Maxim Godin

AI has a product problem. Not a model problem. Models are making capability leaps every few weeks but AI-native product innovation hasn’t kept up. Most products are forcing AI into existing UX patterns rather than rethinking an AI-native experience from first principles. Parallels to early mobile (2007-2010) - for years, mobile meant shrinking your website into your phone until Uber reimagined transportation. To be clear, there are lots of amazing AI-first products. See NotebookLM, Lovable, Stitch, Flow. But the pace of innovation on models has been faster, for a lot of reasons which I will explore in future posts.

554

48 Comments

Every product PRD going forward must be written for two audiences: the people using your product and the agents acting on their behalf. Most teams write PRDs around human workflows: jobs to be done, user goals, friction points. That’s necessary. But it’s quickly becoming insufficient. We’re entering a split-mode world. In the near future, every great product will require two modes: 1. How I work for humans 2. How I work for agents If you ignore the second, you’ll ultimately be solving the smallest part of the problem. Agents will outnumber users. Most “users” won’t be people, they’ll be LLMs choosing tools based on affordances, metadata, and reward functions. You won’t be chosen because of a slick UX. You’ll be chosen because you’re legible, callable, and reliably useful in a multi-agent stack. The products that work best for agents will get picked first. That’s the new first-mover advantage. We’re already seeing the early contours of this with MCP servers and autonomous workflows. This wave is here. The time to add this layer to your product sense isn’t “someday.” It’s now. To make your product great for agents as well as people, it needs to be usable by code with clear inputs, clear outputs, and a clear reason to call you. Your PRDs need to define this.

148

14 Comments

Are Product Ops people doomed from the start? I had a conversation with a former client last week that's been stuck in my head. "By definition, our job is to surface inefficiencies. How do we escape being disliked when we're there to show the organisation everything they're bad at?" He had a point. Product Ops roles typically get staffed because something isn't working. Often, we're hired to diagnose problems, challenge established ways of working, and ask uncomfortable questions about efficiency and effectiveness. We poke and prod. We inquire. We hold up a mirror. And then we wonder why we sometimes face resistance, why transformation feels like an uphill battle, why some stakeholders seem to actively work against the changes we're trying to implement. But if there are only two things I've learned in my years as a Product Ops specialist, it's these: 1️⃣ Resistance isn't personal 2️⃣ Resistance is inevitable When your role is fundamentally about revealing gaps and weaknesses, you're bound to make people uncomfortable. When you're asking teams to change habits that have worked "well enough" for years, pushback is natural. But there is a deep difference between organisations who look in the mirror and get indignant about what they see, and organisations who look to understand. The companies that succeed with Product Ops aren't the ones without problems - companies without problems don't exist. They're the ones who are mature enough to want the mirror held up. The ones that hired you not despite the discomfort you'll cause, but because of it. Remember: Any change - whether personal, professional, or organisational - is painful at first. Because in order to want to change you need to first realise that you have to change. Our job as Product Ops experts is to guide our teams through that change. And discomfort - even our own - plays a crucial part in doing that successfully.

88

18 Comments