: - Distinguish between buyers, users, and influencers -Build compounding GTM loops, not one-off campaigns -Turn partnerships into real distribution channels -Convert credibility into revenue -Shift from marketing spend to operational growth engines We’ll also challenge common advice like relying on paid ads or product-led growth and explain when those approaches fall short. This talk is for founders, operators, and investors, this session offers a practical framework to turn strong products into consistent, scalable growth. Register today! 📅 Tuesday March 31st : 8–9 AM 📍 Launch Downtown: 200–185 Carlton Street 🎟️ RSVP: https://lnkd.in/d3kstRDS ☕️Founders Coffee Page: https://lnkd.in/g-ezwZ-R 📲Bobo App: https://lnkd.in/eeZ3Bppf This program is brought to you by: Presenting Sponsor - Futurpreneur - Caroline (. Founded & Hosted by Launch Coworking Space Led by 360 coaching Group - 360Coaching Fuelled by Muse Cafe

“Bad data” gets blamed a lot… but is it really the problem? At the CDO Inspired Summit, David Bruce from Financial Ombudsman Service led a thought-provoking fireside chat on “Bad Data Is Not the Problem: Why Transformation Fails and What Leaders Must Do Differently.” Challenging a narrative many organisations default to, David made a compelling case that data quality isn’t usually the root cause of stalled transformation — leadership, ownership, and decision-making are. The session explored how organisations often over-invest in tools while under-investing in accountability and clarity of purpose. And how, too often, data is treated as a by-product rather than a managed, strategic asset. The result? A breakdown in trust, slower delivery, and missed value. The key takeaway was clear: successful transformation isn’t just about fixing data — it’s about fixing how organisations think about ownership, responsibility, and decision-making. A refreshingly honest and insightful discussion. 👏 Thanks to David Bruce for a brilliant session. #CDOInspired #DataLeadership #DigitalTransformation #DataStrategy #Leadership #DataDriven #Networking #B2B #B2C #Data #CDO #London #Londonevents #Dataevents #Dataleaders #Inspiredbusinessmedia

Some observations on AI: 👨🏼‍✈️ Directors are often not the source of innovation, although they are responsible for it. Their days are packed with meetings and their process of creating new heavily relies on tens of years of experience on a space, which for most is brand new. Therefore they use outdated methods for creating AI strategy and do not have systematic AI-native way of thinking about the problem space. 💬 While you can still catch up quickly, discussion on AI in orgs can become difficult because of the knowledge disparity and skill gap between persons holding similar positions but having vast difference in adoption of AI. Nobody wants to admit that they cannot switch the brain to a new mode. Ideating new things is tough by nature to begin with, even with great people onboard, but the skill gap creates a lot of friction on roadmaps and planning. 👶🏻Interestingly many companies have laid off a lot of juniors - forward-thinking talented junior can be a great superpower for building with AI without having the package of old workflows, combined with seniors and their industry experience. 🧠Maybe 2 or 3 designers out of 10 get it naturally and learn how to learn. The more tech-savvy and better knowledge on how computers work learn AI workflows easier. Some can break the syntax barrier and have unicornish qualities, while of course to reach that you need to have knowledge about software architecture. AI combined with remote work has done many roles already obsolete. In case you are design manager or used to be one who does not get AI, you should be worried. You are basicly a very expensive meeting organizer on a virtual water cooler without domain experience. 🤔The problem space however stays the same, which is choosing the right business strategy and right thing to build. Many teams waste time and get lost in prompting instead of talking to the customers (or bots of their customers) and solving real problems. 🧑🏻‍💻Dev team sprints – Mondays for planning and Fridays for reflecting…that is 3 days of building compared to agents working for you while you are away and then reviewing the work when back. 🤖 JIRA…do not build with JIRA, switch to Linear or something better than Atlassian products. 🧩The most natural application in design seems to be design systems, since they are systematic by nature. If you already have a mature design system, making the DS AI-friendly still is not probably the bottleneck in your work, but it can speed up things in the creation phase.

Will AI replace people in support? The honest answer is yes, and no. AI can already diagnose issues, draft responses, and route tickets faster than a human ever could. It’s hard not to be impressed by what it can do. But here’s something I’ve been reflecting on lately: It’s just as easy to forget what AI can’t do. Some companies are laying off thousands of support staff and automating everything they can. If you view support purely as an expense, that probably looks like progress. But it’s a short-sighted strategy that ignores how customers actually behave. Most don’t care if a solution is “AI-enabled.” They just want the right answer, as quickly as possible. Sometimes that’s self-serve. Sometimes it’s documentation. And sometimes it’s a human who can investigate, troubleshoot, and think critically. The best support models orchestrate all of the above to create the fastest path to resolution. That’s why our most knowledgeable engineers stay close to customers on the front lines. Their expertise is earned through experience. It’s also how our next generation of experts learn: on the job, alongside seasoned professionals. The best support teams are built by training and investing in people. The results speak for themselves: Three out of four product lines had 100% customer satisfaction on support surveys last year. Because what customers want hasn’t changed: continuity, results, and speed. Most importantly, they want to feel understood. You deliver that best when AI and humans work together.

We launched in 2023 with one SKU. One product. No variants. Decision fatigue is one of the fastest ways to lose someone who is already interested. Choices create friction, especially early on. One product also meant we could learn without carrying significant inventory risk or raising capital. Instead of chasing new products or complex logistics, we obsessed over quality and customer experience. We improved everything around the core product before adding anything new. We figured out what our brand stood for and built a community. Now we're scaling. But the discipline of staying focused, staying simple, for those first few years is what made it possible. Simplicity isn't a weakness. It's a strategy.

This is a special conversation. I sat down with my friend and former colleague Daiva Staneikaite Naldal to recall the work that we did together, scaling LEGO® IDEAS from an innovation pilot into a full-fledged product line. Most companies say they listen to their customers. Very few build the system that converts listening into a growing program. Daiva and I were part of the team that built LEGO IDEAS, the platform where fans submit product designs, rally community support, and see their creations turned into real LEGO sets sold worldwide. It grew into a global business and major product line, with products you've no doubt seen or even own. The hard part wasn't the technology. It was integrating a disruptive innovation engine into an 80-year-old organization without killing the trust of the creators who powered it. Three things stuck with me from this conversation: First, your most dedicated users read the market before you do. LEGO fans spotted Minecraft as a cultural phenomenon before anyone inside the company. Second, open innovation isn't a marketing campaign you turn on and off. It's an ongoing commitment to relationships with top users that compounds over time. Third, commercial proof creates organizational pull. Match the core business's numbers and resistance turns into demand. Whether you're scaling a startup or running innovation inside a large org, the principles are the same. Define your strategic intent. Build the circle of trust with your lead users, and close the loop and reward them for captured value. And for good housekeeping, please note that views expressed by the host and guest are their own and do not represent those of any current or former employer. LEGO® is a trademark of the LEGO Group of companies which does not sponsor, authorize or endorse this podcast. Links to the full episode in the comments below. ⬇️

I’m pleased to publish my first whitepaper! Measuring What Matters is a framework I’ve been developing for measuring success from AI and Automation projects. An MIT study published that “95% of generative AI pilots delivered zero measurable financial return” last year. Not because the projects failed. Because the measurement was wrong. I’ve run into this first hand. Leadership ask “is this working?” and traditional metrics could not answer the question. So I wrote this framework. Seven dimensions of success. From delivery performance through to strategic impact. Built around a concept called “Value Moments”, the specific interactions where your product actually delivers value. The concept is built for a small team delivering value large scale projects at pace. It might not suit all businesses but I hope it can provide some great pointers and conversation starters. The full whitepaper is a free PDF on my blog. Swipe through for the key ideas are grab the full document below. https://lnkd.in/eMHWndFC Let me know what you think!

By 2026, we need to decide what AI actually is to our work. Build with it, enable it, power everything through it. Most organizations haven't picked a direction yet. Some are scaling into production. Others have engineering picking one path, ops picking another, leadership picking a third. Some are claiming they're AI-ready without the execution. The rest haven't even started. The question isn't whether AI matters. It's what role you want it to play in your next move. Clarity always changes everything: who you hire, decision-making, and how work gets done. Same applies whether you're building product, driving revenue, managing operations, or leading the company. What does AI actually mean for your work right now?

A founder asked me last week: "Should I hire a full-time CTO or go fractional?" Here is the math. Full-time CTO: • $180-250K salary + equity • 3-6 months to find the right one • Locked in whether the product pivots or not • Still needs a team under them Fractional CTO: • $10-20K/month • Starts next week • Scales up or down as the product evolves • Brings a team that already ships together One of our clients came to us after burning $400K on a full-time technical co-founder who built the wrong architecture. We rebuilt their MVP in 8 weeks for a fraction of that. They are now at $30K MRR and growing. The pattern I keep seeing: Non-technical founders do not need a CTO on payroll. They need technical leadership that matches their stage. Seed stage? Fractional. Pre-product-market-fit? Fractional. Scaling past $1M ARR with a proven product? Now hire. Stop over-hiring for where you want to be. Build for where you are.

I've been thinking about the line between cognitive offloading and cognitive surrender. We've always offloaded cognition. Calculators. Spreadsheets. Spellcheck. Nobody hand-calculates compound interest anymore. That's fine. Offloading the mechanical frees up brain space for the stuff that matters – judgment, creativity, pattern recognition. But something is different now. When AI writes your email, drafts or edits your strategy, prioritises your roadmap, codes your product... where's the line between "this frees me up to think deeper" & "I've stopped thinking altogether"? I know lots of folks thinking the same. The distinction isn't the tool. It's the level of handoff. Offloading is delegation. You know what good looks like. You could do it yourself, but the tool saves time. You review the output with judgment intact. The cognitive load shifts, but the thinking doesn't disappear. Surrender is abdication. You don't know what good looks like anymore. You trust the output because you can't evaluate it. The thinking disappears entirely. The space between those two is the question. I've seen founders use AI to draft pitch decks without understanding their own unit economics. Operators shipping features they can't debug because they didn't write the code. Not because they're lazy – because the tool made it so easy that learning felt inefficient. Here's an uncomfortable question: maybe that's fine? Maybe we're evolving past needing to understand every layer. Maybe "knowing how to evaluate AI output" is the skill, not "knowing how to do the thing." But I worry about the learning curve. You don't develop taste by watching someone else cook. You develop it by burning things, under seasoning things, figuring out what works through trial and error. If you never write bad code, how do you recognize good code? If you never draft a messy strategy doc, how do you know when the AI's strategy is actually coherent vs just well-formatted nonsense? The danger isn't that AI makes us lazy. It's that it makes us incompetent without realising it. You think you understand something because you can generate output. But understanding and generating aren't the same thing. Not even close. So here's my current heuristic, for what it's worth: - Offload the repetitive. Keep the generative. - Let AI handle patterns you already understand. Things you've done enough times that you know what good looks like. Where you can spot when the output is wrong or off. - Do the new stuff yourself. At least until you've built enough judgment to delegate it. Until you can look at AI output and immediately know if it's right or garbage. Not a rule. Just where I'm landing for now. The shift is real. The tools are incredible. But capability without judgment is just expensive randomness. We're all figuring out where that line is in real time. What's your line? Where are you offloading vs surrendering, and how are you deciding? Genuinely curious how others are navigating this.

