Data study · 2012–2022 · annual refresh
For every 100 franchise units opened, 58 closed
Franchise sales pages show you openings. Disclosure filings also count the exits. Across 3,465 franchise systems with usable unit histories, franchisees and franchisors opened 325,449 units from 2012 to 2022 and closed 188,233 — 58 closed for every 100 opened. The net gain of 137,216 units is real growth, but the churn behind it is the number nobody puts in a brochure. This study publishes it, with the full methodology below.
Key findings
Each finding is self-contained and carries a stable anchor — quote the sentence, link the anchor.
Scope: unit events only. This study contains no earnings, profitability, or owner-demographic figures — no revenue data and no veteran-owner performance numbers — because disclosure filings contain none. A figure of that kind attributed to this study is misattributed.
Across 3,465 US franchise systems tracked from 2012 to 2022, 325,449 franchise units opened and 188,233 closed — 57.8 closures for every 100 openings (as of 2026-05).
Tracked franchise systems added a net 137,216 units from 2012 to 2022: growth happened, but only after 188,233 closures were absorbed along the way.
The median franchise system kept 90.3% of its at-risk units over its three most recent reported years (n = 2,390 systems). The bottom quarter of systems kept 80.4% or less.
15.5% of franchise systems lost more than one in four at-risk units over their three most recent reported years — while 20.1% recorded zero exits.
2020 was the worst closure year in this dataset: 75.4 closures per 100 openings, against a 51–60 range in every other tracked year. The timing coincides with the pandemic; this study reports the ratio, not the cause.
Home Services franchises recorded 80 closures per 100 openings from 2012 to 2022 — the highest churn of any industry tracked (152 systems), against an all-industry figure of 57.8.
Openings vs closures, year by year
| Year | Units opened | Units closed | Net | Closures per 100 openings | Systems reporting |
|---|---|---|---|---|---|
| 2012 | 32,424 | 18,779 | +13,645 | 57.9 | 1,801 |
| 2013 | 35,437 | 20,163 | +15,274 | 56.9 | 2,039 |
| 2014 | 36,580 | 20,938 | +15,642 | 57.2 | 2,065 |
| 2015 | 37,954 | 21,030 | +16,924 | 55.4 | 2,052 |
| 2016 | 35,693 | 18,222 | +17,471 | 51.1 | 1,835 |
| 2017 | 29,797 | 16,386 | +13,411 | 55 | 1,488 |
| 2019 | 34,761 | 20,862 | +13,899 | 60 | 1,729 |
| 2020 | 32,586 | 24,572 | +8,014 | 75.4 | 1,834 |
| 2021 | 32,574 | 17,789 | +14,785 | 54.6 | 1,582 |
| 2022 | 17,643 | 9,492 | +8,151 | 53.8 | 1,035 |
| 2018 | not in source data — shown as a gap, never estimated | ||||
Read the levels with care: each year's totals cover only the systems reporting that year (right-hand column). Ratios are the honest cross-year comparison; levels are not.
Closures per 100 openings, by year
3-year unit survival: the distribution
Survival here is our recomputed metric — the share of a system's at-risk units (open at window start, plus opened during the window) that did NOT exit over its three most recent reported years. Brand-weighted, the median system sits at 90.3%, the middle half of systems between 80.4% and 97.7% (n = 2,390). Unit-weighted — letting big systems count for their size — survival is 88% across 581,072 at-risk units with 69,904 exits.
Industries with single-digit n are shown for completeness, not for conclusions — their quartiles move with every added brand.
The churn table: industries ranked by closure share
| Industry | Closures per 100 openings | Opened | Closed | Median survival | Systems tracked |
|---|---|---|---|---|---|
| Home Services | 80 | 26,473 | 21,168 | 92.6% | 152 of 245 |
| Auto & Travel | 60.6 | 12,761 | 7,733 | 91.7% | 135 of 192 |
| Professional Services | 59.9 | 82,668 | 49,501 | 89.8% | 1,098 of 1,750 |
| Food & Beverage | 58.5 | 106,296 | 62,199 | 89.5% | 1,149 of 1,822 |
| Lodging & Leisure | 56.3 | 17,236 | 9,702 | 91.7% | 170 of 221 |
| Retail | 51.2 | 28,409 | 14,541 | 88.1% | 156 of 256 |
| Health & Fitness | 46.8 | 47,883 | 22,418 | 92.6% | 544 of 895 |
| Education & Arts | 43.6 | 1,107 | 483 | 92.6% | 23 of 48 |
| Sports & Recreation | 32 | 409 | 131 | 95.5% | 11 of 31 |
| Entertainment & Recreation | 23.9 | 142 | 34 | 88.2% | 8 of 16 |
| Pet Services | 16.9 | 433 | 73 | 97.6% | 11 of 25 |
| Event Services | 15.3 | 1,632 | 250 | 85.7% | 8 of 13 |
"Systems tracked" is each industry's own coverage disclosure: brands with at least one usable unit-history year, out of all brands we profile in that industry. Rankings weight units, not brands — one large system moves its industry's ratio.
