How Many ISOs Are There in the U.S. and Canada? — ResidualMatch 2026 Market Estimate

"How many ISOs are there?" sounds like a question with an answer. It does not have one in any published source. There is no licence, no regulator filing, and no industry register that produces a clean count of active independent sales organizations in the United States and Canada. What exists instead is a set of adjacent datasets — card network compliance registries, acquiring M&A trackers, merchant acceptance counts, software payments revenue — each of which measures something real and none of which measures this.

So ResidualMatch Research builds the number instead of quoting one. This article gives a central estimate, a range, a confidence level, and the full arithmetic behind it, so that a buyer sizing a market or an owner sizing a competitive set can challenge a specific assumption rather than the conclusion in general.

Why No One Publishes the Number

The definitional problem comes before the statistical one. ISO status does not come from a public register; it arises from a commercial registration with an acquirer or processor sponsor, governed by card network rules. Nothing in that chain produces a public, deduplicated list of operating businesses.

• There is no ISO licence. A company becomes an ISO by contracting with a sponsor, not by filing with an authority that publishes counts.
• Entity formations carry no payments classification. New ISOs incorporate as ordinary LLCs or corporations, indistinguishable in state and provincial records from any other small business.
• One operating company can hold several registrations across multiple sponsors, and one registration can cover several brands or DBAs.
• Agent offices frequently operate under a larger ISO's registration and never appear as separate businesses, while dormant registrations and rebrands linger in lists long after the underlying business stops or changes.
• The category boundary is moving. A software company monetizing payments through a payfac-as-a-service partner performs an ISO's economic function without ever using the label.

That last point is not a technicality. McKinsey's January 2026 ISV research reported that U.S. payment-processing revenue flowing through independent software vendors grew roughly 20% annually over five years, reaching a projected roughly $16 billion in 2025 against roughly $6.5 billion in 2020. Some of that revenue sits with businesses that are economically ISOs and would never self-describe that way. Any count depends on where you draw that line, so we draw ours explicitly.

Why Visa registry entries are not a census

The most common shortcut is to point at Visa's Third Party Agent registration program or the Visa Global Registry of Service Providers and treat the entries as ISOs. Visa does require third-party agents performing ISO solicitation activities in the U.S. and Canada to be registered, and those programs are real and enforceable. But they are compliance registries, not business censuses, and reading them as a population count overstates in several directions at once.

• The registries span many service categories beyond ISO solicitation — including processors, gateways, encryption support, fulfilment and other third-party services — so entries are not homogeneous.
• A single operating company can appear multiple times through different registration types, sponsors, brands or legal entities.
• Registration status persists through periods of inactivity and is not a statement that a business is currently writing merchants or owns a portfolio.
• Agents writing business under a larger ISO's registration are economically distinct businesses that the registry does not separate out.
• Registration exists to manage risk and compliance obligations, not to describe market structure — so it says nothing about whether an entry owns residual economics, which is the only thing that matters for valuation and M&A.

Used properly, the registries are a useful order-of-magnitude cross-check: they confirm the North American ISO population is in the low thousands of registered agent relationships rather than in the hundreds or the tens of thousands. They cannot confirm a specific count, and we do not use them to produce one.

Why directory rows cannot be counted either

The same discipline applies to our own dataset. The ResidualMatch directory tracks roughly 3,000 North American payment companies, and it would be easy — and wrong — to present that as the ISO population. The directory deliberately includes multiple company types: processors and acquirers, sponsor banks, gateways, hardware and terminal vendors, ISVs, consultancies and service suppliers alongside ISOs and agent organizations. Counting rows would answer a different question than the one buyers and sellers are asking.

The ResidualMatch Estimate: ~1,200 Active Firms

ResidualMatch estimates that roughly 1,200 active ISO and merchant-services firms in the United States and Canada own residual economics on a merchant portfolio, with a plausible range of 900 to 1,800 and medium confidence. "Owns residual economics" is the operative definition: the firm holds a sponsor or processor relationship and earns residual income on merchants it controls contractually. That is the population that can be valued, bought or sold, which is why we use it rather than a broader marketing definition.