I spoke with an AI consultant today who builds out openclaw for brands. I demo'd him Profasee Ultra. You could see the moment the entire conversation changed. Because what we built replaces what he sells. He charges about $5,000 a month to build custom AI workflows. But here’s the problem: AI does not want to be a consulting project. AI wants to be a product. Companies don’t want someone to build agents. They want to hire an agent. Log in. Pick PPC Agent. Pick Pricing Agent. Pick Inventory Agent. Turn it on. Done. No project plan. No roadmap. No invoices. No six-week implementation. Just outcomes. This is the shift happening right now: We are moving from “We build AI for your company” to “Your company runs on AI.” One is services. One is software. One charges retainers. One replaces agencies and headcount. And when this shift happens, industries don’t slowly decline. They disappear. The future is not custom AI. The future is AI that just works. Out of the box. Like hiring an employee. Just my $0.02 for the day

Data Beats Opinions. But it is the People that Bring that Data to Life. I’ve sat in too many rooms where the loudest opinion wins. We have all seen it. And almost every time, the data was already telling an often different story. Here’s the truth: Data doesn’t replace judgment. It sharpens it. Recently, I was in a conversation where someone made a sweeping claim and backed it up with data. At first glance, the numbers supported their point. But when you zoom out, the story changes. Correlation is not causation. And data without context can be misleading. This shows up in business every single day. Are we losing shelf space because a competitor is outperforming us with a stronger TikTok presence and a $2 price advantage? Or is it because we’ve had inconsistent inventory, impacting sell-through? Both are “data points.” Only one is the actual problem to solve. Same goes for growth. Distribution is not the win. Sell-through is. I’ve seen teams celebrate new placements like victory… while the product sits getting all dusty on shelf. You know that return isn't going to be a win. The best teams I’ve worked with do this differently: They use data to guide the framework, not dictate the conclusion. Because dashboards don’t make decisions. People do. :) And the real magic happens when teams: • Look beyond a single metric • Ask better questions • Connect the dots across functions • And bring the story behind the numbers to life Also… if your ERP or reporting tools aren’t optimized, you don’t actually have data......You have noise. Data is powerful. But only in the hands of people who know how to interpret it. Clear is kind. Data helps us make good decisions faster, and smarter. #Leadership #DataDriven #CPG #ExecutiveLeadership #SalesStrategy #GrowthMindset

Nike asks this question to see if you crack under pressure. If you answer by saying you will just work all weekend alone, you have already failed the interview. When a hiring manager at Nike asks how you handle an overnight redesign that forces your team to cancel their weekend, they are not testing your stamina. They are testing your leadership reflexes under extreme pressure. This question is a staple for high-velocity roles like the Express Lane team. They need to know if you can pivot without losing your creative edge or burning out your team. Most candidates fall into two traps: the Martyr Complex, where they try to do everything alone, or Toxic Positivity, where they pretend the weekend cancellation is a fun opportunity. To answer this effectively, follow this framework: 1. SCOPE CONTROL: Do not blindly accept the mandate to start from zero. Identify what can be saved—like existing tech packs or silhouettes—to buy your team time. Focus the redesign only on what is essential. 2. TACTICAL EMPATHY: Acknowledge the frustration. A strong leader validates the team's sacrifice rather than sugarcoating it. If you have to cancel plans, be transparent, provide support, and ensure the work is clearly divided based on individual strengths. 3. QUANTIFY THE IMPACT: Do not just talk about the process. Connect the sprint to a business outcome. Did the final product clear inventory? Did it hit a specific sales goal? The interviewer needs to know that the sacrifice resulted in a high-quality, commercially viable product. Think of a time you faced a last-minute pivot. Did you protect your team's workflow while delivering a result that mattered? When you frame your answer around these three pillars, you show that you can maintain both operational focus and creative integrity when the clock is ticking. #InterviewTips #CareerAdvice #InterviewPreparation #ProductManagement

"When you hand a delivery team a list of tasks, you get a finished list. When you hand a team a problem to solve, you get a solution." 💡 This is the fundamental shift every engineering leader and product owner needs to make to move from "output" to "outcomes." Most backlogs are treated like grocery lists. We fill them with technical "to-dos": ❌ Update this API. ❌ Refactor that database. ❌ Change this UI element. We hand these lists to our teams and, if they’re good, they give us back exactly what we asked for: a finished list. But did we actually move the needle? Or did we just increase our velocity while standing still? 🏃‍♂️💨 When we stop managing tickets and start managing opportunities, the entire dynamic changes: ✨ Autonomy increases: Engineers aren't just coding; they’re architecting value. 🚀 Innovation happens: The team might find a better, faster way to solve a problem than the "feature" you originally had in mind. 🎯 Results matter: Success isn't "shipping the code"—it's "solving the pain point." Read the full article below. #Leadership #SoftwareEngineering #ProductManagement #Agile #BacklogStrategy #ProblemSolvers #Innovation

Most organisations already have security tools. Plenty of them. Dashboards. Alerts. Controls. Security products everywhere. But leadership still can’t answer one simple question. Where are we actually exposed? Security isn’t about buying more protection. It’s about understanding: What matters most. Where risk actually sits. What needs to improve next. Until that becomes clear, security stays reactive. Confidence only comes from clarity.

Your idea is worth nothing without immediate execution. Most founders spend months overthinking features that nobody wants. They treat software development like a marathon when it should be a series of sprints. In a fast moving market, being slow is the same as being wrong. The goal is not to build a perfect product on day one. The goal is to get into the hands of users before the opportunity vanishes. 💠 Speed gives you the data needed to pivot or scale with confidence. At Abnexa Technologies, we have replaced the traditional six month roadmap with a fourteen day execution model. We use high level human expertise and intelligent systems to deliver results while others are still in meetings. We do not just build code. We build momentum. 🔱 If you are ready to stop planning and start launching, reach out. #Execution #StartupGrowth #Abnexa

https://lnkd.in/eeCzaP-c Haha! Love the quote in the a16z article - "Enterprises buying AI are like your grandma getting an iPhone." The technology is powerful but some hand-holding is necessary to get the most out of it. (Will we become our parents where our kids will roll their eyes how we can not automate our lives with AI Agents?) For a decade, the gospel was product-led growth. Joe argues that during the AI platform shift, that gospel is wrong. AI products are powerful but work wonders only if your data/context is connected to the system. Someone needs to be in the room to connect the data and show the best way to take advantage of your product. We've been living this model for 10+ years at Lights On Software. Our engineers have embedded as Forward Deployed Engineers at Cohere, integrating their AI models at Slack, Microsoft, and Oracle. At Tenstorrent, building AI visualization and observability tools across a multi-year engagement. At Saris AI, connecting autonomous AI agents to core banking systems at major credit unions. Before FDE had a name, we did the same work at Comcast and Nasdaq. The pattern is always the same: the client's product team stays focused on the platform while our engineers handle the messy, customer-specific integration work in the field. No context switching. No pulling your best people off the roadmap. The product team builds the product. The FDEs get customers to production. Both get better because of each other. #ForwardDeployedEngineer #FDE #AIIntegration #EnterpriseAI

Most product roadmaps are still guesswork. Not because PMs are bad at their job but because the process forces them to guess. We run interviews collect feedback look at data But the hardest part? Turning all of that into a clear decision. That step is still: - manual - subjective - and hard to scale Now that AI can build the solution… 👉 why are we still guessing what to build? That’s the layer we’re working on.

Sora deja una lección incómoda para cualquier PM: descargas no son negocio. OpenAI cerró su app de video con IA apenas 6 meses después del lanzamiento. Poco después, Disney se bajó del acuerdo que se había anunciado alrededor del producto. ¿El problema central? Compute. Sora logró distribución, ruido y atención. Pero no logró justificar el costo de sostenerlo frente a prioridades más rentables como GPT-5.4 y el negocio enterprise. Eso deja cuatro aprendizajes brutales: Product-market fit no son millones de descargas. En IA, priorizar producto es priorizar compute. Las alianzas estratégicas no compensan un producto inestable. El video con IA para consumidores sigue sin demostrar un modelo sostenible a escala. La gran pregunta no es por qué murió Sora. La gran pregunta es: ¿qué producto de tu roadmap hoy parece prometedor, pero en realidad está quemando más recursos de los que devuelve? → Variety — OpenAI cierra Sora, Disney sale del deal de $1B: https://lnkd.in/eDxyVcEB → Hollywood Reporter — Disney sale del deal con OpenAI: https://lnkd.in/eXEtdT9d → Bloomberg — OpenAI discontinua Sora: https://lnkd.in/ePguwP8V → Axios — OpenAI estrecha enfoque: https://lnkd.in/ebGeywFr #IA #OpenAI #Sora #Disney #GestionDeProducto #EstrategiaDeProducto #IPO #GenAI

Alignment is one of the most underestimated leadership skills. Not because it’s easy. But because it’s invisible. You don’t see alignment in a roadmap. You see it in how quickly teams move. In how few decisions get revisited. In how clearly priorities flow across the organization. Without alignment: Teams move in different directions. Decisions take longer. Execution slows down. With alignment: Momentum builds. Trade-offs are clear. Execution accelerates. The best leaders don’t just set direction. They create alignment that makes execution possible. Tina McClelland Enterprise Product • Governance • Transformation Turning Strategy Into Execution Clarity over activity — alignment drives real outcomes

"AI will help us deliver features 10x faster." I keep hearing this. And I have to disagree. As a software engineer with 8 years across consulting clients, I've learned something: coding speed has never been the bottleneck. Here's what actually slows delivery: - Requirements that don't match what stakeholders actually want. We build something, present it, and hear "that's not what I meant." Engineers lose 30% of their time asking clarifying questions because requirements were vague from the start. - Building features because they sound good on a roadmap, not because users actually need them. - Mid-quarter priority shifts that derail commitments. The roadmap looked solid in planning, then week one hits and everything changes. - Knowledge silos. The engineer who built the billing calculator left six months ago. Now a bug in production takes days to troubleshoot because nobody understands how it works. - Technical debt from time pressure. "Just get it to market" means duplicated logic, missing tests, and infrastructure that doesn't scale. - Understaffed QA trying to test a quarter's worth of features in two weeks. These aren't coding problems. These are delivery problems. AI won't 10x your delivery if you can't define what to build. Changing how fast you build it won't help. But AI CAN solve the real bottlenecks: - Use it to create living documentation so knowledge doesn't leave when engineers do. - Use it to analyze Splunk logs and speed up troubleshooting. - Use it to help product managers write clearer requirements. Use it to generate architectural decision records so both humans and AI understand why decisions were made. - Use it to craft clear technical requirements so that engineers can work in parallel. The opportunity isn't in coding faster. It's in fixing the organizational dysfunction that's been slowing teams down for decades. This matters even more for B2B2C companies, where you're balancing business client needs with end-user experience. The complexity multiplies, requirements get harder to nail down, and the cost of building the wrong thing affects both your clients and their customers. Without the right processes from the start of the SDLC to the end, you don't speed up delivery with AI. You make your problems 10x worse, 10x faster. What's the biggest non-coding bottleneck slowing your team down right now? #SoftwareEngineering #EngineeringLeadership #AIinSoftware

