Capable businesses, wrong tools, every time
The same problem runs through lending, payments and trade. I have spent a career getting to the bottom of it.
Twenty years into a career and taking a step back, you get clarity on the problem you’ve been circling the whole time.
I did not plan it this way. My path through financial services and technology looked, from the outside, like a series of different problems: lending, payments, trade, product, platforms, different markets, different companies, different scales. Each move felt like a new chapter. It was only recently, looking back with the benefit of some distance, that I clearly saw the thread running through all of it.
I keep ending up at the same gap. The capable business, underserved by tools that were built for someone else.
It has taken a different shape each time. But it is always, underneath, the same gap. And I have come to believe it is one of the most consistently underestimated problems in the entire economy.
The shape it took in lending
It started with credit, extending finance to businesses that needed it to grow. And the gap showed up immediately, though I did not yet recognise it as a pattern.
The problem was that lenders kept trying to force a small business into one of two moulds. Either it was assessed like a scaled-down enterprise, judged against financials, structures and reporting it did not have and did not need or it was pushed toward a consumer credit model, treated as a slightly larger individual borrower. Neither fit. A small business is not a small enterprise, and it is not a big consumer. It is its own thing, with its own rhythms of cash and risk, and the credit products on offer had no category for it.
So the capable operator running a real, profitable business would be turned away or priced punitively, not because they were a bad bet, but because the tool had no shape that matched them. The business was sound. The model was blind to it. That was the gap, though I would not have named it that way at the time.
The shape it took in payments
Later, working in payments across emerging markets, I watched the same gap appear in a completely different guise.
The orchestrators and payment service providers had built for card, for bank transfer, sometimes for vouchers. What almost none of them had genuinely integrated was the mobile money ecosystem, the way an enormous share of people in these markets actually move money, day to day. The infrastructure was built around the payment methods the industry was comfortable with, not the ones the customer was actually using.
Which broke the most fundamental rule there is: meet customers where they are, not where you wish they were. The tools assumed a payment reality that did not exist for most of the people who needed them, and then treated the resulting low adoption as a customer problem rather than a design one.
So these businesses did what capable people always do when the tools do not fit. They worked around them. Same gap. Different shape.
The shape it took in trading
Then came the wave of digitising the supply chain, connecting the large FMCG producers all the way down through distributors to the informal trader. And the intent was genuinely good. Done well, digitising that chain could benefit everyone in it.
But it forgot the fundamental principle of how these people actually trade: unplanned, and in person. Trade happens in the moment, on the pavement, in the shop, through relationships and conversation, not through a scheduled checkout flow.
So the tools that got built missed on both ends. Consumer-grade e-commerce journeys, designed for a planned purchase from a catalogue, simply did not map to how a trader buys. And enterprise-grade technology, imposed from the top, created friction where there needed to be ease, process where there needed to be speed. Neither reckoned with the basics of the environment either: patchy connectivity, unreliable electricity, the realities of operating where the infrastructure everyone else assumes is not a given.
Same gap again. Capable participants, moving real volume, handed tools designed for a context that was not theirs.
Why the gap persists
At some point the repetition stopped looking like coincidence and started looking like structure.
The gap persists because of how the economics of building tools actually work. It is easier and more profitable to build for the top of a market than the middle of it. The large enterprise can absorb complexity, pay for implementation, justify the cost. So the sophisticated tools get built for them. At the bottom, the volume is huge and the willingness to pay is low, so you get cheap, general-purpose, shallow tools that handle the basics for everyone and the specifics for no one.
The capable business in the middle, too substantial for the shallow tools, too small to warrant the sophisticated ones, falls into the gap. Not because nobody could build for them. Because the incentives point everywhere else.
Where the gap is now
Which brings me to where I have found the gap most recently, and most acutely.
Mid-market manufacturers and producers, businesses doing real revenue, employing real people, running genuinely complex operations, are living inside this exact gap right now. The enterprise operational systems are built for organisations ten times their size, priced and architected out of reach. The accessible tools handle the basics and nothing like the operational depth these businesses actually run on.
So they make it work, in the way capable businesses in the gap always have, until the business grows enough that the workarounds cannot hold, and the very capability that got them here becomes the ceiling they cannot break through.
It is the same gap I saw in lending. In payments. In trade. The capable business, underserved by tools built for someone else.
The answer the market is slowly arriving at
There is a name for the answer to this, and the industry has been circling it for a while: vertical SaaS. Software built deep rather than broad, for the specific reality of a specific kind of business, rather than general-purpose tools stretched to fit everyone and fitting no one. The whole premise of vertical software is that the gap I have been describing is real, and that the way to close it is to build for the actual shape of the business rather than forcing the business into a shape the tool already had.
I think that is right. It is, in a sense, the conclusion twenty years of ending up at the same gap has led me to.
But there is a structural reason the gap has stayed open even as the answer became clear, and it is worth being honest about. Venture capital is sized for scale. A fund of meaningful size needs opportunities large enough to move it, which biases capital structurally toward horizontal platforms and enterprise software, markets whose theoretical ceiling is enormous. A vertical, by definition, addresses a bounded market. It is not that vertical opportunities are bad. It is that they are shaped in a way the dominant funding model is not built to see clearly. The mismatch is not about merit. It is about how capital is sized and what that sizing makes visible.
So the answer exists, and the capital has been slow to arrive at it for reasons that have more to do with the structure of funds than the quality of the businesses. That lag is itself part of why the gap has stayed open as long as it has.
Something has changed, recently, in what it is now possible to build and at what cost, enough that I think the economics of closing this gap are finally moving to the right side of the line. That is a subject for another time, and I am not going to get ahead of myself here. But after twenty years of ending up in the same place, this is the first time it has felt genuinely closable.
Why I keep coming back to it
I have thought about why this particular problem is the one that follows me around.
Part of it is that it is genuinely large and genuinely underserved which, to anyone who likes solving problems, is the most interesting combination there is. But part of it is less analytical. The businesses in this gap are not failing. That is the thing that gets me. They are succeeding despite the systems around them, not because of them. And the distance between how they are doing and how they could do, if the infrastructure actually fit, is enormous, almost entirely invisible, and paid for in the working hours and stalled growth of people who deserve better tools than they have been given.
I have spent twenty years circling this gap. I do not think that was an accident anymore. And I know that I am not
done with it.