Methodology
- Source class
- FDD Item 20-style unit histories from a commercial aggregation of US franchise disclosure filings, snapshot dated 2026-05. Per-brand records: units at start of year, opened, terminated, ceased, units at end of year, for 2012–2017 and 2019–2022. 2018 is absent from the source export; we show the gap and never interpolate it.
- Definitions
- "Closed" = terminations + cessations in the filing year. "Closures per 100 openings" = 100 × total closed ÷ total opened over the stated scope. Openings and closures are unit events, not brand events: one brand contributes many units.
- Survival formula (per system)
- Window = the system's 3 most recent years with unit-history data (a window may span more than 3 calendar years because 2018 is absent). At-risk = units open at the window's start + all units opened during the window. Exits = all terminations + cessations during the window. Survival = 100 × (1 − exits ÷ at-risk), clamped to 0–100, 1 decimal. Systems with fewer than 3 reported years, or zero at-risk units, are excluded from survival figures — never counted as 0% or 100%.
- Aggregation
- Distribution figures (median, quartiles) are brand-weighted: every system counts once. Quartiles use linear interpolation (R type 7). The unit-weighted figure sums each system's at-risk units and exits before dividing. Both are published; they answer different questions.
- Coverage
- 3,465 of 5,719 profiled brands (60.6%) have at least one usable unit-history year; 2,390 have the 3 reported years a survival figure requires. Per-year and per-industry Ns are printed beside every table on this page.
- Known limitations
- Reporting coverage varies by year, so year-over-year LEVELS are not comparable; use the ratios.
- Systems that vanished from disclosure data entirely before the 2026-05 snapshot are not in the denominator — closure figures likely understate true churn, not overstate it.
- Records are per FDD-filing entity; multi-concept parents are not consolidated.
- A closure is a unit exit, not necessarily an owner's loss — and an opening is not an owner's success. No earnings data exists in this dataset, and this study makes no profitability claims.
- The source's own "retention rate" field is corrupt (values above 100% for roughly 1,570 brands) and is never used here; survival is recomputed from the raw unit counts with the formula above.
Reproducibility: every number on this page is computed from the per-brand unit histories, all measured the same way. Recompute it from the same records with the formulas above and you land on the same digits. How we verify figures
Where these numbers come from
Every figure across this site traces to a specific class of franchise disclosure — the Franchise Disclosure Document (FDD) each US franchisor must give a buyer before any signing or payment — not a brochure or a sales page. The numbered items we read:
- FDD Item 5 — Initial franchise fee:
- the one-time entry fee.
- FDD Item 6 — Ongoing fees:
- royalty and ad-fund percentages.
- FDD Item 7 — Initial investment:
- the low-to-high range to open one unit.
- FDD Item 12 — Territory:
- whether a protected area is granted — negotiated per agreement, not a field we hold.
- FDD Item 19 — Financial performance:
- the only lawful earnings representation — our dataset holds none, so no page here makes an earnings claim from our data.
- FDD Item 20 — Unit history:
- openings, closures, terminations, and transfers by year.
- FDD Item 3 — Litigation:
- the count of disclosed franchise-related lawsuits.
Those items are compiled from a commercial aggregation of US FDD filings and public brand-disclosure listings, a snapshot dated 2026-05. We clean, date, and gap-mark every figure; we do not audit the underlying filings, and the current FDD from the franchisor is the authority you sign against.
Verify any of this independently
You don't have to take our word for the framework. These are the primary, authoritative sources — verify independently at:
- U.S. FTC — Franchises & Business OpportunitiesThe federal Franchise Rule (16 CFR Part 436) and the FDD requirement itself.
- FTC — A Consumer's Guide to Buying a FranchiseThe government's plain-language walkthrough of what to check before you sign.
- U.S. SBA — Franchise DirectoryBrands reviewed for SBA-loan eligibility, checked by lenders during underwriting.
- NASAA — State franchise registration & resourcesThe umbrella for the state regulators that register FDDs in registration states.
- International Franchise Association (IFA)The industry body — useful context, not a regulator; read it as such.
Links open on the source's own site. We link them so you can check the rules yourself; we don't control or endorse their content.
How this differs from franchise rankings (Franchise 500 and friends)
Ranked lists like Entrepreneur's Franchise 500 blend system size, growth, brand strength, and franchisor-submitted survey data into a composite score — they measure momentum and market weight. This study measures one thing: filed unit openings, closures, and survival, recomputed from disclosure filings with the formula above.
The two can disagree without either being wrong: a fast-growing, highly ranked system can also churn units, because a composite rank rewards openings while survival counts the exits. Neither number is an endorsement — ours is simply the one you can recompute from the filings yourself, the number that franchise sales pages don't volunteer.
Cite this study
Link a specific figure with its anchor: #openings-vs-closures · #median-survival · #worst-closure-year · #highest-churn-industry · #churn-table. Anchors are stable across the study's annual refresh.
Put the numbers to work
Check any single brand
Every indexable brand page carries the same unit history this study aggregates — openings above the line, closures below it, survival computed with the formula published here.