How to build a licensing market outlook from observable demand signals without inventing a market size

There is no single official series counting the English market for business-licensing support. An outlook must therefore start with what each available measure represents and resist turning broad business counts into invented application or revenue totals.

What to take away

Define the market question

Separate at least four groups: businesses discovering possible permissions, applicants managing an active case, operators maintaining conditions, and organisations buying external advice or workflow services. A figure about one group cannot automatically describe another.

The Office for National Statistics publishes business activity, size and location datasets. They can provide a denominator under stated definitions and regional classifications. They do not say how many licences each business needs, how often it applies or what it spends on support.

Map the observable route

The Department for Business and Trade's Find a licence service is a UK discovery catalogue with activity and place filters. GOV.UK's council finder routes a postcode to a local authority site. Together they show that a user's journey can cross national discovery and local administration, not that demand is recorded consistently end to end.

A measurement plan should mark where data could exist: searches, result clicks, council enquiries, started forms, submissions, acknowledgements, decisions, renewals and support purchases. Record which stages are unavailable rather than estimating them from a neighbouring number.

Create a useful internal baseline

For an operator, count current sites, regulated activities, open questions, cases by state, conditions, upcoming dates and hours spent resolving gaps. For a support provider, add enquiries by task, qualified opportunities, delivered scope, rework and reasons for declining work. Use stable definitions for at least one reporting period before claiming movement.

The Government Service Manual's introduction to using data to improve a service links measures to user needs, service goals and a baseline. That approach is more defensible than collecting figures because they are easy to display.

Test an estimate before publishing it

Suppose an internal register contains 120 sites and 18 active licensing questions. It would be wrong to describe 15% of sites as needing a new licence: one question may cover several locations, another may concern a renewal, and some may be resolved as not applicable. Report the 18 open questions by their actual case states instead.

For external demand, document the numerator and denominator with equal care. A count of enquiries divided by all businesses mixes awareness, need and channel choice. Segment enquiries by activity, place and stage, then explain missing coverage. Do not extrapolate from one client's portfolio to England.

Use scenarios instead of a point forecast

A stable case assumes similar authority routes and focuses on renewal and recordkeeping demand. A digital-growth case tests whether structured forms or status feeds reduce manual coordination while increasing integration needs. A disruption case considers moved pages, supplier withdrawal, staffing gaps or an unavailable system.

For each case, write the observable trigger, operational effect and action. Do not attach a precise probability unless a documented method supports it. The purpose is to prepare decisions, not to decorate uncertainty with percentages.

State the evidence limit

Label UK statistics as UK-wide and local observations as specific to the participating authority or business. Avoid national claims from one council's publication. Recheck dataset releases, catalogue coverage and council routes before publishing a 2027 update.

The honest conclusion may be that demand is visible but market value is not. That is still useful: it tells the team which measures to collect, which comparisons are unsafe and what new evidence would change the outlook.

Before you act

Common questions

Why can't ONS business counts be turned into licence application or spending totals?

The ONS business activity, size and location datasets can provide a denominator under stated definitions and regional classifications. They do not say how many licences each business needs, how often it applies or what it spends on support. A figure about one group cannot automatically describe another, so broad business counts cannot become application or revenue totals.

How should an internal register of sites and open questions be reported?

Suppose an internal register contains 120 sites and 18 active licensing questions. It would be wrong to describe 15% of sites as needing a new licence, because one question may cover several locations, another may concern a renewal, and some may be resolved as not applicable. Report the 18 open questions by their actual case states instead.

What should each scenario in the outlook contain?

A stable case assumes similar authority routes and focuses on renewal and recordkeeping demand. A digital-growth case tests whether structured forms or status feeds reduce manual coordination while increasing integration needs. A disruption case considers moved pages, supplier withdrawal, staffing gaps or an unavailable system. For each case, write the observable trigger, operational effect and action.