A bank innovation lab can have great people and still stall. The reason is structural: three forces no team out-works. Here is what they are, and the fix.
It's almost never the people. It's the structure they're working inside.
Walk into most bank innovation labs and you'll meet sharp operators. Ex-founders, product leads who've shipped at scale, designers who could work anywhere. The talent is real. So is the budget. Canada's largest banks spend an estimated $1–3 billion a year between them on technology and innovation, by industry estimates. And yet the output keeps disappointing the people who signed the cheque.
The instinct, when that happens, is to look at the team. Re-org it. Hire a bigger name to run it. Bring in a consultancy to "reset the culture."
That instinct is aimed at the wrong target. The lab isn't underperforming because the people are weak. It's underperforming because three structural forces are pulling against them, and no amount of individual talent out-works a bad structure.
Inside a large institution, the projects that survive tend to be the ones with the most senior sponsor, not the most customer demand. One industry survey put it bluntly: a majority of corporate innovation projects are shaped by executive preference rather than market signal (Innovation Leader, 2024).
One sentence holds the whole problem. The work bends toward the org chart.
A founder outside the building answers to one thing: whether a customer will pay. A lab leader inside the building answers to a customer and a sponsor and a steering committee and whoever's budget the venture touches. Each of those is a gravity well. The venture drifts toward whichever one pulls hardest, and that is rarely the customer.
A lab doesn't fail because the team can't build. It fails because the building keeps voting on what gets built.
Now the uncomfortable part. The operators who are best at venture creation, the ones who can take a thin idea and find the buyer in three weeks, rarely spend a career inside an internal lab. They start companies. Or they join studios where the upside is equity in something they built, not a band on a comp grid.
So internal labs end up competing for exactly the talent that's hardest to retain, against compensation structures the bank can't match without breaking its own pay bands. The lab can hire well. Holding the very top of that distribution, year over year, is a different fight. It's structural, not a recruiting problem you can solve with one good req.
This isn't a knock on anyone who works in one. It's an observation about incentives. People follow ownership. Internal labs, by design, have very little to offer.
The third force is the quietest and the most expensive. Internal labs are bad at stopping things.
Not because the people lack judgment, but because stopping costs someone politically. Whoever shuts a project down has to explain the months already spent. The sponsor who championed it loses face. The team gets reassigned, which feels like a demotion. Every incentive in the room points toward keeping a weak venture alive a little longer. So it limps on, quarter after quarter, well past the point anyone still believes in it.
The result shows up in the timelines. Analyses of corporate innovation have found internal efforts running well past the point the market signal justified before anyone pulls the plug (McKinsey). A studio reaches that decision in weeks. An internal lab can take a year and a half, and the difference isn't diligence, it's the cost of saying no.
Put a number on it. A venture that gets stopped at week six costs a studio under $50K (AOS operating model). The same idea, carried inside a lab until it finally collapses, can run $2–5M before anyone calls it (McKinsey). The lab didn't lose that money building the wrong thing. It lost it not being able to stop building the wrong thing.
If the problem were talent, you'd be stuck. Talent is scarce and expensive and slow to assemble. But the problem isn't talent. It's structure, and structure is fixable.
The fix is to put venture creation somewhere the three forces don't reach. Outside the org chart, so political gravity has nothing to bend. Tied to equity, so the operators who can actually do this have a reason to stay. And run on a cadence with hard decision gates, so stopping a weak venture is a scheduled, blameless event instead of a career risk.
That's the logic behind a corporate venture studio. The institution keeps what it's genuinely good at: customers, distribution, regulatory standing, capital, brand trust. The studio supplies what the structure inside a bank actively works against: speed, ownership-aligned operators, and the discipline to stop early. (We've written separately on how a six-week decision cycle replaces an eighteen-month process and on why a studio's first job is sorting, not building.)
None of this means an internal lab is a mistake. For some mandates (deep platform work, anything that has to live inside the core) it's the right home. But if the goal is to create new ventures on real customer demand, and the lab keeps stalling, look at the structure before you look at the people. The people are probably fine.
The money is already being spent. The only variable is what it buys.
Nothing here is an offer to sell a security or investment advice.
Field notes on venture building, AI, and capital. No spam, unsubscribe anytime.
By subscribing you agree to receive AOS Insights e-mails. We use your address only for this newsletter - see our Privacy Policy.