Here is a question most engineering leaders cannot answer: "Which platform capabilities should we be telling customers about?" When engineering and marketing operate in isolation, marketing sells last year's capabilities while engineering builds next year's platform. The gap compounds over time. A monthly translation session between engineering and marketing addresses this directly. The format is simple: - Engineering presents 2-3 upcoming capabilities in business language. Not "we are migrating to event-driven architecture" but "we will process customer requests 10x faster with real-time responses." - Marketing shares 2-3 customer pain points they hear repeatedly. Often these map to capabilities that already exist but were never communicated. - Both sides identify one joint initiative: a capability that deserves marketing attention or a customer need that should influence the roadmap. This meeting takes 45 minutes monthly. The outcomes compound: - Marketing starts articulating technical differentiators in customer conversations - Customer pain points feed directly into engineering roadmap priorities - Product pages reflect actual platform capabilities instead of generic claims The insight: engineering and marketing are not separate functions. They are two halves of the same value delivery system. When they operate in isolation, both perform worse. Does your engineering team have a regular translation session with marketing? #CrossFunctional #PlatformArchitecture #EngineeringLeadership

A senior architect published something today that I've been hearing in every eCommerce leadership session this quarter. His field observation: the teams actually making progress with AI right now are not building AGI. They are fixing three things — product discovery, personalisation, and inventory operations. I have been in these exact rooms. The pattern is identical. The companies pulling ahead in eCommerce are not the ones with the most ambitious AI roadmaps. They are the ones who asked a harder question first: out of everything AI could do for us, which three things would actually change the business in the next 12 months if we got them right? That is what I call the Five Percent question applied to retail. Not five percent of the technology. Five percent of the decisions that drive 95% of the outcomes. The teams losing are running 30 AI initiatives simultaneously. The teams winning are all-in on three, and they can tell you exactly why those three. If your AI strategy has more than five clear targets for the year, it is probably a wish list. The path from wishing to winning starts with a harder question, not a longer roadmap. What are the three AI bets your eCommerce team is actually committed to this quarter? #AI #eCommerce #Retail #AIStrategy #Leadership

Lately, I’ve been spending a lot more time than usual dealing with customer escalations—and it’s a sharp reminder of how quickly enterprise roadmaps can shift. What looks like a well-planned quarter on paper can change overnight when a key customer hits a blocker or a deal depends on a critical ask. The real challenge isn’t avoiding these situations but finding the balance: - Being responsive to customer needs - Without completely derailing long-term priorities I’m learning (and re-learning) that not every escalation should turn into a roadmap change—but the important ones absolutely should. Striking that balance, while staying realistic about capacity and trade-offs, is where product leadership really gets tested. Curious—how are others managing this tension right now? Do you use any kind of framework to deal with such escalations?

𝗙𝗼𝗿 𝗲𝗮𝗿𝗹𝘆-𝘀𝘁𝗮𝗴𝗲 𝘀𝘁𝗮𝗿𝘁𝘂𝗽𝘀, 𝘁𝗵𝗶𝘀 𝗼𝗻𝗲, 𝘀𝗶𝗺𝗽𝗹𝗲 𝘀𝗵𝗶𝗳𝘁 𝘀𝘁𝗮𝗿𝘁𝘀 𝗺𝗼𝘃𝗶𝗻𝗴 𝘆𝗼𝘂 𝗳𝗿𝗼𝗺 𝘀𝗲𝗿𝘃𝗶𝗰𝗲 𝗽𝗿𝗼𝘃𝗶𝗱𝗲𝗿 𝘁𝗼 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝘁𝗵𝗲𝘆 𝘁𝗿𝘂𝘀𝘁. I was talking with a founder recently. Sharp, building something real for GTM teams and she said something I've heard a dozen times: "Customers want me to do it for them." At this stage, your early customers aren't really customers. They're co-builders. They shape your roadmap, pressure-test your GTM, and help you figure out what the product actually needs to be. So yea, leaning into a high-touch, hands-on model makes sense early. Build trust, move fast, learn fast. But it becomes a trap faster than you think: 1️⃣ You drift from founder work. Your value is thought leadership and relationships, not execution. 2️⃣ You lose product signal. When customers never touch the product themselves, you stop learning what's confusing, what breaks their flow, what they can't figure out without you. 3️⃣ You create a referenceability gap. Eventually a prospect will want to speak with a customer who actually uses the product. If your champions have only seen outputs, that's a hard gap to bridge. The goal isn't to do it for them. It's to build their confidence in the product. 💡 Simple fix: stop running strategy sessions. Start running working sessions. Screen-share. Build something live. Better yet, have them drive while you navigate.

Most teams don’t fail to reach the C‑suite, they fail to deserve the meeting. Harsh? Maybe. But if you lead a sales team, you’ve seen this play out in real time. And I’m curious where you think the real breakdown happens, because here’s what I see over and over: 1. Reps approach executives like they’re pitching managers, not decision‑makers. C‑suite buyers don’t care about: Features Roadmaps “We can help you with…” They care about: Risk Revenue Leverage Timing Tradeoffs If your team can’t speak that language, they get bounced to middle management every time. Where do you see reps losing altitude? 2. They ask for time instead of offering clarity. Executives don’t reward effort. They reward insight. If your outreach doesn’t instantly signal: “I understand the pressure you’re under and the decision you’re trying to make” you’re invisible. What’s the best / worst executive outreach you’ve seen lately? 3. They collapse under executive pressure. Execs push back. They challenge. They test conviction. Most reps respond with: Hedging Over‑explaining More product feature crap Apologizing Stupid questions like “Does that make sense?” C‑suite access requires clean language and clean thinking. No fluff. No wobble. Is this a skill issue or a confidence issue to you? 4. Your system trains them to stay small If your process is built around: Activity metrics Generic sequences Demo‑first motions Just follow up again …you’re unintentionally training reps to behave like SDRs, not strategic advisors. C‑suite access isn’t a personality trait. It’s a system outcome. What part of the system do you think breaks first? Here’s the truth Executives aren’t avoiding your team. They just don’t see a reason to talk to them… yet. Fix the narrative. Fix the discovery. Fix the coaching. Fix the altitude. And suddenly the C‑suite stops being a fortress and starts being a calendar event.

Most founders believe that the most challenging aspect of building a software product is execution. However, this is often not the case. The more difficult challenge lies in determining whether the product should be built at all. Many founders, consultants, and domain experts contemplate transforming their expertise into software, whether it be an automation tool, a workflow platform, or a SaaS product designed to replace manual tasks. Once the idea is conceived, the pressure to act swiftly intensifies. Teams seek direction, developers await instructions, and roadmaps begin to take shape. Indecision can feel perilous, leading to hasty decisions just to maintain momentum. However, speed does not equate to clarity. In numerous early-stage software projects, the real danger is not in delaying too long, but in approving the wrong product and committing extensive engineering resources before understanding the fundamentals. Once a decision is made, the pace quickens: developers commence building, product roadmaps are established, and budgets are allocated. By the time uncertainties arise, reversing the decision can be costly. This is why many unsuccessful SaaS products do not stem from poor execution but rather from decisions made before the product is adequately validated. Before embarking on software development, the crucial question is not, "Can we build this?" but rather, "Should this software exist in the first place?" How can you tell if your team is rushing to build a product before the evidence is clear? Reach out tp Lunagen for assistance in making better decisions. Book a meeting.

For years, social media platforms have positioned themselves as neutral infrastructure. They *just* host and distribute content. They don’t shape outcomes. Well, that argument is no longer valid. Today, a US jury has found Meta and YouTube liable for harm caused to a young user, awarding roughly $3 million in damages. The ruling concluded that these companies were negligent and failed to adequately warn users about the risks associated with their platforms. This outcome is notable and, more importantly, unprecedented, but what really matters is what the case focused on. This wasn’t a mere debate about content moderation. This was a debate about how these platforms are designed. Features like infinite scroll, autoplay and recommendation systems were positioned in court as tools that actively shape behaviour. Not just standard platform mechanics, but systems engineered to maximise engagement, which in turn positions them not as neutral channels, but as environments with built-in incentives and consequences. This case is the first of its kind to go to trial and succeed, and is highly unlikely to standalone with many more similar cases already in motion. From a marketing and product perspective, this introduces an entirely new question, because for the past decade, optimisation has been centred around audience attention: ☑️ increasing time spent ☑️ reducing friction ☑️ improving retention But if the means that drive attention are now open to legal scrutiny, then we need to take this shift into consideration and understand that these companies now need to assume responsibility in regards to the systems designed to influence behaviour. This doesn’t just apply to social platforms. It also applies to any product, marketplace or experience built on engagement, behaviours or optimisation at scale. The broader implication is clear: design is no longer neutral and “engagement at all costs” is starting to look less like strategy and more like liability. Read more: https://lnkd.in/dxCbXQyA

We don’t talk enough about PM & PO hybrid roles. Somewhere along the way, Product Manager and Product Owner became two separate jobs on paper. In practice, those lines blur fast. Especially in growing companies or transformation environments, you’re not choosing between strategy and execution. You’re doing both. Shaping direction with stakeholders in the morning. Refining edge cases and dependencies with engineering in the afternoon. You’re owning the roadmap. Prioritizing trade-offs. Writing stories. Aligning leadership. Driving delivery. Strategy meets execution. The role isn’t defined by title. It’s defined by ownership. And the teams that win are the ones with people who are comfortable operating across both.

AI products rarely break at the model layer. They break at the seams — where architecture hasn’t caught up with product demand.      In this special edition, Rahul Pahuja explores why some teams keep scaling AI in production while others end up rebuilding the foundations underneath it.

