Most payment-processing M&A looks rigorous and tells you nothing. The bulk of diligence hours go to contracts, projections, and trailing EBITDA. The operational layer gets a checkbox. In payments, that is exactly backwards. The real value lives in the platform underneath the P&L — the data, the integrations, and the operational detail that makes revenue repeatable. Miss it, and you are buying a narrative, not a business.
I have been on both sides of the diligence table, and the pattern is consistent. Buyers review the customer contracts and the financial model and stop there. They take management’s word that the revenue is sticky, the relationships deep, the technology solid. What almost nobody does is dig into how the revenue is actually produced. Is a critical integration maintained by a single engineer who knows where the bodies are buried? Is a key vendor relationship held in one founder’s head? Is the workflow documented, or is it tribal knowledge that walks out the door when that person leaves?
If the work depends on a person rather than the platform, and that knowledge is not encoded anywhere, the forecast is fiction. That is the sausage-making. Most of the deal-makers I have sat across from have never made sausage — they have never run an operating company, never been on the ground figuring out how the pipe connects to the valve. It is not a character flaw. It is a blind spot, and it is structural: you cannot diligence a mechanism you have never operated. So the numbers get read, the narrative gets trusted, and the machine that produces both goes unexamined.
The other thing they underweight is the platform itself. Strategic M&A is not only about the financial case, though the financial case has to be there. It is about what the underlying technology is and whether it can be leveraged at greater scale or across a different customer base. Data, software workflow, and integrations are the next form of leverage. It is the same playbook the AI labs are running: ingest data, learn from it, build new products on top. When you buy a payments business and look only at the P&L, you are treating a software company like a vending machine.
Financial services is, at its core, pipes and plumbing. Nothing sexy, but it takes a lot of code and a lot of maintenance. AI is collapsing the cost of building on and around those pipes, which means the acquirer who already owns the integrations and the data can move on them far faster than a competitor starting from scratch. The value of existing plumbing goes underappreciated precisely because it does not show up cleanly on a balance sheet. It is also what determines whether you can actually scale the business once you own it.
None of this means you ignore the financials. The financial element must exist. A great platform that does not make money is a hobby. The discipline is in the weighting.
So if you are evaluating a payment processor or an ISO, here is where I would spend the diligence hours. Start with an integration inventory: what systems talk to what, who maintains them, and what happens to recurring revenue if one breaks. Map the key-person dependencies: which roles, if vacated tomorrow, would stop revenue from flowing. Audit the data rights: can you actually use the transaction history and customer behavior data the way the investment thesis assumes, or do the merchant agreements and network rules say otherwise? Then read the sponsor-bank and processor agreements for the terms that decide whether the business survives the transaction at all: assignment and change-of-control clauses, exclusivity, term and termination rights, and how much of the volume sits with a single counterparty.
Those answers will not show up in the model. They determine whether the model means anything.