# Greg Coticchia

Greg Coticchia writes and advises on product management, helping companies connect product decisions to customer problems and market needs.

- Profile: https://b2binfluencer.marketing/creators/greg-coticchia/
- LinkedIn: https://www.linkedin.com/in/gregcoticchia
- LinkedIn followers: 12,382
- Topics: artificial-intelligence, startup, product-management
- Based on 10 dated posts, latest 2026-08-27

## Summary

Most useful for product strategy and customer-problem discovery workshops or advisory.

- Best for: A B2B SaaS company seeking product strategy and customer-problem discovery workshop or advisory.
- Not for: Campaigns needing paid distribution, audience targeting, or verified case evidence.
- Check first: Confirm delivery format, recent post results, and audience data before contacting them.

### Takeaways

- A feature request is data, not a decision.
- Capability is not a value proposition.
- An acquisition changes who the product is for.

## Main ideas

- **A feature request is data, not a decision**: Greg Coticchia describes a call where a CEO told a CTO to build a feature because it would make the product stickier—without data, customer input, or a clear problem to solve. The CEO’s opinion carried something data did not: authority. Coticchia draws a hard line between letting customers design your solution and understanding the problem they’re trying to solve. They treat executive intuition, a large customer’s demand, and a sales emergency as three forms of input that are each insufficient on their own. Before deciding to build, they ask: Does it solve a meaningful problem, and for whom? How many customers have it, and how painful is it? What information exists? Does it strengthen differentiation or advance strategy? What won’t be built instead? Saying yes to everything produces what they call Frankenstein: a collection of individually defensible features that collectively make no strategic sense. Their priority is protecting strategy from the loudest voice in the room.
- **Capability is not a value proposition**: Writing about AI agents, Greg Coticchia argues that the gap between Silicon Valley enthusiasm and ordinary users isn’t a technology problem but an old product management failure: technology in search of a problem. They describe a pattern they say they’ve watched across more than 100 product launches: engineering builds something impressive, the demo lands inside the building, everyone assumes it’s obvious, the product ships, and the market shrugs. The problem isn’t bad technology; nobody started with a customer problem worth paying to solve. Their rule: “It can” isn’t “I need to.” Winning teams work backward from a real job, a real frustration, a moment when someone would pay to never do something again. Sequence matters to them: problem first, customer first, capability in service of both.
- **Knowing too much can be why a company misses a market**: Greg Coticchia argues that the companies that miss a market aren’t inexperienced; they know too much. Expertise stops them from asking the basic question that matters. They list healthcare, legal technology, enterprise software, manufacturing, and cybersecurity as fields where they’ve watched this play out. They frame their own lack of domain knowledge as a structural advantage, letting them see assumptions insiders could no longer notice.
- **An acquisition changes who the product is for**: On AI startup acquisitions, Greg Coticchia says the technology is the less interesting question; the customers are more interesting. The pattern they report from coaching CEOs through deals is that acquisitions struggle not because the AI wasn’t good enough, but because the acquiring company changes who the product is for—often exactly what made the startup valuable in the first place. They frame the tradeoff as a question, not a settled answer: should a large acquirer preserve what made a startup unique, or is changing it simply the price of scaling? They say they’ve observed both approaches and one works far better, without naming which in the post. Their priority is to put customer definition and product identity on the table before locking in changes to scale.

## Work to read

- [Can We Build It? Is the Wrong Question.](https://www.linkedin.com/posts/gregcoticchia_productmanagement-productstrategy-innovation-activity-7498758951560359936-4Ppr)
- [Why aren't 'normal' people using AI agents?](https://www.linkedin.com/posts/gregcoticchia_why-normal-people-arent-using-ai-agents-activity-7498398960391475201-8kqA)
- [The wrong question when acquiring an AI startup](https://www.linkedin.com/posts/gregcoticchia_i-think-most-companies-are-asking-the-wrong-activity-7481311129831505920-9-Bq)
- [McDonald's, milkshakes, and the cost of knowing too much](https://www.linkedin.com/posts/gregcoticchia_product-leadership-domainexpertise-activity-7458238331017973760-brhQ)

## Brand relationships

- **Thomson Reuters** (mention)

## Partnerships

Greg Coticchia’s four LinkedIn posts cover product management practice: customer-problem discovery, AI product adoption, acquisitions, and domain expertise. They don’t describe sponsored content, paid collaborations, media-kit terms, rates, or a route for business inquiries. The posts mention advising CEOs through acquisitions, teaching product management, building and selling an eBilling company to Thomson Reuters, and turning down a largest-customer feature request to protect strategy. These are business credentials, not partnership offerings. Before planning a brand campaign, ask for a media kit, sponsored-post history, or partnership contact.

## Alternatives

- Omar Chaudhry: Choose Chaudhry when the need is specifically for agentic AI, not product-strategy discipline in general. Their posts draw a hard line between tools that assist and agents that execute: a tool waits for a prompt, click, or decision, while an agent detects a signal, decides, acts, and learns. A tool reduces effort; an agent absorbs responsibility. The closing argument is that the winner will have the fewest workflows still needing a human in the loop. Both posts work through recommerce and logistics examples, including pricing inventory in real time, negotiating freight, and rerouting supply. The second post offers a discovery method for a slow research cycle: simulate buyer personas with AI before spending on real users, then ask the model to attack the idea, identify where the economics fail, and surface edge cases. A B2B SaaS team with an agentic roadmap and a discovery process that takes weeks may find this closer to its problem than a general feature-gating exercise.
- Eric Seader: Choose Seader when a B2B SaaS company sells to legal, compliance, or professional-services buyers and needs to understand what they weigh when evaluating AI. Writing from inside a legal-tech vendor, Seader makes two buyer-facing arguments concrete: a low sticker price can hide the true cost, illustrated by Spirit Airlines' internal data sold in bankruptcy court to train AI models; and model control and continuity matter as much as capability, illustrated by firms building their own GPU clusters and lawyers writing their own tools. For a team whose product decisions hinge on how in-house counsel think about data ownership and vendor lock-in, this is useful voice-of-customer material. Like Coticchia, Seader argues from consequences rather than capabilities.
- Lars Ling: Choose Ling when the task is to convene the surrounding ecosystem, not run an internal product-strategy workshop. Their posts are organizing notices: a Sunday Walk & Talk ahead of Wasa Future Festival, and the festival's startup programme, run with the University of Vaasa and the Vaasa startup ecosystem, including a session on the connection from Vaasa to Silicon Valley. A brand seeking presence and relationships in a Nordic energy-transition and startup cluster, or wanting its name attached to a regional innovation event, fits this material well. The posts show how a local ecosystem is brought together, who the organizers are, and what the programme looks like.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/