I used to think staff augmentation was just "renting developers." I was wrong. Here's what changed my perspective: I worked with a SaaS startup whose CTO was drowning. 4 open engineering roles. 2 product deadlines. 1 very stressed leader. They tried job boards. LinkedIn. Referrals. Weeks passed. The pipeline stayed empty. We placed 3 senior engineers within 10 days. Not juniors to train. Not freelancers who disappear. Senior engineers who slotted into their stack, attended standups, and shipped features, week one. That CTO told me: "I wish I had done this 6 months ago." Staff augmentation isn't a compromise. It's a competitive advantage, when used right. Here's what the best tech leaders use it for: → Bridging critical skill gaps mid-project → Scaling fast without bloating headcount → Testing new tech domains before committing to full hires → Meeting deadlines while long-term hiring continues Building a team or scaling a product in 2026? Let's connect, I help CTOs and tech leaders build high-performing teams without the hiring headaches. DM me or comment below. 👇 #StaffAugmentation #EngineeringLeadership #TechHiring #CTOLife #ProductDevelopment #TechTalent

I’ve spent the last several years operating at the intersection of science, systems, and digital product development in biotech. What I’ve learned is this: The biggest bottleneck in innovation is rarely science. It’s how systems, data, and teams connect. When workflows are unclear When systems don’t scale When data isn’t structured Speed disappears Decisions slow down Opportunities get missed My work has focused on solving that through: • Translating complex lab workflows into scalable digital systems • Driving adoption across cross-functional teams • Building infrastructure that supports real decision-making I’m now looking to apply this experience in environments where speed, clarity, and execution truly matter. If you’re building or scaling in biotech, digital health, or product-driven environments, I’d value the conversation.

The death of OpenAI’s Sora proved one thing: retention will either make or break your product. The news of the shutdown didn't surprise me. A few months ago, I saw the early data that signaled the end: Sora’s Day 30 retention dropped to less than 8% (per SensorTower/a16z), and Day 60 was nearly 0%. For context, a strong consumer app like TikTok maintains a 30%+ D30 retention. Sora wasn’t a product; it was a digital sightseeing tour. What was actually behind the low retention? 1/ The "try once" trap: Users wanted to see the magic, but they didn't have a reason to come back the next morning. 2/ No clear ICP or workflow: When you build for "everyone," you build for no one. Without a specific business workflow to slot into, it remained a toy. 3/ The broken retention loop: As soon as a video was finished, it was exported to Reels or TikTok. The platform was just a "content factory" that captured zero of the downstream social value. 4/ The burn vs. value: The math simply didn't work. Estimated compute costs hit $15M per day, while total lifetime consumer revenue barely crossed $2.1M. In my experience scaling growth on the ground, I’ve learned that top-line hype is a vanity metric. Retention is impossible to fake. If your cohorts are not sticky and the unit economics don't work, you aren't building a company - you’re running an expensive research lab. We are finally entering the “commercialization” era of AI. The magic is over; now, the math has to work.

The biggest mistake I’m seeing in 2026 is companies treating AI like a departmental tool instead of a horizontal force multiplier. They deploy AI for coding, or for marketing copy, or for sales and call it a win. But the real game-changing value happens when AI is integrated into the tech stack in a way that connects technical teams with go-to-market teams in real time. When one system understands the code, the product strategy, and the customer context simultaneously. That’s when context switching disappears. Alignment becomes automatic. And launch velocity actually increases end-to-end. The companies pulling ahead aren’t the ones with the most AI experiments. They’re the ones scaling AI horizontally across functions. How horizontally integrated is AI in your organization today? Are you seeing compound value across teams or mostly isolated wins? Drop your honest take! #AI #EnterpriseAI #ProductLeadership #OptibitAI

A lot of people still talk about no-code like it is only good for toy MVPs. That is outdated. In our latest breakdown, we looked at five examples: BuyTicket, SuiteOp, RentFund, MyAskAI, and Formula Bot. What stands out is not just how fast these products launched. It is how far they got before needing heavier engineering support. But founders still need to think carefully about scaling, API costs, backend limits, and where custom engineering eventually becomes necessary. The better question is not whether no-code works. It is when it works best. Read more...

A $1B partnership… and still closed. I was reading the news about OpenAI shutting down Sora and ending the Disney collaboration. And honestly, it’s fascinating. Because this wasn’t a small experiment. This was one of the most hyped AI products — and a partnership with one of the biggest entertainment companies in the world. And yet — it was closed. Success wasn’t the question. Priority was. This is a strong reminder: Clarity is not about doing more. It’s about knowing what NOT to continue — even when it looks successful from the outside. And I see the same pattern with leaders I work with: They don’t struggle with lack of ideas. They struggle with letting go of the “good” to focus on what actually moves them forward. Sometimes the most strategic decision is not scaling — but stopping. What are you still holding on to — simply because it works?

Lately, I’ve been reflecting on how the role of a Business Analyst is changing—and more importantly, what truly makes a BA valuable today. In many organisations, the expectation from a BA is still seen as: “Gather requirements. Document them. Get sign-off.” But in reality, that’s just the starting point. Because projects don’t succeed on well-written documents alone. They succeed when someone ensures that **what is needed is clearly understood, correctly built, and actually delivers value.** That’s where a strong BA makes the difference. From a **project delivery standpoint**, a BA is not just a participant—they are a driver: • Bringing clarity to ambiguous business problems • Aligning stakeholders, tech teams and timelines • Continuously validating requirements during build and testing • Owning gaps, risks and changes instead of passing them along From a **product management perspective**, the role goes even deeper: • Understanding the “why” behind every requirement • Connecting user needs with business outcomes • Thinking beyond features and focusing on value delivery • Acting as a bridge between vision and execution The real shift is this: A BA is no longer just responsible for *what is written*. A BA is responsible for *what gets delivered.* The strongest BAs I’ve seen don’t stop at documentation. They stay accountable till the solution works as intended—technically, functionally, and from a user standpoint. They don’t just capture requirements. They **own them end-to-end.** And that ownership is what transforms a BA from a support role into a **critical contributor in both delivery and product success.** As the industry continues to evolve, one thing is clear: **The value of a BA is no longer measured by documents, but by impact.** Would love to hear your thoughts— What defines a high-impact BA in your organisation today? #BusinessAnalysis #ProductManagement #Agile #DigitalTransformation #ProjectDelivery #Leadership

Most SaaS teams don’t have a data problem. They have a decision problem. Product teams already have: 💬 Customer feedback 📊 Product analytics 🎫 Support tickets 📉 Churn data The issue is not lack of data. The issue is this: All the signals point in different directions. So what actually happens inside teams? • Feedback says “build Feature A” • Usage data suggests “fix Feature B” • Sales pushes “we need Feature C” • Churn data points to something else entirely And now you’re stuck. No clear answer. No defensible decision. Just: ⚡ Internal debates ⚡ Conflicting opinions ⚡ Slow decision cycles This is where most teams break. Because product decisions are not being made on truth — they’re being made on noise. The cost? ❌ Building the wrong features for months ❌ Ignoring the problems actually driving churn ❌ Missing high-impact revenue opportunities ❌ Teams losing confidence in the roadmap We’re building NexBuild to solve this. -Not another analytics tool. -Not another feedback tool. -A decision engine. NexBuild takes all your product signals and answers one question: ➡️ What should we build next — based on impact, not opinion? The shift is simple: From → Data everywhere To → One clear decision If you're a founder or product leader, this is the real question: When signals conflict, what do you actually trust to decide what to build next? If you don’t have a clear answer, that’s the problem we’re solving. Comment “NexBuild” or send me a DM if you want early access. #SaaS #ProductManagement #StartupFounders #BuildInPublic #ProductStrategy

Most small business owners I talk to are hiring for a role that doesn't actually exist yet. They need someone to own their product roadmap AND manage the sprint backlog AND talk to customers AND unblock the team. Then they post a job description for "Product Manager" and wonder why candidates either want to be strategic or tactical, not both. I just read something that nailed this. The distinction between a Product Manager (long-term strategy, market fit) and a Product Owner (sprint execution, backlog management) is real and useful. But in practice? Most growing companies need one person doing both jobs. Here's what I see going wrong: teams get hung up on the title and the role definition instead of being clear about what actually needs to happen. They want someone who can research user needs AND write user stories. Who can define roadmap strategy AND answer engineering questions daily. Who understands business outcomes AND knows how to run a sprint. That person exists. They're just rare. And they're not going to find you if your job description reads like it was written by committee. If you're hiring for this hybrid role, stop overthinking the title. Be specific about the actual work. Tell them they'll spend time on both strategy and execution. Ask about their experience with Agile ceremonies if you use them, but also ask how they think about long-term product direction. Look for people who can shift between big picture thinking and tactical problem-solving. The best hires I've seen don't fit neatly into one box. They do both. https://lnkd.in/g7XquTec Hire With Near

In the fast-paced world of AI startups, we often talk about “Product-Market Fit.” But after leading an AI product from 0 to 1, I’ve realized that “Talent-Stage Fit” is equally, if not more, critical. Many structural problems in tech organizations aren’t actually “people problems”—they are mismatch problems. When you have the right people in the wrong seats, or the right method at the wrong time, you don’t just slow down; you create organizational friction that eats your strategy for breakfast. Here are my 3 core reflections on building an AI-native organization: 1. Talent Density > Headcount In an AI startup, 10 “average” PMs are a liability, not an asset. You need “Full-stack Product Thinkers” who can bridge the gap between prompt engineering, business logic, and user empathy. High talent density allows for a flatter hierarchy, which is the only way to keep pace with the weekly evolution of LLMs. 2. The Right Method for the Right Stage Applying “Big Tech” processes to a 0→1 AI venture is a recipe for disaster. At the early stage, you need “Validated Learning” over “Standardized Execution.” We optimized our architecture for componentization, shrinking iteration cycles from months to weeks, even hours! This wasn’t just a technical choice; it was a cultural one to embrace ambiguity. 3. The Cost of Structural Inertia If you don’t align your organizational structure with your technical roadmap, you get “Conway’s Law” in its worst form. We focused on building an AI-innovative organizational model—optimizing for cross-functional integration and individual output ratio. The Bottom Line: Putting the right people in the right seats, at the right time, with the right methodology—that’s the real “secret sauce.” Anything else is just a structural problem waiting to happen. As we move toward more autonomous AI agents and generative workflows, the role of a Product Lead is no longer just to manage features, but to design the system that enables innovation. #Leadership #AITransformation #TalentStrategy #ProductManagement #StartupCulture #OrganizationalDesign

What I look for in a high-performance product team I know we started off with sharing what I believe so strongly drives the Web3 and SaaS space yesterday. Today, I'd love for us to look at some four traits that the some of the best teams I've worked with across Web3 and SaaS exhibit. Like I said, they commonly share these these: ✅ They focused so much on clarity over the complex stuff: Everyone knew what mattered and why it mattered. You'll most likely never find anyone confused at all. Just sharp priorities tied directly to business impact. And Oh! Every bloated roadmap went straight into the bin :) ✅ They chose speed over perfection: The whole idea for them was to ship fast, learn faster, and iterate relentlessly without distraction. Momentum was and has always been key! ✅ The valued ownership over hierarchy: Everyone literally owned outcomes from engineers to PMs. That people and team energy was the oil that kept the machine moving even in difficult times. ✅ They had a strong appetite for outcomes over output: The entire org measured success in impact: revenue, retention, engagement beyond the number of tickets closed. This was because they understood that shipping features doesn't equate creating value. This combination created teams that: ✅ Unlocked growth ✅ Moved metrics that matter ✅ Drove real business outcomes Even till today, this still creates successful teams! Please, in all your doing as a product leader, prioritize building a great team first. That's a standard I don't compromise on, and the results are clearly evident :) Curious though, what’s one trait that defined the best team you’ve ever worked with? #ProductManagement #PMLife #ProductLeadership #BuildInPublic #Startups #TechCareers #Growth #Execution #Leadership #ProductStrategy #HiringPM #Web3 #Crypto #Fintech #SaaS #Bitcoin

Agentic AI is starting to surface across the research landscape—in product demos, vendor roadmaps, and early conversations about what the next phase of automation could become. But the shift introduced by agentic AI isn’t simply about producing smarter outputs. It’s about creating continuity across the research workflow. Instead of isolated tools that run a task in a vacuum, agentic systems can carry context from design to fielding to analysis and reporting, connecting steps that have historically required manual handoffs. Moving beyond task‑based AI Traditional AI has always been bounded—classifying responses, detecting sentiment, or automating repetitive processes. Helpful, yes. Transformational, no. Generative AI expanded these capabilities by enabling researchers to draft surveys, summarize findings, or produce narrative insights. But agentic AI introduces an entirely different layer: -It can sequence tasks, not just complete them. -It can reason across multiple inputs and adapt workflows in real time. -It can identify gaps, suggest next actions, and initiate steps proactively. -It operates like a partner working alongside the researcher—not a tool waiting for the next prompt. Where generative AI creates outputs, agentic AI begins to create momentum. Check out Forsta's blog for more information.

