8 Commercial Weaknesses Quietly Slowing Fintech Growth
The technology scales. The commercial discipline around it usually does not — and the gap only becomes visible when growth stops paying for it.
- Fintech firms scale technology brilliantly and commercial discipline much less so. While funding was cheap that was rational. It is no longer.
- The eight below describe where the gap opens: pricing without knowing cost to serve, a customer base nobody has segmented by profit, compliance treated purely as cost.
- The position worth holding is not “we are growing fast”. It is “we know which customers make money, at what cost to serve, and we price accordingly”.
UK fintech has been genuinely good at building things. Products ship fast, engineering is strong, the customer experience is often better than the incumbents can manage.
What has scaled far less well is the commercial discipline underneath. While capital was cheap and growth was the only measure that mattered, that was a reasonable trade. Investors funded acquisition and asked about users rather than about margin.
That period has ended. The questions now are about profitability per customer, cost to serve, retention economics and whether the business works without the next round. Firms that built the commercial discipline while they were growing are answering easily. Firms that deferred it are discovering how much has to be reconstructed.
The eight below describe where the gap usually opens. None of them is a technology problem, and none is fixed by shipping faster.
Pricing was set against competitors, not against what it costs to serve
The pricing came from the market. What incumbents charge, what challengers charge, what would win the deal. It is defensible and it is how most fintech pricing gets set.
What is usually missing is a genuine view of what each customer costs to serve. Support contacts. Compliance checks. Payment processing. Failed transactions. Account maintenance. Those costs vary enormously between customer types, and the pricing rarely reflects it.
So some segments are priced well below what they cost and the average conceals it. Growth in those segments actively destroys value, and the faster it grows the worse it gets. The fix is to build a cost-to-serve model by segment before the next pricing decision rather than after the next investor question.
Nobody has segmented the base by profit
The customer base is segmented — by product, by channel, by size, by how they were acquired. All of it useful for marketing and product decisions.
What almost never exists is segmentation by contribution: which customers actually make money once cost to serve, support load and compliance overhead are counted properly. Without it, decisions get made on averages, and averages hide the shape.
The consequence is that acquisition spend keeps buying more of whichever segment is cheapest to win, which is frequently the least profitable one. The fix is genuinely a few weeks of finance work, and most mid-market fintechs have not done it because nothing forced them to.
Onboarding is manual behind a slick front end
The customer sees a clean digital journey. Behind it, a meaningful share of applications go to manual review. Identity checks that need a human look. Anti-money-laundering alerts to clear. Documents to verify by eye.
Each intervention costs real money and real time. It also sets a ceiling on growth, because the operations team scales in a straight line with volume while the technology does not.
Nobody usually owns this number. The product team owns the journey, operations owns the queue, and the cost per onboarded customer sits between them uncounted. The fix is to measure the manual rate and the cost per intervention, then treat reducing it as a commercial project rather than an operations one.
While capital was cheap, deferring the commercial discipline was rational. The bill for that decision arrives at the next raise.
Compliance is run as a cost line rather than as something that wins work
The compliance function is real, growing and expensive. Regulatory change, reporting, monitoring, audit, remediation. It is budgeted, managed and treated as the price of operating.
That framing misses what it could be. Corporate customers ask for evidence of controls during procurement. Partners run diligence before integrating. Investors examine it closely. Firms that can demonstrate this well win business that firms which cannot will lose.
The same spend, designed with both purposes in mind, returns differently. The fix is not more budget — it is deciding at the point of investment that the money should buy two things. Almost nobody in the mid-market frames it that way.
Knowing which customers make money is the first step. Pricing accordingly is the next.
A free Executive AI Opportunity Assessment. Forty-five minutes, online, with your leadership team. We arrive with a model of how firms in your sector make money and where AI has moved those numbers — and you tell us where it is wrong.
Book a free assessmentGrowth targets are set on volume, so the mix quietly worsens
The commercial team has targets. New customers, transaction volume, balances, accounts opened. Clear, measurable, and everybody understands them.
What the targets do not carry is any quality dimension. A customer worth nothing counts the same as one worth a great deal. So the team optimises for what is easiest to win, which is rational given how they are measured.
The board sees growth. The finance director sees margin failing to follow. The gap between the two widens quarter by quarter and gets explained as a timing issue. The fix is to put contribution into the target, not just volume — which is unpopular, straightforward, and one of the highest-return changes available.
