Item 19 Average vs Median: Decoding Franchise Survivorship Bias

Why “Average Revenue” Is the #1 Liar in the FDD

When a sales rep points to a $500K or $800K “average unit volume” on the Item 19 page, that single figure is doing more rhetorical work than any other number in the FDD. It anchors expectations, frames the financial model, and shapes the offer you’ll accept.

Average revenue is the easiest number in the FDD to inflate without lying. The FTC requires Item 19 figures to be truthful and have a reasonable basis — it does not require them to be representative. A franchisor can take a system with deeply uneven performance, calculate a perfectly accurate arithmetic mean, and disclose a number that almost no franchisee actually earns. This is the dominant pattern in disclosures we read every week.

NASAA’s Rule: Median Must Accompany Average (and the Loophole Most Franchisors Use)

In 2017, NASAA issued its FPR Commentary — the controlling interpretive guidance state franchise regulators still apply. The key passage: if a franchisor includes an average in Item 19, it must also include the median or another measure of dispersion such as range or quartile breakdown.

An unaccompanied average tells you nothing about distribution, and distribution is the only thing a prospective franchisee cares about. In practice, franchisors comply with the letter and bypass the spirit three ways.

Subset averaging. The franchisor reports “average for units open at least 24 months” or “average for franchised units in the top 50% by revenue.” A measure of dispersion has been added — the disclosure has been stripped of the bottom half.

Tier-only reporting. Item 19 shows averages for top, middle, and bottom thirds, but thresholds are chosen so the bottom tier looks tolerable. A “bottom-third average of $220K” sounds livable until you realize that tier contains units earning $40K, $80K, and $120K.

Company-store blending. Company-owned units outperform franchised units because they get the best locations and trained corporate operators. Blending them into the system-wide average inflates the headline number for franchise buyers, who will never run a company store.

None of this is illegal. All of it is legible if you read the footnotes.

Worked Example: How 9 Franchisees at $300K and 1 at $2M Becomes a $470K “Average”

A ten-unit franchise system with the following annual gross revenues.

Franchisee	Annual Gross Revenue
Units 1-9	$300,000 each
Unit 10	$2,000,000
Sum	$4,700,000
Average (mean)	$470,000
Median	$300,000
Top quartile avg.	$2,000,000
Bottom quartile avg.	$300,000

Nine of ten franchisees — 90% of the population — earn $300K. The disclosed “average” of $470K is 57% higher than what 90% of operators experience. A buyer who builds a pro forma around $470K, expects to land 10-20% below it, and finds themselves at $280K in month 14 was always going to find themselves at $280K.

A disclosure-compliant Item 19 under NASAA’s commentary shows both the $470K average and the $300K median. The misleading version shows only the average. Both can be drafted by competent franchise counsel; only one lets you make a real decision.

Want help extracting median, top-quartile, and bottom-quartile figures from a specific FDD — and a clear flag when only the average is disclosed? Our $49 single-franchise report does that analysis and surfaces the dispersion the franchisor preferred to hide.

Survivorship Bias: The 40 Closed Units That Disappear from Item 19

The next layer of distortion is which units the franchisor counted in the first place. Item 19 disclosures typically report figures for currently operating units as of fiscal year-end. Units that closed, transferred under distress, or were terminated are excluded. This is survivorship bias, and it is built into nearly every Item 19 you will read.

Consider a franchisor that opened 50 units over five years. Ten closed. The Item 19 reports the average of the remaining 40. Were the 10 closures performing above or below the survivors? Almost certainly below. Units close because they ran out of cash, the operator quit, the franchisor terminated for underperformance, or the location proved unworkable. The closures are the bottom of the distribution, silently removed.

If those 10 closures averaged $150K before failing, the true system-wide average across all 50 units is meaningfully lower than the published average across 40. The figure isn’t false — it answers the wrong question. You aren’t asking “how do surviving units perform?” You’re asking “what is the realistic range of outcomes if I join this system?”

This is why comparing Item 19 to Item 20 closure and transfer rates is non-optional. A rising closure rate alongside a stable “average revenue” signals the disclosure is being scrubbed of bottom-end data each year.

The Top-Quartile Trap: When Franchisors Report Only the Top Performers

Beyond outright survivorship bias, a subtler filter is the operational threshold: “mean revenue for stores in continuous operation for 24 months or more,” “the figure shown covers locations that operated for the full reporting fiscal year,” “data is limited to shops classified as ‘mature,’” “owner-operated stores only.”

Each filter is defensible on its face. The cumulative effect is that the disclosed average reflects only franchisees who survived long enough and performed well enough to clear the filter. This matters most for first-year buyers, who by definition spend their first 24 months as exactly the unit type being excluded. Year-one Item 19 benchmarks almost always come in 20-40% below the “mature unit” averages that dominate published disclosures.

Common disclosure tactics — what they look like and what they hide.