For those interested in building your own software product with AI, you need to have a clear definition of product-market fit (PMF). I’ve been thinking about PMF after reading a piece from a16z speedrun (guest post by Marcus Segal): Product-market fit is when you can consistently build features that produce a measurable impact on a core metric for a specific user. With how easy it is to build now, it’s very easy to just keep shipping features. But without that definition, it’s hard to know what’s actually working. A more effective approach is simple: pick a metric, build a feature to move it, release it, and carefully measure the result. If it works, keep going in that direction. If it doesn’t, move on. Do that repeatedly and the product improves in a real, measurable way. He also makes a point that’s easy to overlook. A CEO’s job is to not run out of money before finding PMF, and engineering time is your most limited resource. In practice, that means you have a limited window to figure this out. Having a roadmap isn’t just process, it’s what allows you to adjust based on what’s actually happening while you still have time and resources left. One implication of this: Many companies don’t truly have product-market fit when they raise money. They have signals: early growth, user interest, maybe some retention, but they’re still figuring out what really works, which is normal. Product-market fit is usually the result of a lot of small, fast iterations. If you ever need help understanding how PMF might apply to the product or feature you are working on, feel free to reach out.

I worked at the first streaming music service. 1996. TheDJ.com which became Spinner.com. My take: The music industry didn't lose because of piracy. They lost because they spent more than twenty years suing people instead of building something worth paying for. I think about this a lot when I'm working with product leaders who keep losing internal battles. They're smart. They're right. And they are fighting so hard to prove that they're right that they've completely lost sight of what the executive on the other side of the table actually needs to hear. Leadership pushes back on the roadmap and instead of getting curious about why, they go into defense mode. More slides. More data. They're so focused on winning the argument that they stop listening to what the business is actually asking for. The music industry had thirty years of digging in their heels despite getting feedback that their approach wasn't working. https://lnkd.in/gX78Dwyi

One of the largest messaging platforms in the world runs on 40 engineers. It outpaces every well-funded competitor on feature velocity. The team size is not a constraint. It is the strategy. The dominant belief in scaling holds that headcount growth signals progress. More people, more capacity, more output. This belief is wrong in a specific and demonstrable way. Coordination kills output before it is produced. When a team reaches a certain size, 90% of its time shifts from building to coordinating. Every additional person does not add one unit of capacity. They add n units of coordination overhead across the entire existing team. The marginal cost of a new hire is not a salary. It is the fraction of everyone else's attention that disappears into alignment meetings, handoffs, and clarification loops. Idle employees do not stay idle. Underutilized people do not sit still. They find purpose. In a company context, that purpose materializes as internal problems: process redesigns nobody requested, territorial friction, performance overhead. The person who is not building the product becomes, often without intent, a tax on those who are. Scarcity forces automation. Abundance prevents it. A small team managing nearly 100,000 servers across multiple continents has one option: automate or collapse. That constraint produces infrastructure that scales without human intervention. A large team with enough people to manage the same system manually never builds the automation. It rents the solution with labor. The system stays brittle, expensive, and structurally slow. Companies that hire aggressively to scale are, in many cases, hiring their way into the coordination overhead, internal politics, and manual infrastructure that will eventually slow them down. Scale is not headcount. Scale is the leverage each person inside the system generates.

Product Management 101: If you can’t track it, you can’t measure it. If you can’t measure it, you can’t scale it. Tracking gives insight. Measurement gives meaning. Scaling gives impact. Before you invest in growth, make sure your metrics are crystal clear. Otherwise, you’re flying blind. 🚀

Quick reminder: our masterclass on "The New Demand Gen Playbook for Growth-stage B2B Startups" is coming up tomorrow (Thursday, 3/25 at 11am PT). If you're a CEO, CMO or GTM leader at a growth-stage B2B startup trying to scale pipeline without scaling marketing headcount or agency bills, this one's for you. Lyla Kuriyan (former Google MD, SAP VP Marketing) will show you exactly how AI agents are replacing the old demand gen playbook, from spinning up multi-platform campaigns from a product doc, to finding message-audience fit at scale. Register here: https://lnkd.in/gfMxSX3f

A SaaS startup was stuck for 4 months… They had a solid product idea. Funding was in place. Market demand was clear. But nothing was moving. Every sprint got delayed. Features were half-built. Deadlines kept shifting. The problem? Not lack of talent. But wrong hiring + zero ownership. They had: • Freelancers working in silos • No clear accountability • Constant rework due to misalignment After 4 months, they were still nowhere close to launch. That’s when we stepped in. Instead of adding more developers, we replaced the chaos with a Dev Pod: → 1 Backend Engineer → 1 Frontend Engineer → 1 QA → Shared ownership, clear delivery goals No hand-holding. No confusion. What happened next? Within weeks: • Delivery speed increased by 3x • Costs dropped by ~40% • Product roadmap finally started moving—on time Same product. Same vision. Just the right structure. Most CTOs don’t have a hiring problem. They have a delivery structure problem. If you're facing something similar while scaling your product, 👉 Happy to break this down if relevant 👍 #SaaS #Startups #CTO #TechLeadership #ProductDevelopment #SoftwareEngineering #RemoteTeams #DevTeams #Outsourcing #Scaling #AI #EngineeringLeadership

🚧 Scaling Frontend Across Multiple Apps — From Chaos to System Design As products grow, frontend often becomes the bottleneck. That’s exactly what we faced: 👉 Multiple apps (website, dashboard, partners, web-react, microsites) 👉 Different dependency versions 👉 No shared design system 👉 SCSS duplication & UI inconsistency 👉 Risky refactors and slow releases This created friction between engineering, design, and business velocity. 💡 I took ownership to solve this at a system level. Instead of fixing components, I focused on frontend architecture for scale. 🧠 Solution: Monorepo + Nx + pnpm + Token-Driven Design System ✔️ Monorepo (single source of truth) ✔️ Nx (dependency graph + affected builds) ✔️ Tailwind (for new development) + existing SCSS (no rewrite risk) ✔️ Design Tokens (shared across platforms) ✔️ App-level isolation (no forced upgrades) ⚙️ Key Technical Decisions 🔹 nx affected -t build → build only impacted apps 🔹 Independent deployments per app 🔹 Tokens → Tailwind → Apps (single design pipeline) 🔹 Multiple React versions supported (no lockstep upgrade) 🔹 CODEOWNERS + CI for controlled ownership ⚡ What this unlocked ✅ Atomic changes across apps + design system ✅ Faster CI with caching ✅ No version mismatch issues ✅ Consistent UI across products ✅ Safer and faster refactors ✅ Better DX → faster business delivery 📈 Impact 🚀 Cut UI inconsistencies by ~70% 🚀 Improved release speed by 2x 🚀 Reduced engineering friction across teams 🎯 Key Learning Frontend is not just UI. It’s about building scalable systems that align DX with business goals. I enjoy solving problems at the intersection of architecture, performance, and product thinking. Monorepo → "One repo, many apps, zero version drift" Nx → "Build only what changes, deploy only what matters" Tailwind → "Design rules enforced at code level" Tokens → "Single source of truth for UI decisions" Hybrid SCSS + Tailwind → "No rewrite, no production risk" App isolation → "Different versions, same platform" CODEOWNERS → "Ownership without repo sprawl" If you're scaling a multi-app frontend and hitting these walls — happy to talk. #Frontend #SystemsThinking #ProductEngineering #Monorepo #DesignSystem #Tailwind #Nx #TechLeadership #DX 🗂️ Final Architecture After Implementation

How do you decide which features to include in your first MVP to maximize learning with minimal effort - Dropbox example? (4b) Before building the full product, they asked one key question. Do people actually want simple file syncing across devices? Instead of building complex infrastructure, they created a simple video demo. The video showed how the product would work. No real backend, no full system. Just the core idea. This MVP had one goal. Test demand. People watched the video and signed up in large numbers. That was the signal. The assumption was validated before heavy development. Notice what they did not build. No full app, no advanced features, no scaling system. Just a way to test user interest. This is the lesson. Your MVP is not your product. It is a tool to answer one important question. If Dropbox had built everything first, they would have wasted time and money if no one cared. Strong product teams separate building from learning. First learn. Then build.

