Franchise opportunities desk research: evidence and limits

Mistake: Treating a desk verdict as permission to invest

For franchise opportunities, a desk conclusion should not sound like investment approval. Desk research can narrow choices, but cannot prove that an opportunity suits a particular buyer or will deliver a stated return; legal, financial and operational checks are still outstanding.

Mark the outcome preliminary and use it only to narrow options before due diligence. It is not a recommendation to proceed, a forecast, or a finding of suitability. Keep estimates and claims within their evidential limits; definitions and assumptions need an accountant approval before calculations appear in a presentation score.

A separate return on investment method can frame figures, but it cannot turn desk findings into a forecast. State what remains unknown, and distinguish documented facts, unverified assertions and matters requiring further advice. This boundary should accompany every conclusion rather than being left implicit.

Mistake: Letting a strong score conceal a failed gate

A candidate can look attractive on paper while failing a minimum requirement. Scoring before applying the gates risks allowing a strong brand impression to disguise a material problem; neither a high rating nor missing information can repair a failed gate.

Check the correct entity, written territory availability, complete draft agreement, full fee schedule, sufficient downside funding and access to independent advice before scoring. These are the six hard gates. Publish the weighted criteria and scoring anchors only after applying them.

Relevant criteria can include role fit, cash exposure, territory support, training, supply restrictions, renewal and exit. A five-point support rating is not reproducible unless the documents, timetable and reference evidence needed at every rating level are specified. Record missing material as unknown, not zero or average, and say whether the gap blocks advancement.

Keep financial calculations outside the presentation score until an accountant has approved their definitions and assumptions. A strong brand score cannot cancel a failed gate or settle an unknown. Where evidence is incomplete, the record needs to identify what is missing and explain why the gap matters to the decision.

Mistake: Choosing the buyer profile after finding candidates

An undefined buyer makes candidate inclusion arbitrary: the reviewer can reshape the field after seeing appealing names. That weakens comparison and makes the search difficult to reproduce, because the criteria may shift to favour a candidate already found.

Before opening provider pages, set the intended buyer, English location, working role, verified funding ceiling and excluded sectors using the entry checklist. Record search terms, filters, results, reasons for including or removing each candidate, discovery routes and the period covered by the search.

Also set capital, required income, time horizon and risk limits. Specify whether the search concerns a new territory or an existing unit, and whether the buyer will become a franchisee or convert an independent business. These are buyer-specific limits, not universal thresholds or a statement of current live thresholds at Franchise Scout.

Mistake: Calling a directory a complete picture of the market

A directory listing does not prove that the whole market has been searched. It may reflect membership or paid placement, while other opportunities may be found through different routes; treating one directory as comprehensive overstates what the research established.

Describe every discovery route used and keep paid prominence, membership and editorial selection visible as separate pathways. For example, a defined study could review relevant entries in the public BFA directory alongside candidates nominated independently by the buyer. That is a selected set, not the entire market.

BFA membership alone does not establish operational performance. The same brand can vary by territory, premises, staffing, permissions and finance. Public sources establish some details, but cannot settle every question about a specific opportunity or unit.

Mistake: Promoting a claim beyond what its evidence supports

A claim without traceable support can be mistaken for an established fact. Without consistent records, reviewers cannot compare opportunities reliably or revisit findings when records change; weak material must not be upgraded into certainty.

Use a selection checklist evidence log with the exact URL, publisher, access date, supported claim, geography and a captured copy where permitted. The row below illustrates the format; it is not a completed check or live Franchise Scout finding. Compare company accounts and financial values using the same unit and period.

A practical desk-research ladder, not a statement of live Franchise Scout grades, is:

The Companies House search guide describes available company information and warns about register limitations. Use it to identify the contracting entity and filings, not to certify operational quality. The Start Up Loans guide to buying a franchise offers prompts on territory, fees, support, restrictions and existing franchisees; turn each into a request, an accepted evidence type and a decision owner.

Search the Companies House register and check the brand through the Intellectual Property Office trade mark search, recording proprietor, status, classes and relevant marks. Ownership does not prove that a buyer will receive the described licence or territory. Request the current prospectus, draft agreement, accounts, fee schedule and territory map; the Companies House register search can help locate the contracting entity and filings, but cannot certify operational quality.

Mistake: Reporting documents as if they tested an operation

Public material is not a test of a provider or a franchise operation. Without a site visit, account, enquiry or reference interview, desk findings cannot establish implementation, service quality or return; reporting them as observed results gives the work more reach than it had.

Label each item as tested, observed or read, and state the limits of the work. If a test took place, record its task, version, conditions and date, and say whether an account was used, a site visited or a reference interview held. Do not imply a test occurred when the work consisted of reading public information.

Reading the public BFA franchisor directory is desk research, not an operational test. It does not create an account, submit an enquiry, contact a provider, inspect a private prospectus, verify testimonials or test a franchise operation, so it produces no operational test result or verified customer testimony.

The reviewer was the Codex editorial research desk. No payment, free access, referral fee or affiliate link was received. A named human editor, solicitor and accountant remain required before publication. Keep those limits with the findings so readers can distinguish desk work from checks that still need to happen.

Mistake: Treating selected testimonials as a network-wide result

One chosen franchisee account cannot describe an entire network. Selection, incentives and relationships can shape what people report, so a convenience sample does not establish a network-wide satisfaction rate or justify a general claim about experience.

The Competition and Markets Authority guidance on online reviews and endorsements says businesses publishing them must meet rules on authenticity and accuracy. For every quoted experience, record its source, date, role, relationship, incentive and selection route.

Ask current and former franchisees comparable questions, but do not presume their accounts represent everyone. Record how the people approached were selected, too. Report only what those accounts can support, and leave wider conclusions unclaimed unless independent evidence supports them.

Mistake: Changing the review method without recording the change

A revised weight, exclusion rule or evidence standard can change a result without any change in the opportunity itself. If the earlier method and result are not retained, readers cannot tell whether the candidate changed or the protocol did.

Version the protocol before each update, retain the earlier calculation, explain the change and identify the verdicts it affects. Set verdict rules before searching. This method offers no universal score cut-off and does not report current live verdict mapping at Franchise Scout.

Apply the four actions consistently:

Ask someone who did not research the candidates to recreate one score from the saved sources and anchors. A second reviewer should also reproduce the candidate set and score from the saved protocol. Refresh the review when territory, fees, documents or provider records change, and retain the previous version for comparison.