Precision needs something to stand on
Leonid Goriev
Founder Alty
June 4, 2026
A reaction to McKinsey's Global Banking Annual Review 2026, from the team that builds the products underneath these decisions.
McKinsey's 2026 Global Banking Annual Review has a title worth sitting with: Precision with speed.
The argument is hard to disagree with. Banks posted another strong year — net income up 7% to $1.3 trillion (McKinsey, via Idea Farm).
But the ground is moving underneath the numbers. Among the world's 1,000 largest banks and 1,000 largest fintechs, fintechs' share of joint revenue has climbed from 10% in 2021 to 17%. Generative AI reached 45% of US working-age adults two years after ChatGPT launched, rising to 55% the following year — a curve that took digital banking fifteen years to climb. The report's conclusion: banks can no longer afford to be slow followers. They need to move with precision and speed, and become what McKinsey calls multispeed organizations.
We've built banking products for fifteen years, including the first version of monobank and platforms now used by tens of millions. So I read that conclusion and thought: yes, and that's the easy part to say.
The hard part is that precision and speed aren't decisions a bank can make. They're outcomes a structure either permits or blocks. When a capable bank can't move quickly, the cause usually sits below ambition and talent: the foundation underneath has made every fast decision dangerous.
Speed is a property of the foundation, not the team
Here's what the "move faster" advice misses in practice.
When we meet a bank that has slowed down, the team is rarely the bottleneck. The people are good. The strategy deck is clear. What's actually happening is that years of incremental fixes have left the underlying system fragile, and now every change carries risk. So the organization does the rational thing. It hesitates. It adds review layers. It slows down because it has good reason to fear breaking something that works.
The report puts a number on the cost of this. Banks spend more on technology than the next four sectors combined, yet their productivity has lagged. McKinsey's own explanation is structural: where other industries reinvented themselves and jumped to the next curve, banks layered call centres, online banking, mobile apps, and now AI on top of the existing core. The spending was real. The core underneath it never changed.
You cannot instruct your way out of that with a "be faster" mandate. Speed returns only when the foundation becomes safe to change.
We saw this directly with GTBank. By 2021 it had been one of Africa's most innovative banks, and it still had the ambition. But its mobile platform had grown difficult to change without risk. The fix wasn't a redesign or a speed initiative. We mapped what was constraining the platform, simplified the product architecture, and rebuilt the technical foundation so changes stopped being dangerous. The app rating moved from 3.4 to 4.7 stars. The quieter result mattered more: the team got its release confidence back. The speed didn't come from a faster team. It came from a foundation that was finally safe to push against.
Precision is a decision problem before it's a capital problem
The report reframes precision as capital discipline — allocating line by line, rebalancing portfolios toward structurally advantaged segments rather than chasing scale. That's where the value it points to sits.
But precision in capital allocation rests on precision in decision-making, and that's the layer most banks never made legible. Most product failures we see aren't execution failures. They're decision failures: priorities that were never truly agreed upon, tradeoffs nobody made explicit, systems still designed for a business that no longer exists. You can't allocate capital with surgical precision on top of a decision process that runs on assumptions.
This is why, with a Swiss private bank, we paused product work entirely before touching the interface. A previous agency engagement had left them with deliverables but no direction they could safely build on. So we started underneath: which systems consolidate, where communication lives, what stays non-negotiable under Swiss regulatory standards. Only then did client-facing work begin — and it passed independent cybersecurity and regulatory audits. The precision McKinsey describes became possible only because the decisions underneath it were made defensible first.
Multispeed is right — but it needs a spine
The idea of the multispeed organization is the most useful in the report. McKinsey lays it out as three speeds: a core that runs incremental, proven work at roughly 90% success rates; an incubator layer testing new propositions at around 50%; and a frontier of seed-stage bets where maybe 20% survive. Different parts of a bank, deliberately running at different speeds.
The logic is sound. But notice what the report also finds: around 70% of the industry's innovation ideas cluster in the same crowded, high-popularity areas, and the bets on AI and digital assets sit in the highest-risk, lowest-penetration corner of the map. The appetite for speed is already there. What's missing is the structure to run three speeds at once without the fast lane quietly destabilising the core.
That's the real risk with "multispeed": it becomes a permission slip for incoherence — every team at its own pace, no shared logic holding them together. Multiple speeds only work if there's a single decision structure underneath that everyone can see and defend. Otherwise, it isn't multispeed. It's just fragmentation with a strategy label.
That structure is the thing we spend most of our time on, before a single screen gets designed.
What this means if you're inside a bank reading the report
The report states the bind plainly: banks now have to hold risk discipline and resilience and operate at significantly higher speed — something it calls, until recently, a contradiction in terms. The era of slow followership is closing. I'd add one line to that.
Precision needs something to stand on. Before the next transformation programme, the more useful question isn't "how do we move faster." It's "why has moving become dangerous, and what would make it safe again." That answer usually sits in the structure, not the roadmap.
If you're working through that, three places to start:
How we approach product decisions before execution — how we work
When an existing product needs to move again — improving a product
When you're building something new and want the foundation right from day one — new product
Leonid Goriev is the founder of Alty, a product partner for fintech and banking. Alty works with banks and fintechs on the decisions that shape what gets built, before the build starts.
Spotted by Yelyzaveta Yatsuk, our Partnership Manager.
Source: McKinsey & Company, Global Banking Annual Review 2026, "Precision with speed" (May 2026). Figures cited from the report and secondary coverage; see the full report.
