Franchise Factbook
Trust & methodsscore_v1

How the Factbook works.
Everything here is built to be checked.

Where the data comes from, how the score is computed, how reviews will be verified, and what we do when we get something wrong. Looking for the tools instead? Start at for franchisees.

1,748
brands covered — updates automatically, never hand-edited
729
built from complete current FDDs
0.85
extraction-confidence bar — below it, the page says so above the figures
5 days
correction deadline, logged publicly in the changelog

Where the facts come from

Every US franchisor must publish a Franchise Disclosure Document under the FTC Franchise Rule — fees (Items 5–6), full startup costs (Item 7), litigation and bankruptcy (Items 3–4), earnings representations if any (Item 19), and outlet-by-outlet openings and closures (Item 20). We collect them from public state registries and franchisors, and extract with automated parsing plus human review. See how to read the numbers yourself.

Facts are presented verbatim — numbers come from the filing, with a link to the source document.
Absence is disclosed — if a brand shares no earnings data, we say so — that's information, not a gap.
Every fact is traceable — to its source, on every brand page.

Extraction confidence

Separate from the score, each recently-added brand carries a 0–1 confidence that the captured figures faithfully reflect the filing — highest from a primary regulator-hosted PDF, lowest from secondary aggregators. We publish it on the brand page rather than keeping it internal.

We'd rather show a figure with an honest caveat than hide it or present it as more certain than it is. Brands added before scoring show "extraction not scored" — which means unknown, not high. Want one filing read closely? Independent FDD review →

The Factbook Score

0–100, computed from filed data only — never assigned, never edited, never sold. Four components, each a percentile within the brand's category:

Owner retention
40%
Inverse of 3-yr average turnover rate, from Item 20
Unit growth
30%
3-yr net franchised-unit growth rate, capped
Disclosure quality
20%
Whether the brand discloses earnings (Item 19) and its depth
System maturity
10%
Years franchising (log-scaled)
Rules that never bend
  • Under 3 years of Item 20 history → "Score pending — insufficient history," never a low score for missing data
  • Under 3 scoreable peers in a category → "Score pending — insufficient peers"
  • No manual overrides — not for sponsors, not for anyone

A score is a percentile against the brands we can currently score in that category — and categories differ widely in how many that is. A score drawn from a small category is a narrower statement than the same number in a large one, and we'd rather say so than present the two as equivalent.

Four checks before any brand is scored

The score reads Item 20 tables straight from filings — so a brand failing any check keeps its figures on the page, shown as filed, with "Score pending" and the reason.

1
The table has to reconcile
Opening count + openings − terminations, non-renewals, closures and reacquisitions must equal the closing count. When it doesn't, a column usually went missing in extraction.
2
The years have to join up
One year's closing count should match the next year's opening count, within 2 outlets or 5% of the system — multi-state filings restate between years.
3
Closures must be disclosed as something
Zero terminations, non-renewals and closures in every year is far more often an unparsed column than a system that lost nothing — we don't claim a retention score from it.
4
The system must be big enough to rank
Under ten outlets, a single closure swings the rate so far that a percentile against peers stops meaning anything.
Owner reviews · coming

Only verified current and former owners, matched against the brand's own franchisee roster (Item 20 exhibits) or a document we view and don't retain. Anonymous publishing at 5+ reviewers; aggregate score at 20+.

What we don't do

Sell rankings or scores. Resell contact information. Accept payment to remove, soften, or delay facts, reviews, or changelog entries. How we make money →

Corrections

Published an error? We re-check against the source filing and correct within 5 business days, logging the fix in the profile changelog. This page is versioned; substantive edits are public.

Information, not investment, legal, or financial advice — always review the FDD with a franchise attorney