Data standard

What we verify, what we estimate,
and what we refuse to publish

Franworld sells decisions, not impressions. That only works if you can tell which of our numbers are hard, which are estimates, and which we deliberately left blank. This page is that answer — including the claims we removed after they failed verification.

Four labels, used consistently

Every figure on Franworld falls into one of these, and we aim to make which one obvious in context.

Verified

Traceable to a named primary source we checked — a regulator, a national franchise association, a government filing, or a published industry benchmark we can cite with its date.

Brand-supplied

Provided by the franchisor in its own listing: investment ranges, fees, unit counts, markets offered. Refreshed when the brand updates its listing. Directionally reliable, not audited by us.

Estimated

Our own calculation or editorial assessment, with the rubric published alongside it. Used where a real figure does not exist publicly — never presented as fact.

Not published

Data we could not verify to a standard we would want to be quoted on. We leave the gap visible rather than filling it with an impressive-sounding number.

Where each dataset comes from

DatasetSourceLabel
Franchise regulation, 35 marketsNamed national regulators and statutes — for example the FTC Franchise Rule (US), the Franchising Code of Conduct and ACCC (Australia), Decree 35/2006 and MOIT (Vietnam), the Franchise Law 2019 (Saudi Arabia), GR 42/2007 and the STPW regime (Indonesia).Verified
Franchise ecosystem (associations, expos)National franchise associations and their published event calendars — IFA, BFA, FCA, AMF, PFA, AFI, FLA and equivalents.Verified
Brand listings (cost, fees, units, markets)The franchisor's own Franworld listing.Brand-supplied
Industry statistics pagesPublished industry research with the publisher and year stated on each figure, including its age where that matters.Verified
Emerging-market classificationIMF World Economic Outlook grouping of emerging market and developing economies.Verified
Expansion Index & Franworld Index scoresOur own scoring engines, with the full weighted rubric published on each index's methodology page.Estimated
Franchise conversion benchmarksNo current public benchmark we could trace to its publisher's own report. The widely quoted portal lead-to-close figure is listed below, under claims we no longer use.Not published
Whether AI matching lifts conversionNo credible public benchmark exists. We are instrumenting our own funnel and will publish the number when we have enough data to stand behind it.Not published

How often data refreshes

  • Brand listings — whenever the franchisor edits them; scores and counts recompute automatically.
  • Directory and market pages — regenerated hourly, so counts and ranges on a page reflect live listings.
  • Regulatory summaries — reviewed when a market changes its franchise law; the applicable statute is always named so you can check it yourself.
  • Indices — continuously recomputed, published as a stamped annual edition.

Claims we removed after verification failed

Publishing this list costs us some persuasive-sounding numbers. We think a data platform that hides its retractions is worth less than one that shows them.

“APAC is the largest franchise market, at 30% of global franchises.” — Widely repeated, failed verification. We do not use it, including in pitch material. Our emerging-market focus is justified by competitive whitespace and data gaps, not by this statistic.

“Portal leads are resold 5–15 times.” — The resale pattern is real and documented; this specific multiplier is not. We describe the behaviour and give no multiplier.

“Portal leads convert at about 0.3% — roughly 350 leads per closed franchise.” — Credited to FranConnect's 2020 Franchise Sales Index, but FranConnect's own report pages do not state it; the only public source we found is a September 2020 article on Franchising.com. We listed it here as Verified until 15 September 2026 and no longer quote it as a benchmark.

“Qualified leads cost $50–90; high-net-worth leads $100–200+.” — Could not be verified. We publish only the $28–$100 (typically $30–$60) range that could.

Various answer-engine optimisation statistics — for example that structured data lifts AI-answer selection by a specific percentage. Every such figure we tested failed verification, so none appear anywhere on this site. We run our own monthly citation test instead.

Known limitations

  • We do not audit brand-supplied financials. A brand can overstate its numbers; our defence is disclosure scoring and investor-facing prompts to request the current disclosure document.
  • We do not yet measure franchisee satisfaction or unit-level performance, because we cannot do either at scale without selection bias.
  • Regulatory summaries are informational, not legal advice, and change without notice. Always confirm with qualified local counsel.
  • Coverage is deeper in the markets we focus on than elsewhere; a thin country page means thin data, not an empty market.

One standard across three sites

Franworld, franpulse.ai and franbase.ai are sister products from the same team and hold to the same rule: a figure is either traceable to a named source, labelled as our estimate with the method shown, or not published. If you find something on any of the three that breaks that rule, write to data@franworld.com and we will correct or remove it.

See the standard applied

Both of our indices publish their full scoring rubric — including what they refuse to score.

Franworld Index method →Expansion Index