Merchant Category Codes: Who Assigns Yours, and Who Pays When It Is Wrong

Nearly every merchant we speak to believes, at least initially, that the merchant category code is something they select — a field on an application, a preference to be negotiated. It is not. The acquiring bank assigns the code, and it is required to assign the one that reflects what the business actually does.

Once that is clear, a great deal of otherwise baffling underwriting behaviour becomes legible. The scrutiny of your website, the questions about which products drive revenue, the interest in what you sell that you did not mention: none of that is caution for its own sake. The acquirer is determining a classification it will be held to, and it is the party that answers for getting it wrong.

Card network rules place the obligation on the acquirer to assign a code matching the merchant’s genuine business activity, and to perform risk-based due diligence at onboarding and on an ongoing basis to detect merchants operating outside their assigned code. This is not a one-time formality at signup — it is a monitoring duty that persists for the life of the relationship.

Where a merchant operates several lines of business, the rules are specific rather than permissive. The acquirer assigns the code representing the line with the highest sales volume, or assigns distinct codes to the separate lines. What it may not do is quietly classify a mixed business under whichever of its activities attracts the least attention.

Some categories carry more than a code. Higher-integrity-risk classifications come with registration requirements and specific controls, and in those categories the acquirer typically registers merchant URLs and re-attests to them periodically. That has a consequence merchants underestimate: the site reviewed at onboarding is a standing representation, not a snapshot. A new landing page, an added product line, or an affiliate microsite can put a compliant account out of alignment with its own registration without anyone intending it.

The enforcement design here is the part worth internalising. When a merchant is persistently coded wrong, the network’s remedy runs against the acquirer: it can face assessments and be compelled to correct the classification. Intentional miscoding, particularly to disguise a category the networks watch closely, is treated as a serious violation rather than a clerical matter.

That reframes the underwriting conversation entirely. Detailed questions about your product mix are not obstruction; they are an institution protecting itself from a liability it cannot delegate. Understood that way, the fastest path through underwriting is a complete and unflattering description of what you sell — because the acquirer will find the rest eventually, and finding it later is materially worse for both parties than finding it now.

It also explains why a correction, when it comes, arrives abruptly. The acquirer is not renegotiating with you; it is remediating its own exposure on a timeline set by someone else.

Somewhere in the shopping process, a merchant in a difficult category will be offered a more favourable code. It is usually framed as expertise or flexibility — a provider who understands the business and can classify it more advantageously than the last one.

Consider where the risk in that arrangement actually sits. The acquirer takes the assessment if the miscoding is caught, which is real but survivable for an institution. The merchant takes the consequences of the correction: repricing, a reserve that appears without warning, an account that no longer fits the bank’s appetite, and a mid-stream migration to a new processor. The party being done the favour absorbs the larger share of the downside.

A correctly coded account in a category a bank has knowingly accepted is a durable arrangement. An advantageously coded account is a deferred problem with a date on it that nobody has told you.

This category runs into the multi-line rule more often than almost any other, and usually without realising it. A business selling research compounds that also carries general supplements, or adds consumables, or bundles ancillary items alongside the primary product, has created exactly the mixed-activity situation the rules address — and the classification will follow the highest-volume line rather than the one the operator considers definitive.

The bundling question deserves particular attention here, because it is where classification and product presentation intersect. What is sold alongside a product contributes to how the whole offering reads to an acquirer, and adding items that imply a different use than the one stated is a recurring trigger for review. This is the same totality-of-evidence logic that governs how these businesses are assessed generally: the classification follows the observable facts of the operation, not the description the operator prefers.

The practical approach is unglamorous and effective. Present the full product mix at application with honest volume proportions, keep the reviewed site aligned with what was registered, and raise new lines with your provider before launching them rather than after. A code that matches reality is the cheapest stability available in a category where stability is the scarce commodity.