Tactic	What It Looks Like	What It Hides
Mean without median	”Average gross revenue: $620,000”	Distribution skew; the midpoint may be $400K
Subset reporting	”Mean for stores open 24+ months”	Struggling first- and second-year locations
Top-tier disclosure	”Top quartile averaged $1.1M”	The bottom three quartiles
Survivor-sample basis	”Based on 84 of 96 currently operating shops”	The 12+ locations that closed
Company-store blend	”System-wide AUV including corporate stores”	Franchise-side performance is lower
Mature-cohort inclusion	”Locations classified as mature per footnote 4”	The ramp-period reality

How to Reverse-Engineer the Real Distribution (Item 19 + Item 20 + Item 3 Triangulation)

You will rarely get the full distribution handed to you cleanly. You can reconstruct it from three independent sections of the FDD.

From Item 19, extract the disclosed average, any median or quartile data, sample size, footnote inclusion criteria, and time period.

In Item 20, look for the unit counts at the start and end of each of the last three fiscal years, plus transfers, terminations, and non-renewals. A system reporting strong averages while shedding 8-15% of franchisees annually is exporting its bottom performers out of the dataset every year.

Then turn to Item 3 and read the franchisee litigation patterns. Multiple suits alleging Item 19 misrepresentation is direct corroboration that disclosed numbers diverge from operator experience.

A franchisor with a $500K average, 12% annual closure rate, “mature unit” inclusion, and pending franchisee misrepresentation suits is publishing a number that overstates the median outcome. A franchisor with a $500K average, $440K published median, 2% closure rate, all-unit inclusion, and a clean Item 3 is publishing a number you can build a plan around.

5 Questions to Ask Existing Franchisees About the Item 19 Distribution

Validation calls are where disclosed numbers meet reality. Most buyers ask “what’s your revenue?” and get answers that don’t help them locate themselves in the distribution. Ask these instead.

1. “Where do you sit in the system — top third, middle, bottom?” Forces a quartile answer and tells you whether the operator is representative or an outlier.

2. “How does your revenue compare to the Item 19 figures?” “About 80% of the disclosed average” tells you the disclosure runs high; “110% and middle of the pack” tells you it understates — rare but real.

3. “What did your first 18 months look like vs. the Item 19 sample criteria?” If Item 19 excludes units under 24 months and the franchisee lost money for the first 20, you’ve located the missing data.

4. “Of the franchisees who started when you did, how many are still operating?” Cohort survival rates are far more informative than Item 20 aggregate counts.

5. “What’s a realistic first-year revenue range for a new unit in a market like mine?” The figure operators give is almost always 25-40% below the disclosed average.

If the franchisor’s verified earnings claims line up with reality, 10-15 calls produce a consistent picture. For the underlying framework, see Item 19 financial performance representations and Item 19 red flags and misleading data presentation.

Want the full average-vs-median-vs-quartile breakdown for a specific franchise, plus survivorship-bias flags pulled from Item 20 trends? The $49 single-franchise report extracts the dispersion the franchisor disclosed and flags the dispersion they didn’t.

Frequently Asked Questions

What is the difference between average and median in Item 19? The mean is the sum of all franchisee revenues divided by the number of units. The midpoint — the median — is the middle value when revenues are ranked low to high. A few high performers pull the arithmetic mean above that midpoint. If nine franchisees earn $300K and one earns $2M, the calculated average is $470K but the middle value is $300K — the latter represents the typical franchisee far more accurately. Put another way: the middle store, not the brag store.

Why do franchisors report averages instead of medians? Averages are almost always higher than medians in franchise revenue distributions because top performers skew the mean upward. A higher headline number makes the opportunity more attractive. NASAA’s 2017 FPR Commentary requires a median or other measure of dispersion alongside any average, but franchisors work around it with subset averages and excluded categories.

What is survivorship bias in franchise Item 19? Survivorship bias is the error of analyzing only units that survived while excluding those that failed. Item 19 typically includes only currently operating units. If a franchisor opened 50 units and 10 closed, the disclosure reflects only the 40 survivors — and the 10 closures were almost certainly the worst performers. This systematically inflates every metric.

Can Item 19 numbers be manipulated without being illegal? Yes. The FTC requires Item 19 data be truthful and have a reasonable basis but gives wide latitude on presentation. Subset averages, excluding units under 24 months, separating company-owned from franchised locations, and omitting closed units are all legal techniques that produce misleading-but-defensible disclosures. The footnotes are where the truth lives.

How do I verify Item 19 earnings claims against the real distribution? Triangulate three sources. Read every Item 19 footnote to understand inclusion criteria. Compare Item 20 unit counts across three years — high transfer or closure rates signal survivorship bias. Call 10-15 franchisees from the Item 20 list and ask where they sit in the distribution, not just what they earn. If operators consistently report figures below the disclosed median, the published number is suspect.