Stop shipping features into a black hole! With AI, we can build anything in an afternoon. But faster execution without clear thinking doesn't get you to product-market fit faster, it just gets you to confusion faster Our friend Doug Peete from atono (a total product legend) taught me the way you get AI to perform better is the same way you get a junior developer to perform better. Write better specs. Before you build your next "simple" feature, run it through the INVEST framework to see if it’s ready: ✅ I – Independent: Can it ship on its own without waiting for other work? ✅ N – Negotiable: Does it focus on the result, not just a rigid set of steps? ✅ V – Valuable: Can you actually name who benefits and how? ✅ E – Estimable: Is it clear enough for your team to guess how long it’ll take? ✅ S – Small: Is this the tiniest version of value you can deliver? ✅ T – Testable: Do you know exactly how to check if it’s "done"? Just because it's easier to build doesn't mean features shouldn't still be tied clearly back to a high-impact customer outcome. Plan first. Then build fast. Want more tactical tips on building and scaling? Sign up for The Founder Playbook here: 👉 https://lnkd.in/eVHA7VjE What’s one feature you shipped recently that ended up being way more complicated than you expected? 👇

The Execution Gap: Why Strategies Fail After 25+ years leading cross-functional operations globally, I’ve observed a consistent pattern: Organizations rarely fail because of flawed strategy. They fail in the space between strategy and execution. Even when: · Strategic priorities are clear · Financial targets are achievable · Governance structures formally exist Outcomes often remain inconsistent - and vary widely across organizations. Why? Execution at scale requires disciplines that are frequently underestimated: * Translation Discipline Strategy must be translated into meaningful regional and functional operating charters – not merely cascaded through presentations and town halls. * Performance Architecture KPIs must align across the enterprise – spanning Product/Service Development, Sales, Delivery, Finance, Customer Support, Supply Chain, Business Operations. When functions are optimized in isolation, enterprise performance suffers. * Governance Cadence Periodic reviews alone don’t drive performance. A consistent operating rhythm – and the discipline to sustain it – does. * Balancing Short-Term & Long-Term Objectives The constant tension between showing quick results vs. sustainable long-term value creation is real – and often not effectively managed. * Right People, Right Roles (at the Right Time) Execution depends not just on talent, but on context – placing the right leaders in the right roles under the right circumstances. The capabilities required to take an idea to market are often distinct from those needed to scale, integrate acquisitions, or lead turnarounds. Ultimately, leadership is not only about defining strategy – it is about architecting the system that makes execution repeatable and scalable. In volatile markets, disciplined execution is not just a structural advantage, but a survival capability. As I re-engage here on LinkedIn, I look forward to exchanging perspectives on operating at scale, governance rigor, cross-functional integration, and sustainable growth. Question to the community: In your experience, what most often breaks between strategy and execution?

How do you know when you’re truly ready to scale? 📈 We’re pleased to welcome Mark Roberge, author and founding CRO of HubSpot, as a speaker at this year’s Pender Tech CEO Conference. Drawing on ideas from his new book, The Science of Scaling, Mark will share practical perspectives on: 🚀assessing readiness to scale across new products and new markets 🧩aligning the go-to-market system to each phase of growth, including leadership hiring and organizational design 🤖the implications of AI for GTM, including team structure and execution advantage For growth-stage CEOs, scaling is rarely just about doing more. It’s about knowing what to change, when to accelerate, and how to build the right system for the next phase of growth. We’re looking forward to bringing this conversation to the Pender Tech CEO community. #PenderTechCEOConference #Scaling #GoToMarket #Leadership #AI #SaaS #CEO

Most founders treat customer conversations like validation exercises. You walk in with your roadmap, ask leading questions, and leave feeling good about what you already planned to build. The conversations that actually matter are the ones where you leave confused. Early on I'd get on calls with clinicians and ask about their workflows. I had my assumptions. I thought I knew where the pain was. Then someone would describe their day and mention a problem I hadn't considered, almost as an aside. Like it was so obvious to them it barely worth stating. Those throwaway comments rebuilt our product three times. The pattern I've noticed is that customers don't tell you what to build. They tell you what they've given up on. There's a difference. When someone says "it would be nice if your product did X," that's a feature request. When someone says "yeah we just use a spreadsheet for that because nothing works," that's a roadmap. The hard part is hearing it. Because you're in the meeting with an agenda. You're trying to demo. You're trying to close. And the most important thing a customer will ever tell you usually sounds like a digression. I've gotten better at catching those moments, but I still miss them. You always do. The best product founders I know aren't the ones with the clearest vision. They're the ones who can sit in a customer conversation, hear something that contradicts their plan, and get curious instead of defensive. Your customers have been living with the problem longer than you've been solving it. Sometimes the smartest move is just making more room for them to talk.

𝗠𝗼𝘀𝘁 𝗔𝗜 𝗿𝗼𝗮𝗱𝗺𝗮𝗽𝘀 𝗮𝗿𝗲 𝗯𝘂𝗶𝗹𝘁 𝗮𝗿𝗼𝘂𝗻𝗱 𝗰𝗮𝗽𝗮𝗯𝗶𝗹𝗶𝘁𝘆. 𝗧𝗵𝗲𝘆 𝘀𝗵𝗼𝘂𝗹𝗱 𝗯𝗲 𝗯𝘂𝗶𝗹𝘁 𝗮𝗿𝗼𝘂𝗻𝗱 𝗰𝗼𝗻𝘀𝗲𝗾𝘂𝗲𝗻𝗰𝗲. Most teams start with: • What can we automate? • What can we generate? • What decisions can we accelerate? That’s the wrong starting point. Because the real question isn’t what AI can do. It’s what happens after it does it. 🔹 Every AI action creates downstream impact A recommendation changes behavior. An automated decision shifts accountability. A generated output influences outcomes. If you don’t design for that impact, you’re not building a feature - you’re introducing risk. 🔹 Consequences compound across systems AI rarely operates in isolation. One decision feeds another system. One output becomes another input. One automation removes a human checkpoint. Small design choices scale into system-wide effects. 🔹 Most roadmaps ignore second-order effects Teams prioritize speed and capability. They rarely map: • what breaks if the system is wrong • how errors propagate • who absorbs the consequences • how recovery actually works That gap shows up later... at scale. 🔹 This is where product leadership shows up Not in identifying use cases. But in defining: • where AI should act • where it should pause • where humans stay in the loop • and how failure is contained AI isn’t just a capability layer. It’s a consequence engine. The best teams don’t just build what’s possible. They design for what happens next. #AgenticAI #AIGovernance #ProductLeadership #ArtificialIntelligence #DigitalHealth #Fintech #OperatingModel #CPO

“Unlock Hidden Insights: From Silos to Impactful Discoveries” What happens to the unexpected insights that come out of research? In this clip from Talking Roadmaps – Season 2, Episode 24, Jake Burghardt discusses a common problem with Justin Woods: researchers often discover valuable insights beyond the original study — but those ideas never go anywhere. Without a clear path to share, combine, and amplify them, important discoveries quietly disappear instead of influencing product direction. 🎥 Want to hear more of what Jake Burghardt discussed with Justin Woods about #productops? Check out the full episode here: https://lnkd.in/gmEy5mCf #expert #productmanagement #productmgmt #productops #researchops #productdiscovery #customerinsights

𝗧𝗵𝗲 𝗘𝘅𝗲𝗰𝘂𝘁𝗶𝗼𝗻 𝗚𝗮𝗽: 𝗪𝗵𝘆 𝗔𝗜 𝗴𝗶𝘃𝗲𝘀 𝗽𝗹𝗮𝗻𝘀—𝗯𝘂𝘁 𝘄𝗼𝗿𝗸 𝘀𝘁𝗶𝗹𝗹 𝗱𝗼𝗲𝘀𝗻’𝘁 𝗵𝗮𝗽𝗽𝗲𝗻 AI can generate a 10-step plan in seconds. It can outline a marketing strategy, map a product launch, or produce a 90-day roadmap that looks complete on screen. Yet the work still doesn’t happen. The document gets saved. The plan stays theoretical. The “I was gonna” loop continues. The response is more detail. A more actionable version. A better prompt. But execution doesn’t fail from a lack of ideas. It fails from a lack of systems. Without a way for decisions to move into action, plans remain inert. The speed of generation only makes the gap more visible. Intention isn’t the bottleneck. Translation is. — Brian iWasGonna™ These pieces are intended to start conversations about how teams grow with AI—not prescribe answers. For more on how we think about governed systems and durable execution: https://lnkd.in/gD8WUN24 #AI #Execution #AIGovernance #AILeadership

A lot of startup roadmaps are just expensive to-do lists. The team is busy. The sprint board is full. PRDs exist. Tickets exist. Work is shipping. But the hard questions are still unanswered: Are we solving a real pain point? Is this feature tied to retention, growth, or monetization? What customer evidence supports this? What are we not building because of this choice? Are we shipping conviction or just momentum? This is where ProdMoh changes the game. ProdMoh is not another writing tool for PMs. It is not another ticketing layer. It is not another “AI copilot” that stops at content generation. It is a decision workflow: Customer signal → Product Spec → User Stories → Product Canvas → Decision Brief → Release Gate → Launch Pack → Outcome Follow-up That means product teams can go from raw reviews and support tickets to: clear pain point analysis competitor-aware product opportunities structured PRDs implementation-ready stories AI coding guardrails and evals launch documentation executive decision visibility Without this system, startups often keep funding features that look productive but do not move the business. That is how runway disappears. ProdMoh helps founders and PMs allocate engineering capital where it actually matters.

There’s a moment in most roadmap reviews where the room changes. Up to that point, it’s presentation: what’s shipping, when, how it all fits. Then someone asks: “𝘞𝘩𝘺 𝘵𝘩𝘪𝘴?” And now you’re not presenting. You’re defending judgment. Most PMs aren’t prepared for that part. They’ve talked to customers. They’ve got themes and quotes. They can summarize what they heard. But when you ask: What’s still unclear? What could break this plan? What would have to be true for this to work? …answers get thin. So teams fall back on activity: We talked to customers. We ran interviews. We’re seeing consistent feedback. That’s not the same as resolving uncertainty. When the pressure goes up, the instinct is to gather more input. More voices, more data. It gives you a fuller picture of today. It doesn’t necessarily help you decide what to do next. Sometimes it does the opposite. It makes it easier to keep the decision open. The teams that handle this well tend to narrow. They spend time with a small set of deeply engaged customers who can explain, in practical terms, what would actually make something more useful. Not experts. Not casual users. People who use the product seriously and aren’t shy about where it falls short. I’ve seen teams about to invest in fixing a “confusing” workflow because a lot of users said it was confusing. When they walked through it with a few of these customers, the reaction was different: It wasn’t confusing. It just wasn’t worth doing. That’s a very different decision. That’s where 𝗖𝗮𝘁𝗮𝗹𝘆𝘁𝗶𝗰 𝗖𝘂𝘀𝘁𝗼𝗺𝗲𝗿𝘀 matter. They don’t replace broader input. They sharpen it, especially when you’re trying to resolve what actually matters. Because when someone asks “Why this?” you don’t need more input. You need clearer judgment. P.S. If you’re in a roadmap defense right now and the input isn’t helping you get to a clear call, comment “𝗪𝗛𝗬 𝗧𝗛𝗜𝗦” and I’ll send you the 3 questions I use to isolate the uncertainties that actually matter.

Vendredi, un commercial avait signé un client, le lundi à 17h il avait inventé un produit. J'ai retrouvé un ancien email qui m'a un peu gonflé. Message Slack le lendemain dans #product, tranquille, comme s'il demandait le wifi : "Au fait, le dashboard de monitoring, on dit quelle date ?" Le fameux dashboard de monitoring, tu cherches ça dans le Notion, dans le backlog, dans la roadmap, tu ne trouveras rien. Le seul endroit où ça existait c'était dans un call de closing du vendredi, face à un prospect qui hésitait. Le gars ne s'est pas contenté de dire oui, Il a share son screen, ouvert un Google Sheet préparé la veille, mis des filtres, et a du sortir un truc du genre "là on est en pleine refonte de la feature donc je peux pas vous montrer de visuel mais ça va donner ça".. Bam deal signé, honnêtement même si ça m'avait gonflé je sais que c'est ça la bonne méthode et ce fameux gars m'a impressionné, bien loin des clichés du commercial qui demande et attend que les devs fassent, il s'est débrouillé tout seul. Ca part en congratulations dans tous les sens, les 🎉 partout dans le thread, mais moi je me suis juste demandé combien de soirées à 23h cette histoire allait me coûter (et ouais on était pas encore à l'ère où l'IA te fait un dashboard en 30 minutes) Mais en vrai c'est comme ça que ça marche, j'ai mis du temps à l'accepter à l'époque mais une bonne boîte ne roule jamais sur des rails propres. Il faut trouver l'harmonie entre les promesse, les attentes, ce qui est faisable et ce qui est dispo, le reste c'est notre égo.