Paid channels have inflated across the sector. Most firms have felt this directly on their own numbers.
Not because it is small, but because most firms have never segmented by contribution.
Usually higher than the product team assumes and higher than the board has ever been shown.
Permanent and growing, and framed as overhead rather than as something that could win work.
This describes a pattern we see rather than a measured dataset. Your own figures are the ones worth putting against it — and the exercise of producing them is usually more revealing than the comparison would be.
Partnerships get signed before anyone models the economics
Partnerships matter in fintech. Distribution deals, embedded finance arrangements, referral agreements, white-label relationships. They are how many firms reach scale.
They are frequently signed on strategic logic. This partner gives us reach into a segment we cannot access alone. Revenue share gets negotiated. Integration gets built.
What is often missing is a model of what the partnership actually returns once the integration cost, the support load, the compliance overhead and the revenue share are all counted. Some partnerships are excellent. Some consume more than they produce, and nobody finds out because nobody reports partnership contribution separately. The fix is a standard model applied before signing, and reporting per partner afterwards.
Three questions before the next board meeting
- Can you show, this week, which customer segments contribute after cost to serve? If it needs building, decisions are being made on averages that hide the shape.
- What share of onboarding needs a human, and what does each intervention cost? If nobody owns that number, nobody is reducing it.
- Which of your partnerships would survive an honest contribution analysis? If you are not sure, at least one probably would not.
The finance function reports history rather than shaping decisions
Finance is competent. The accounts close on time, the board pack is accurate, the investor reporting is clean. It does what a finance function is supposed to do.
What it usually does not do is sit inside the commercial decisions. Pricing gets set by product and commercial. Partnership terms get agreed by business development. Acquisition budget gets allocated by marketing. Finance reports the consequences afterwards.
The consequence is that the economics get discovered rather than designed. The fix is to put finance into the decisions before they are made, which needs commercial finance capability rather than reporting capability — a different hire, and one most growth-stage fintechs make later than they should.
Nobody owns the commercial design as a whole
The most structural of the eight. Product owns the product. Engineering owns the platform. Commercial owns revenue. Finance owns reporting. Compliance owns the regulatory position. Every function is well run.
What does not exist in most mid-market fintechs is anyone who owns how the commercial model fits together — pricing against cost to serve, segment mix against acquisition spend, partnership terms against contribution, compliance investment against what it could win.
So each of those decisions gets made well inside its own function and badly as a system. That absence is what the other seven describe. The fix is to make the commercial model someone’s explicit responsibility at executive level, with authority across functions rather than inside one. Most firms appoint a chief revenue officer who owns sales, which is a different job.
What this means
None of these eight are technology problems, and none of them slow a fintech down while capital is cheap. They become visible exactly when growth has to pay for itself — which is where most of the sector now finds itself.
The pattern is consistent. Each function optimises well within its own boundary, and nobody owns the way those decisions combine into an economic model. Pricing is set without cost to serve. Growth targets are set without contribution. Partnerships are signed without a model. Compliance is spent without asking what else it could buy.
The economics of fixing it are unusually favourable, because most of the work is analysis rather than build. Segmenting the base by contribution is a few weeks of finance effort. Measuring the manual rate in onboarding is a fortnight. What is expensive is discovering all of it during a raise, when someone else is asking the questions.
Which is the same argument we make about AI. Change the work before you change the technology — and before adding AI on top of a commercial model nobody has examined, it is worth knowing which parts of it currently make money.
Do you know which customers make money, or only how many you have?
Forty-five minutes with your leadership team, online, at no cost. We bring a working model of how firms in your sector make money and where AI has moved those numbers. You tell us where it is wrong. Then we tell you what we think — including “don’t invest”, if that is the honest answer.
Book a free assessment45 minutes · online · no cost · nothing to prepare
Continue reading
7 Operational Resilience Gaps Boards Now Expect Answers On
Where the framework and the operation most commonly come apart.
For chief executives9 Hidden Lifecycle Frictions Quietly Raising Your Acquisition Cost
Rising acquisition cost is rarely an acquisition problem.
For finance directors11 Pressures Mid-Market CFOs Are Quietly Managing
Structural rather than personal, which is why working harder does not help.