Every agency I've worked at has had some version of an innovation lab. Sometimes it was a literal room with beanbags and a whiteboard. Sometimes it was a team of three people with "Future" in their job titles. Sometimes it was just a page on the website that said "We don't just make ads, we make the future" next to a stock photo of someone wearing a VR headset. The details varied. The outcome was always the same: nothing shipped.

I've spent fifteen years watching this pattern repeat. An agency decides it needs to innovate. It hires some technologists, gives them a budget, puts them in a room away from the client work, and waits for magic. Six months later, there's a prototype that nobody asked for, solving a problem nobody has, built without any input from the people who actually understand the audience. The prototype gets shown at a conference, wins an award for "innovation," and is quietly shelved. The technologists leave. The agency hires new ones. The cycle begins again.

This is not innovation. This is theatre.

The fundamental error

The mistake is structural, and it's the same mistake every time: separating the people who invent from the people who understand audiences. Innovation labs are staffed with technologists and product people who know how to build things but have no instinct for who they're building for. Meanwhile, the strategists and creatives who spend every day thinking about consumers are locked in a different room working on campaigns.

The labs produce technology looking for a problem. The strategy teams produce insights looking for a solution. They rarely meet. And when they do, it's in a meeting where someone says "Can we make this into an AR experience?" and everyone nods politely.

The real innovation in advertising was always in the work itself -- not in a separate department with beanbags and VR headsets. Crispin Porter didn't have an innovation lab. They had creative people who understood the audience and were given permission to do something different.

Think about the genuinely innovative advertising of the last twenty years. Nike's "Write the Future." Burger King's "Whopper Detour." Spotify Wrapped. None of these came from innovation labs. They came from teams who understood the audience so deeply that they could see an opportunity nobody else could see, and then used technology as a tool to reach that audience in a new way. The technology served the insight. Not the other way around.

Build with strategy, not in a lab

The alternative to an innovation lab isn't "don't innovate." It's "innovate where the understanding lives." Put the building capability in the hands of the people who understand the brief, the audience, and the cultural context. Don't separate invention from insight. Merge them.

This is, incidentally, exactly what I've done since leaving the agency world. Every product I've built started with a strategy question, not a technology question. Modern Retro didn't start with "What can AI image generation do?" It started with "What would it look like if modern brands existed in the 1970s?" -- a question rooted in cultural insight about nostalgia, brand identity, and aesthetic sensibility. The technology was the tool. The understanding was the starting point.

🧭
The innovation test: If you can't explain who it's for and why they'd care in one sentence, it's not innovation. It's a tech demo. Every genuine innovation starts with an audience truth, not a technology capability.

CultureTerminal started with "People who care about culture want a single place to see what's happening across music, fashion, design, and tech." Trove started with "Your bookmarks reveal your taste, but no tool helps you see the patterns." In every case, the insight came first. The building came second. And because the same person held both the insight and the building tools, there was no gap between understanding the problem and creating the solution.

Why agencies keep getting this wrong

There are three reasons, and they're all uncomfortable to say out loud.

First, agencies don't actually want to innovate. They want to appear to innovate. An innovation lab is a signal to clients and to the industry that you're forward-thinking. It's a slide in the credentials deck. It's a talking point in the pitch. But genuine innovation is risky, messy, and often indistinguishable from failure in its early stages. Agencies, which run on client confidence and quarterly revenue, have very little tolerance for that kind of risk.

Second, the incentive structures are wrong. The people doing the innovating have no connection to client revenue. The people driving client revenue have no time to innovate. Innovation sits outside the P&L, which means it's the first thing cut when budgets tighten, and budgets are always tightening.

Third -- and this is the really uncomfortable one -- most agencies don't actually understand technology well enough to innovate with it. They understand technology as a media channel, as a distribution mechanism, as a thing you buy and place ads on. They don't understand it as a material you build with. There's a fundamental difference between knowing how to advertise on Instagram and knowing how to build something that people use. Agencies are extremely good at the former. They are, with very few exceptions, terrible at the latter.

The people who should be building

The best agency strategists I've worked with have extraordinary audience intuition. They can sit with a brief for ten minutes and identify the real problem, the one the client hasn't articulated. They can look at consumer data and see the story that the numbers are trying to tell. They can walk through a cultural moment and immediately see the opportunity for a brand to participate authentically. These skills are rare and valuable and developed over years of practice.

These are exactly the skills that should be directing what gets built. Not in a lab, divorced from the audience understanding that makes innovation meaningful. But in the work itself. The strategist who understands the audience should be the one who decides what to build for them. The creative director who has aesthetic judgment should be the one who shapes how it looks and feels. The planner who reads culture should be the one who identifies the timing and the context.

The best ideas come from people who understand the audience AND can make things. Not people who understand technology and are looking for a use case. Start with the human truth. Build from there.

AI tools have made this possible in a way it never was before. A strategist doesn't need to learn to code to build a product. A creative director doesn't need a development team to prototype an experience. The gap between "I understand what this audience needs" and "Here it is, I built it" has never been smaller. The question is whether agencies will recognise this and restructure accordingly, or whether they'll keep hiring technologists, putting them in rooms with beanbags, and wondering why nothing ships.

Innovation is not a department

If I could change one thing about how agencies think about innovation, it would be this: stop treating it as a separate activity performed by separate people in a separate space. Innovation is what happens when the people closest to the problem are given the tools and the permission to solve it in a new way. It's not a lab. It's not a team. It's not a line item in the budget. It's a capability that lives inside the people who are already doing the work.

The best advertising has always been innovative. Not because it came from an innovation department, but because it came from people who understood something true about the audience and found a way to act on it that nobody had thought of before. That's still the formula. The only thing that's changed is that the tools to act on it are now available to everyone.

Kill the lab. Give the strategists the tools. Watch what happens.