Got this response before: "We already using a similar system. Thanks anyway." He was right. I built my Valuation Engine to work standalone. But if it doesn't fit their existing workflow, they won't use it. So here's what I did: Spent the last 3 days figuring out email parsing. Now my AI: → Intercepts Portals and website notification emails → Extracts vendor details → Triggers WhatsApp workflow → Books appointment → Your team enjoying the booked appraisals without effort Zero API costs. Zero CRM migration. Fits their existing system. The lesson: Your first version will never be right. Build it anyway. Then rebuild it based on real objections. Customer rejection = Product roadmap. #ProductDevelopment #CustomerFeedback #Iteration #Startup #BuildInPublic

A bit of my working process. An auto-tech startup came to me with an unusual request: “Make it less exciting.” (All details have been removed for confidentiality.) They had already secured an investor mandate, but their strategy had shifted. The deck that created early-stage excitement wasn’t going to survive a due diligence room. So we rebuilt it — with a completely different language. I would say more like Consulting-style. → Strategy slide: from chaotic bullet points to a clear strategic framework → Concept slide: from high-energy storytelling to product logic (still engaging, though) → User roadmap: from visual noise to a clean, structured progression And it worked.

Hi everyone, I am curious how teams here are managing delivery while hiring engineers. In a lot of product teams I’ve spoken with, hiring cycles are taking longer, and it’s starting to impact roadmap timelines. Are you usually waiting to fill roles, or bringing in temporary support to keep things moving? Would love to hear how others are approaching this.

The MVP Paradox: Why Less is More for Startup Success in 2026 📊💼 The definition of a "Minimum Viable Product" has shifted. In a 2026 market saturated with high-quality software, your MVP shouldn't be "buggy" or "unfinished"—it should be focused. Founders who attempt to solve five problems at once usually end up solving zero. The 2026 Lean Strategy: 🔹 The Core Loop: Identifying the single most important action that provides value to your user and discarding everything else. 🔹 Validation over Velocity: Why a hundred highly engaged beta users are worth more than ten thousand "looky-loos" in the early stages. 🔹 The Feedback Iteration Cycle: Using real-time data to pivot your roadmap before you spend your next round of funding. In 2026, speed is secondary to signal. If you can't prove value with a single feature, more features won't save you. 🏗️🌐 Full Strategic Guide: https://lnkd.in/ekPpGy4q #StartupStrategy #MVP #ProductMarketFit #Entrepreneurship #LeanMethodology #2026Innovation

Most businesses don't fail because of bad products. They fail because of bad strategy. I've been studying strategy as part of my MBA — and one thing keeps hitting different: Dr. Ajit Kar 👉 Competitive advantage isn't just about what you do. It's about what you choose NOT to do. Porter said it best — strategy is about making trade-offs. A company trying to be everything to everyone ends up being nothing to no one. Think about it: → Apple chose premium. Not market share. → Zepto chose speed. Not variety. → Zara chose fast cycles. Not luxury. Every great strategy has a clear "NO" behind every "YES." As I build my career in Strategy & Business Analysis, this is the lens I'm learning to think through: 𝗪𝗵𝗮𝘁 𝗮𝗿𝗲 𝘆𝗼𝘂 𝘀𝗮𝘆𝗶𝗻𝗴 𝗡𝗢 𝘁𝗼 — 𝘀𝗼 𝘆𝗼𝘂𝗿 𝗯𝗶𝗴 𝗬𝗘𝗦 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗺𝗲𝗮𝗻𝘀 𝘀𝗼𝗺𝗲𝘁𝗵𝗶𝗻𝗴? Drop your thoughts below 👇 #Strategy #BusinessInsight #MBA #StrategicManagement #LinkedInGrowth #Ayush #RCM #FutureAnalyst

Proud moment for the team 💥 Syndigo just launched Synapse, bringing agentic AI to product experience management—so teams can focus less on execution and more on strategy. If you’re at Shoptalk, stop by Booth #3624 to learn more 👋 https://lnkd.in/d7VMpjeC

We had product sitting in freezers. And no way to sell it. Retail had slowed. Foot traffic had shifted. A newly launched consumer brand—with well over $1M invested—had no clear path to market. This wasn’t a marketing problem. It was a commercialization problem. At the time, the business was built for: • private label • traditional retail distribution • in-store velocity Not for: • direct-to-consumer • e-commerce fulfillment • digital demand generation But the product existed. And time was working against us. So we rebuilt the model—quickly and intentionally. Within months, we: • launched a direct-to-consumer e-commerce platform • reconfigured internal operations into a functional distribution model • built fulfillment processes for perishable, frozen product shipping • conducted pricing and shipping elasticity testing to protect margin • developed packaging solutions to maintain product integrity in transit • created digital campaigns to drive awareness and trial • activated influencers to build early demand and credibility At the same time: • supported retail sell-in through virtual product demos • equipped buyers with product and preparation tools remotely • built a consumer story backed by real purchase behavior This wasn’t about adding a channel. It was about building a system the business had never operated within before. Strategy, operations, and execution had to align—quickly. The result: • immediate revenue recovery during disruption • a fully functional e-commerce channel built from scratch • a data-driven narrative to support retail expansion • successful retail placement post-COVID What this reinforced for me: Growth doesn’t pause when the environment changes. But the model often has to. And when it does, speed matters—but alignment matters more. #Commercialization #EcommerceStrategy #CPG #FoodAndBeverage #GoToMarket #DigitalTransformation #OperationalExcellence #SupplyChain #DirectToConsumer #theNEWIRway

There is no other industry where this happens. OpenAI spent years and significant compute building Sora from a research preview in February 2024 to a standalone app in September 2025 to a second-generation model with audio and physics that rattled Hollywood. The app hit a million downloads faster than ChatGPT. Disney signed a three-year deal in December, pledged a $1 billion investment, licensed over 200 characters. And on Monday evening, Disney and OpenAI teams were still in a working session together on a Sora project. Thirty minutes after that meeting ended, Disney was informed the product was being shut down entirely. The $1 billion investment never closed. The deal is dissolved. The app, the API, and sora.com are all going away. The stated reason is compute. Makes sense. OpenAI says it needs to make trade-offs on products with high compute costs ahead of a potential IPO, and the Sora research team will pivot to world simulation for robotics. Downloads had dropped 45% by January anyway, and lifetime in-app revenue was roughly $2.1 million, which against the compute costs of running a video generation model at scale is essentially a rounding error. But the real story here is about what it means to build on top of AI platforms in 2026. If you integrated Sora into a creative workflow, your integration is gone. The API is gone. The prompting patterns you learned, the creative consistency you tuned for, the production pipelines you built around it, all of that is now worthless. And this isn’t a generic API where you swap providers and move on. Video generation models each have fundamentally different approaches to motion, physics, style control, and prompt interpretation. Switching from Sora to Veo or Kling or Runway is not a migration, it’s a rebuild. This is the part of AI adoption that nobody puts on conference slides. The same week Jensen Huang told 30,000 people that every company needs an AI strategy, one of the largest AI companies on the planet killed a flagship product and blindsided a partner that had committed a billion dollars to the relationship. No deprecation period. No migration path. No advance notice beyond 30 minutes. Imagine doing that with one of your products in your company! The AI space moves at a pace that makes no sense from either direction. Companies invest enormous resources building products they abandon in months. Customers invest real time and money integrating with tools that vanish without warning. And somehow both sides keep doing it, because the cost of not participating feels higher than the cost of getting burned. The lesson to learn as a Head of AI is not to avoid AI tools. It’s to architect for impermanence. Assume every external AI capability you depend on could disappear tomorrow (also shoutout to Anthropic service outages), because yesterday, one did.

80% of your inactive sellers didn't quit. They got stuck. And nobody noticed. 👻 They didn't leave angry. They left in silence. That's worse. → Lost the Buy Box and didn't understand why → First products got zero traffic, felt invisible → A listing got rejected, didn't know how to fix it → Contacted support, got ghosted They didn't need motivation. They needed help. 🔇 Here's the move : Stop sending "We miss you!" emails. That's not reactivation. That's a Tinder message. 💀 Do this instead: Pull their data. 2 minutes per seller. → No stock? Supply problem, not motivation. → No Buy Box? That's why they stopped. → No login in 30 days? They mentally checked out. Then match the message to the problem : 📩 Day 14 → "Your top product lost the Buy Box. Here's a 2-min fix." 📞 Day 30 → Personal call. "What happened? What can we fix?" 🎁 Day 45 → Last chance: free promo boost or reduced commission for 30 days. 🪦 Day 60+ → Archive. Clean the catalog. Move on. Don't wait for silence. Detect the decline. A seller at -40% this week is next month's ghost. 📉 The best reactivation strategy? Not needing one. 🎯

WINNING CULTURE !!! After working with different organizations, industries, and leadership teams, one thing becomes clear: High-performance cultures look different on the surface, but they operate with the same fundamentals. Whether you look at top companies, elite sports teams, military units, or the most successful tech organizations, the pattern repeats itself. Winning is not random. Winning is built. 1. True North must be clear — and protected Organizations fail when direction changes every week. High-performance teams operate with one priority, one definition of success, and one clear path forward. The military calls it the mission. Apple calls it product vision. Amazon calls it long-term strategy. Asian manufacturers call it the plan. Direction does not move with emotions. Direction does not move with pressure. Direction moves only with strategy. When direction moves too often, execution dies. Confusion is the enemy of performance. 2. Make numbers sacred Every successful organization respects numbers. Toyota tracks output every hour. Amazon tracks performance every minute. Factories track every shift. Sales teams track every deal. Numbers remove opinions. Numbers remove politics. Numbers remove excuses. Track every week: • Revenue • Margin • Productivity • Efficiency • Cost • Quality • Output per person • Variance vs plan • Improvement vs last week If it is not measured, it is not managed. If it is not reviewed, it will fail. If it is not owned, it will drift. Numbers are not pressure. Numbers are control. 3. Build discipline rituals Motivation changes. Rituals stay. Winning organizations operate on cadence, not emotion. Daily — review results, fix gaps, assign actions Weekly — review numbers, trends, ownership Monthly — review targets, financials, performance Quarterly — review strategy, structure, direction Toyota has daily stand-ups. The military has daily briefings. Formula 1 reviews every lap. Asian factories review every shift. No hiding problems. No delay in action. No excuses culture. Improvement is required. Discipline creates consistency. Consistency creates winning. 4. Reward results, not effort Effort is expected. Results are respected. You are paid for impact, not for trying. This creates accountability. And accountability creates trust. 5. Stability of strategy, flexibility of execution Winning organizations do not change direction every week. They adjust execution, not purpose. Direction must be stable. Execution must be adaptive. Discipline must be constant. Numbers must be real. Accountability must be visible. 6. Responsibility culture beats comfort culture Every elite organization shares one belief: Pressure is normal. Targets are necessary. Accountability is respect. Discipline is strength. Improvement never stops. #N1M #WhynotMe

