Item 19 Readiness: The Financial Performance Representation as a By-Product

Item 19 financial performance representation work gets expensive when the numbers have to be assembled for it. A network already keeping location financials on one chart of accounts has the average, the quartiles and the trail behind them as a by-product of the year, and the amendment becomes an extract rather than a project.

Most brands experience this as a season. The fiscal year closes, the amendment falls due in the months after it, and somewhere in that window a small group starts pulling numbers together.

The season is optional. Not the deadline — the scramble. What makes an amendment expensive is rarely the disclosure decision. It is that the data underneath does not yet exist in a usable form, so it gets built once a year, by hand, under pressure, then thrown away.

One chart of accounts, settled where the ledger connects

The requirement behind every credible number is dull. Every location records the same thing the same way — and in franchising the recording is done by independent businesses with their own bookkeepers and their own software.

The decisions that bite are small and definitional:

• What counts as revenue — gross, or net of discounts, comps and employee meals.
• Where third-party delivery commission sits: cost of sales, operating expense, or netted out of the top line before you see it.
• Cash basis or accrual.
• Calendar month or four-week trading period, and what that does to a year-on-year comparison.
• Which owner entity the numbers belong to when one person runs several stores.

None of that is a disclosure question. All of it decides whether your average is an average of the same thing.

Consider who keeps those books. Most of the estate sits with a minority of owners — as of 2025, 58.8% of US franchised locations were held by the 19.3% of franchisees running more than one, on FRANdata's outlook research — and an owner with six stores runs one bookkeeper across all of them, often across two brands, on a chart of accounts that was never yours.

The workable answer is to settle the mapping once, where a location's ledger connects, rather than to normalise afterwards in a spreadsheet. A mapping that lives in the connection applies to every period from then on and can be inspected by somebody who did not build it. One that lives in the head of whoever does the annual pull gets re-derived each year and remembered slightly differently.

Quartiles fall out of the benchmarking you already run

If you coach with peer comparisons, you already band locations by volume and compute a median for each band. The disclosure view is that same query with a different date range and exclusion set — and a brand that can filter to units open a full trading year, at any moment, has a cohort it can also use in a coaching conversation next Tuesday.

That is the part founders tend not to believe until they see it. The top and bottom quartile that felt like a decision to commission are numbers your field team has been reading all year for entirely different reasons. Why the distribution persuades a candidate's accountant where a single average does not is the argument under Item 19 FDD requirements, and I would rather link it than rehearse it.

Average unit revenue is the easy half. The questions that follow are about the population — how many locations are in the set, how many reached the figure, how a mature unit compares with one in its second year — and those are answerable in the same breath only if the population is defined by a rule the system applies rather than by a list somebody typed.

The Item 19 financial performance representation your system already produces

Being ready is narrower than it sounds. At any point in the year, for a stated period and population, you can produce the figure, the count of locations in the set, how many met or exceeded it, the exclusions with the date each was decided, and the extract behind all of it.

What belongs in the file behind those numbers is set out under Item 19 substantiation documentation. The question here is the division of labour: which of that file a system produces on its own, and which a person still has to write.

The system produces the extract, the population, the counts and the timestamps, holds the definitions in one place, and records an exclusion when the decision is made rather than reconstructing the reason months later.

A person still writes the scope memo, judges what a fair population is, and signs. Worth naming, because vendors blur it, and the blurring produces a confident number nobody can defend.

The trail is a property of the system, not a folder

The awkward test is not whether you can reproduce this year's figure. It is whether you can reproduce last year's, on last year's definitions, after this year's definitions changed.

That only works if a definition change is an event with a date and an author rather than an edit somebody made. The same applies to a restated month, a late correction, a store that transferred in April: each changes a number you have already disclosed, and the trail's value is that it can say what the number was when you disclosed it.

The property worth having comes up more often than the annual amendment. In a network of 45 locations — an illustrative brand, not a customer of ours — a candidate's accountant asks a sharp question in July about how the figure treats units that changed hands. A brand that can re-run the population that afternoon answers the same week. One that cannot says it will come back after the next renewal, which the accountant reads as an answer.

What none of this does for you

Three honest limits.

Scope is a judgement, and a clean system will produce a defensible-looking number from a badly chosen population. Consistency is not candour.

The consent question is real. A franchisee's ledger belongs to an independent business, and what you see is set by the agreement and by what owners agree to connect. A brand that introduces the connection as a compliance mandate gets thin, late, minimally compliant data — the same problem in a new costume.

Standardisation makes weak results legible too. More than one brand has tidied its financials and found its bottom quartile worse than the story it had been telling candidates. Information arriving at an inconvenient moment beats finding out during diligence.

None of which is a reason to buy software. An Item 19 financial performance representation is a thin reason on its own; the real one is that a network running on a single record produces this alongside everything else — the claim the rest of AI franchise management software rests on.

Which is the version worth aiming at: not a brand that survives the amendment, but one for which the amendment stopped being an event.

Extracts are cheap; the file behind the number is the part somebody eventually reads — what an examiner does with your substantiation.