The product-building culture shift—minimizing the work around the work with Intercom’s VP of product
The forced reset: how do we minimize the work around the work?
Published
25 January 2026
Content
Creative
We’re in the midst of a culture shift when it comes to building products.
For Intercom’s VP of Product, Brian Donohue, this shift has come with an unintentional downstream effect: minimizing the work around the work. With an 11-year tenure at Intercom, Brian has witnessed and contributed to the evolution of a shifting customer service industry. He describes the last few years as totally re-energizing.
We sat down with Brian to discuss the forced reset of the last three years, the evolution of team structures, chatbots, and the ground-level reality of building in 2026.
Converging product roles in the new frontier.
Can you tell me a little bit about yourself and your team at Intercom?
Brian: I run the product management team for the core product at Intercom. I’ve been here over 11 years—a good long while. Like a lot of us in the R&D leadership team who have been here around that decade timeframe, I’ve been totally re-energized by the last three years.
Everything has turned on its head. The opportunity is massive, and it almost feels like starting over, even from within the same company. It’s an incredibly fun time.
The role I have is exactly what you’d expect. It involves the high-level strategy of what we are building and why, the execution to ensure we are actually building a great product, and the customer validation needed to ensure it's working before restarting that loop. Finally, a big part of it is building the actual organization that supports getting all of that done.
You’ve mentioned a shift in how products are being built lately. What does that look like for you at Intercom?
Brian: One of the interesting things that has happened over the last three years is a lot of indirect product cultural change. It’s not just because AI lets you do things you couldn't do before; there is a whole product-building culture shift that has come around as well.
For us, it’s about minimizing the work around the work and focusing on the core work itself. I think this is a broad macro trend, and it’s been liberating. As a product leader, you're now expected to get closer to the heart of what you and your team are doing and not spend so much time around the fringe.
Real-time cognitive adaptation: user sentiment is no longer static.
Product roles have experienced a convergence lately. From your vantage point, how have teams actually changed in the last few years, and what are your predictions for the future?
Brian: We’re right in the center of this fluidity. For a while, everything was thrown into the air. We’ve seen engineers step in as PMs and designers shipping code. I think that role fluidity is a durable part of the future. But the work that needs to get done hasn't changed that much—unless you've got a lot of unnecessary cruft that you can now cut out.
For example, we don’t write PRDs anymore. The goal isn’t necessarily to get AI to write the PRD—it’s to realize you didn’t need the document in the first place. You still have to figure out what to build, write it down in some shape or form, talk to customers, and validate the product. That requires human judgment.
We’re moving away from the cookie-cutter approach to building products. The old way was that you had, let’s say, ten teams; they had their roadmaps, and you basically carved out your product roadmap based on the size of those teams. This model doesn’t work anymore. It’s becoming less about a fixed team structure and way more about adaptability—looking at what the specific project needs, identifying the necessary roles, and then seeing who we have available to execute.
The team is a very powerful unit, especially when you have good relationships and good collaboration. Humans like to operate in teams. Plus, it’s important to have some durable ownership of product remit and expertise in certain areas of the product. Because of this, I think we’ll continue to require some form of team structure, but I think the adaptability and fluidity of roles is a durable thing.
If speed to market is no longer a competitive edge, what differentiates the winners today?
Brian: I think there are four main pillars of a “winner” right now:
- Product vs. consultingware: One major thing that sets companies apart is whether they’re creating an actual product or ‘consultingware’, an offering marketed as a standalone product that actually requires extensive, manual customization to function. The line between consulting and product has gotten way blurrier with AI. A lot of the work is heavy consulting—which is valuable and needed—but you have to be clear on what you’re actually getting. The problem with the "consultingware" approach is that it’s not scalable and its value eventually wanes. While a customer might get the benefits of a highly customized product, they can’t truly own it, iterate on it, or build on it, so they lose the "build vs. buy" advantage. They might as well have just built it internally. Plus, if your product hasn’t improved in a year because you’re too busy doing custom work, you’ve lost the benefit of being a product company.
- The speed of the feedback loop: This increases the value of product judgment. With a speedy feedback loop, there's more product judgment to be able to make the right decisions. I’d say this is an essential ingredient rather than a differentiator.
- Quality: I think the actual differentiator in the space when everyone can build fast is quality; it's the actual performance of your product. It’s easy to build an AI demo that looks “good enough.“ But does it pass a deep test? Can you actually use it if it’s business-critical? It takes time to build a good AI product. It is fast to build a superficial AI product.
- Deep technology stack ownership: As a short-term, medium-term, and even long-term investment, we believe you have to go deep into the full technology stack. That’s why we build our own models. Instead of using off-the-shelf models that anyone can use, a real differentiator is building one that's tuned for your specific use case based on all the data that you have; this uniquely enables you to build a great product. Actually owning the full stack will give you the product performance and efficiency that are essential.
Chatbots used to be the “annoying” part of the internet. How is user sentiment shifting?
Brian: We have to hold our hands up: we helped build the world of “annoying chatbots.” They were essentially virtual IVRs that people tried to bypass as fast as possible. But the shift we’re in now is dramatic.
When people realize they’re talking to a new-generation bot, their behavior changes. They move away from “Google keyword” shorthand and start writing full sentences and paragraphs. They get the back-and-forth dynamic. I saw a quote from an 86-year-old customer who had a ten-minute call with Fin, our AI voice agent. She told it, “You sound like a real person. It’s very artificial intelligence terrific.”
The irony is that right now, we have a complete role reversal. The bot sounds fluent and natural, while the human—still stuck in the old mental model of “dumb bots”—responds in one-word barks like “Refund. Return yesterday.” We’re watching the human brain adapt in real-time.
How will you navigate the uncertainty of the future?
What excites you most about what’s coming next?
Brian: The sheer uncertainty. We are three years into this wave, and there are still diametrically opposed conclusions that feel equally plausible.
Look at a product like Cursor. They have unbelievable growth. But in six months, will they be a durable powerhouse, or will code generation become so good that the tool itself is commoditized? You can say the same for OpenAI. Are they the dominant force for the next 15 years, or will the application layer move so fast that the model becomes a commodity?
Five years ago, the winners were predictable. Today, all the chips are back on the table. The openness of the field is what makes this fun. It’s a crazy, stressful environment, but that uncertainty is the prize.
Subscribe to Outlier
Juicy, inspiring content for product-obsessed people. Brought to you by Dovetail.