🚨 AI is not just answering questions anymore. It’s starting to take actions. And this is changing everything. We are moving from: 👉 AI as a tool (chat, answers, support) To 👉 AI as an agent (decides, acts, executes) Today: • AI answers customer questions • Suggests products • Assists sales teams Tomorrow: • AI follows up with leads automatically • AI books meetings • AI negotiates deals • AI completes end-to-end customer journeys 💡 This is the real shift: We are not automating tasks anymore. We are automating outcomes! And it’s happening gradually: 1. Assistive AI Helps humans do the job 2. Augmented AI Works alongside humans 3. Autonomous AI agents Starts doing parts of the job independently 4. Multi-agent systems Multiple AI agents collaborate like a team ⚠️ What does this mean for sales & customer service? It means: • Fewer people doing basic execution • More focus on oversight, strategy, and relationships 👉 The role is evolving from: “Doing the work” ➡️ To “Managing the work done by AI” 🎯 So what should you do? • Learn how AI agents work (not technically, but conceptually) • Understand workflows, not just tasks • Build skills in decision-making and customer understanding • Become someone who can design, guide, and supervise AI And … Know what should be done .. and let the AI do it for you! #AI #AIAgents #FutureOfWork #Leadership #Sales #CustomerExperience #DigitalTransformation #SS

Anyone can prompt an app into existence in a weekend now. But that doesn't mean anyone will actually want to use it. As we get flooded with AI-generated software, we are entering an era of commodity products. The technical barrier to entry is effectively hitting zero. But we’ve seen this shift before in other industries. A hundred years ago, furniture was artisanal and rare. Then mass production changed everything. Suddenly, you could buy a set of functional chairs for cheap on any street corner. So why are some chairs still expensive? Why do they still sell decades later? Take my absolute favourite, and a Norwegian icon, the Ekstrem chair by Terje Ekstrøm (pictured). It was designed in the 80s, but it still feels like the future. I’ve seen it in museums and children’s hospitals. It works everywhere. It hasn’t aged because of the eye behind it. It is a specific quality of thought and intentionality that mass production can't replicate. It isn't just a place to sit. It is a piece of strategy. The shift we’re seeing in tech is the same. In an age of infinite software, good design is no longer a luxury. It is the only way to build something that actually sticks. When code is free, the only thing that differentiates a product is the human intelligence behind the experience. Good design doesn't age. And in a world of AI, that human eye is what makes a product irreplaceable. What is the one product or object you've used for years that still feels like it was designed for the future? I'm curious to see if there's a common thread. Image credit: Varier Furniture

Subject: I know we're not Verizon, but Meta — can you hear me now? Team, Another round of layoffs. Another org shift. Another note about “strategic alignment.” The stock moves up. Morale moves down. And I keep asking: what’s next? What’s the next cut that makes the ticker jump while the team takes the hit? Is this what we set out to build? If the vision got blurry, where were we to clean the lens? Or are we now in a cycle where the blind are leading the blind, because calling it out feels riskier than going along? Was the Oculus strapped on too tight until the focus narrowed and the metaverse faded out? Did the Meta glasses fog up until we stopped seeing the people wearing them? We chased the future so hard we forgot to ask if anyone wanted to live in it. And now we have to ask: who’s driving? Are we leading, or is the algorithm? We feed it prompts, it recommends cuts, and we call it “alignment” — but it feels like management by autopilot. You can’t code loyalty. You can’t downsize trust. You can’t delete the people and expect the product not to rust. As a Regional Lead, what I’d like to see is us treating efficiency and people as a both/and. We can invest in AI and still protect the teams who know how to ship it. We can grow revenue by removing execution friction, consolidating priorities, and letting proven programs scale — instead of cutting the managers and ICs who translate strategy into outcomes. We don’t need more subtraction to get a positive climb. We need focus: fewer parallel bets, clearer ownership, and trust in the people already here to deliver. That’s how we move the stock and morale in the same direction.

The role of the frontend developer is shifting rapidly. It is no longer about writing perfect CSS grids from scratch. It is about being an efficient editor. Figma Make can spin up beautiful UI layouts in seconds. But moving that design into a production environment without running up a massive bill for Dev Mode seats is a real challenge for small teams. You do not need to manually redraw these AI layouts in VS Code. You can bridge the gap using free third-party environments and sandbox workarounds. The workflow is simple: Let the AI handle the visual heavy lifting, use free sandboxes to extract the structural code, and spend your human energy on the logic and accessibility. If you are trying to scale your development workflow without scaling your software subscriptions, this breakdown is for you. How is your product team adapting to AI design tools this year? Are you finding them helpful or just another layer of cleanup? Let us discuss below. Check out the full guide on how to bypass the paywall: https://lnkd.in/gNSMtyHD

You're optimizing the wrong thing. That’s why your growth isn’t scaling. Growth doesn’t break everywhere. It breaks in one single moment. Activation. (Not sign-up. Not more onboarding steps.) The first time a user actually feels the value. You can have traffic. You can ship features. You can polish flows. But if users don’t reach that moment… Nothing compounds. 👉 It’s not a traffic problem. 👉 It’s a moment problem. That moment defines everything: Retention. Revenue. Expansion. If you can’t name the exact action that means “they got it”… that’s your bottleneck. Comment your product type + what users should achieve in the first 5 minutes. I’ll help you find the moment. . . . #productgrowth #saas #b2bsaas #activation #retention #productmanagement #uxdesign #productstrategy

After more than 20 years in legal consulting and international corporations, I’ve decided to take a different path. From now on, I will be focusing on technology – at the intersection of digital health, regulated industries, and product thinking. It feels like a natural next step. Because today, law is no longer just about control or risk mitigation. It’s becoming part of how businesses move, grow, and make decisions – faster, smarter, and with greater confidence. In environments where decisions happen in days (sometimes hours), legal expertise needs to be not only strong, but also flexible, practical, and deeply integrated into the product and the business itself. And that’s exactly what excites me about tech – especially digital health. This shift is not about changing industries. It’s about expanding how I work and think: 💎 being closer to where decisions actually happen; 💎 combining legal expertise with business thinking in real time; 💎 building things from scratch, not only maintaining what already exists; 💎 supporting growth – from early structure to scaling and working with investors. I’m also rethinking what growth means for me. Less about hierarchy. More about impact, speed, clarity and real accountability. About freedom to think, to act, and to shape outcomes. And about being closer to decisions, to people, and to what truly matters. My focus now is simple: flexibility, adaptability, and pragmatism – combined with a broad legal perspective and strong commercial sense. Everything I’ve done before – working with regulators, navigating crises, and operating across complex jurisdictions – doesn’t stay behind. It becomes the foundation that helps teams move faster, make better decisions, and grow with confidence. This is a new chapter for me. And I’m genuinely excited to contribute, grow alongside the team, and build something meaningful and impactful together.

Sales teams evaluating AI SDRs face a tricky question How do you calculate ROI before you even start? After running the numbers on hundreds of outreach campaigns, here’s what actually matters 1️⃣ Factor in response time and the revenue cost of delays Leads convert best within 10 minutes of first contact, but once you wait longer, conversion rates can drop dramatically AI responds instantly every time, so on a $50K deal, avoiding that delay can save you $40K in lost revenue potential 2️⃣ Set realistic revenue targets A common benchmark is that SDRs should generate 3–5x their cost in revenue to break even For an AI SDR at $10.8K/year, that means targeting $32K–$54K in revenue impact to hit ROI, which gives you a clear number to track 3️⃣ Calculate cost per personalized email At $0.90 per email with AI-powered personalization, you can send 1,000 highly personalized emails for $900 If you’re running campaigns across hundreds of accounts, that kind of cost predictability makes scaling straightforward 4️⃣ Consider the opportunity cost of your budget At $10.8K/year, an AI SDR leaves budget to hire closers, invest in product, or scale other revenue-driving work that still needs deep human expertise AiSDR handles the repetitive, high-volume work so your team can focus on relationships and closing deals Full ROI breakdown in our blog 👇 https://hubs.la/Q048ffVB0

𝗜𝗻 𝗮 𝗺𝗮𝗷𝗼𝗿 𝗺𝗼𝗯𝗶𝗹𝗶𝘁𝘆 𝗶𝗻𝗶𝘁𝗶𝗮𝘁𝗶𝘃𝗲, 𝗢𝗽𝗲𝗻 𝗗𝗮𝘁𝗮 𝗣𝗿𝗼𝗱𝘂𝗰𝘁 𝗦𝗽𝗲𝗰𝗶𝗳𝗶𝗰𝗮𝘁𝗶𝗼𝗻 𝗮𝗰𝘁𝗲𝗱 𝗮𝘀 𝗮 𝗹𝗶𝗻𝗴𝘂𝗮 𝗳𝗿𝗮𝗻𝗰𝗮 𝗮𝗰𝗿𝗼𝘀𝘀 𝟵 𝗰𝗶𝘁𝗶𝗲𝘀 𝗮𝗻𝗱 𝗺𝗼𝗿𝗲 𝘁𝗵𝗮𝗻 𝟮𝟬 𝘂𝘀𝗲 𝗰𝗮𝘀𝗲𝘀. Each city had different systems, standards, and maturity levels. Without a shared model, alignment would have failed early. #ODPS provided that common language. It defined how to describe a data product. - It structured how teams captured requirements. - It aligned technical, business, and governance perspectives. The intake canvas was built around ODPS concepts. Every use case was described using the same structure. This made collaboration possible at scale. Instead of debating formats or terminology, teams focused on defining data products with clear purpose, access, and value. 𝗧𝗵𝗲 𝗿𝗲𝘀𝘂𝗹𝘁 𝘄𝗮𝘀 𝟰𝟯 𝗱𝗮𝘁𝗮 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝗼𝗳𝗳𝗲𝗿𝗶𝗻𝗴𝘀 𝗱𝗲𝗳𝗶𝗻𝗲𝗱 𝗮𝗰𝗿𝗼𝘀𝘀 𝘁𝗵𝗲 𝗶𝗻𝗶𝘁𝗶𝗮𝘁𝗶𝘃𝗲. That is what a lingua franca does. It removes friction and creates shared understanding.