The range is asymmetric on purpose. The downside case assumes our activity screening is too generous and that more directory records represent dormant or duplicate businesses than we credit, which pulls the count toward 900. The upside case assumes directory coverage is thinner than we believe — particularly among small, regional, single-owner ISOs with minimal digital presence — and that a fuller census would find closer to 1,800.

United States and Canada

A country split is defensible only to the extent our directory geography is representative, so we publish it with lower confidence than the headline number. Within the directory, 2,717 of the roughly 3,000 tracked North American companies are United States based and 287 are Canadian; among records carrying an ISO or agent-organization classification the split is 1,938 U.S. to 199 Canadian, or roughly 91% to 9%. Applying that observed proportion to the firm estimate, ResidualMatch estimates roughly 1,090 active firms in the United States and roughly 110 in Canada, with confidence low to medium on the split specifically.

One caution on that split: directory coverage is not guaranteed to be equally complete in both countries, and the Canadian market has a more concentrated bank-acquirer structure that plausibly supports fewer independent firms per merchant than the U.S. Both effects point the same way — treat the Canadian figure as an order of magnitude, not a measurement.

Firms Are Not the Whole Market

The firm count answers one question. For most buyers, the more useful population is the agent and sub-ISO layer beneath it: independent books of business with real residual income whose owners can sell without transferring an operating company. ResidualMatch estimates roughly 9,000 such books in North America that are large enough to be worth transacting, with a wide range of 6,000 to 14,000 and low-to-medium confidence.

That estimate is inferred from typical agent counts per ISO across the directory population, scaled against the 37 million-plus U.S. merchant acceptance locations Nilson reported for 2025, and filtered to books carrying enough monthly residual to justify a transaction. Small changes in the "worth transacting" threshold move the figure substantially, which is why the band is wide and the confidence is lower.

What the Number Means If You Are Buying

• A population of roughly 1,200 firms is small enough to map. Systematic coverage of a defined segment — processor, geography, merchant type, size band — is achievable in a way it is not in most fragmented markets.
• It is also small enough that competitive intensity is real on the visible deals. Any target already running a broker process is being seen by the same buyers you are.
• Segment before you source. Roughly 9% of the firm population is Canadian on our estimate, so a North American mandate and a U.S.-only mandate are not materially different in deal count.
• Size the pipeline honestly. Against a firm base of this size, a target of two firm-level acquisitions a year requires visibility across a large share of a mostly private market, not one relationship.

The related question — how many of these firms actually change hands each year — is modelled separately in our buyer's guide to finding an ISO business for sale, which estimates roughly 60 firm-level acquisitions and roughly 600 residual-book sales annually in North America.

What the Number Means If You Are Selling

• You are not one of tens of thousands of interchangeable sellers. A population near 1,200 firms means a well-prepared portfolio with clean reporting stands out against a competitive set that is measured in hundreds, not thousands.
• Buyer scarcity is the wrong worry; buyer fit is the right one. The constraint in most processes is finding acquirers whose mandate matches your processor, merchant mix and size — not finding acquirers at all.
• Preparation is the differentiator. In a small, private market, the sellers that transact fastest and best are the ones who can produce residual reporting, contracts and attrition history without delay.
• Know your number before the first conversation. Establishing a defensible valuation range is the single highest-leverage step before any buyer contact.

A Living Estimate

This is a modelled view, not a measurement, and we treat it as revisable. Directory coverage improves as we verify and enrich company records; the ISO/software boundary continues to move as more payments revenue flows through ISVs; and consolidation steadily moves books upward into larger platforms. We update the figures when the evidence changes rather than defending a published number. If you operate in this market and believe an input is wrong, the methodology above is deliberately explicit enough for you to say exactly where.