Defining your area and reading public sources

Inès's first search gave her a national figure for spending on cycling, with a great many zeros. She can do nothing with it: her customers live within twenty minutes of the workshop, not across the whole of France. What will decide her €48,000 of turnover is the number of households in her area that have bicycles repaired, and the number of workshops already serving them.

Defining the catchment area

The area is chosen before anything is counted, because it sets the scope of every figure that follows. For a neighbourhood workshop, you think in minutes on foot or by bicycle; for a home service, in the distance you are prepared to travel yourself; for online sales, the area becomes national.

The area must then be matched to a geography in which public statistics exist. The municipality is the simplest. INSEE also publishes some data at a finer level, the Iris, which is a sub-municipal division. An area drawn with a compass matches no dataset: it will have to be approximated by municipalities or Iris.

Two INSEE geographies help with the reasoning. The living area, or bassin de vie, is defined as the smallest territory in which inhabitants have access to the most common facilities and services. For a local business, it is often a good starting point: it shows where people are already in the habit of going.

The urban unit, for its part, rests on continuous built-up land: a municipality or group of municipalities with a continuous built-up zone, with no gap of more than 200 metres between two buildings, and at least 2,000 inhabitants. It helps to tell a town customer base from a scattered rural one.

Counting potential customers with the census

According to INSEE, the population census provides statistics on the characteristics of inhabitants (sex, age, occupation, nationality, employment and so on), of households (composition, size, family ties) and of their dwellings. It is the basis for counting who lives in the area, municipality by municipality.

The first choice is the counting unit. If the purchase is decided per household, like servicing a family's bicycles, you count households. If each person buys for themselves, you count inhabitants, possibly within an age band. Mixing the two inflates or crushes the customer base without anyone noticing.

The second point is the reference year. INSEE explains that results are produced each year from the five most recent annual surveys. A census figure therefore describes a five-year period centred on its reference year, not today's situation. It must be cited with that year.

Listing competitors and facilities

INSEE's permanent facilities database lists a wide range of facilities and services, commercial or not, open to the public, as at 1 January each year. It covers in particular shops and services to individuals, located by municipality, by Iris and even by coordinates.

INSEE presents it as a tool to study the structure of the supply of services in a territory, and mentions the calculation of facility rates by relating facilities to their potential users. That is exactly a founder's question: how many establishments already serve how many customers.

To look for competitors themselves, the official creation guidance suggests the public business directory, the Annuaire des entreprises, which shows businesses located in a territory and by field of activity through an advanced search. The guidance does not say how that field is designated: for each search, note the activities selected and their exact wording.

INSEE maintains the NAF, a classification of productive economic activities in which you can search for a code or a label. According to the same page, NAF rev. 2 applies until the end of 2026 and NAF 2025 from January 2027: a search by code that straddles the two states its version.

No database is exhaustive, and an activity code does not say what a business really does. A large sports store may repair bicycles without its code showing it; a workshop may have closed without yet being struck off. A dated on-site check remains essential to confirm the list.

Reading household spending

INSEE's household budget survey aims to rebuild the full accounts of households: their spending, broken down into a classification of about 900 budget items, and their resources. It helps you place a price within a household budget, and see which items weigh the most.

It has two limits that must be written into the study. It is an average over households living in France, which describes no particular area. And it measures spending, not an intention to buy from you. It serves to check that a price is plausible, never on its own to set the number of customers.

Fieldwork: surveys, questionnaires, interviews

The official guidance lists as tools field surveys, questionnaires, research, and analysis and conclusions drawn from the information collected. Public sources tell you how many households live in the area; only fieldwork tells you how many of them have the need you want to serve.

A useful questionnaire is short, asks about facts rather than opinions, and is run where the target customers live. Have you had a bicycle repaired by a professional in the last twelve months? gives a proportion. Would you like a new workshop? mostly yields politeness.

Two further factual questions are used in the next module. How many times, over those twelve months? gives a frequency. Where was the last repair done? shows what share of the households concerned currently go outside the area. None of these questions asks the household what it would do tomorrow.

The result is cited with everything needed to judge it: number of answers, places and dates of collection, exact wording of the question. A sample of a few hundred people questioned outside a single station is not a poll; it is worth what it is worth, and saying so honestly strengthens the credibility of the file.

Interviews complement the figures. The guidance advises talking to your competitors, entrepreneurs in the sector and your future customers. A repairer established in another town, who has nothing to fear from you, can tell you how many jobs one person handles per day and which months are quiet.

If your customers are businesses

The reasoning does not change when customers are businesses; only the counting unit changes. A maintenance service for shops counts establishments, not households. The same search by territory and field of activity that is used to find competitors is then used to count potential customers.

The activity code then becomes a customer filter. You must state which codes were kept and why: a florist and a bakery do not have the same maintenance needs, and grouping codes that are too different inflates the target. The list of codes kept goes in an appendix, with the date of the search.

In the field, the questionnaire becomes an interview: a few dozen managers questioned, with factual questions about their current provider, how often it comes and what it charges. A business that has signed a multi-year contract will not change provider this year; that can be measured and should be written down.

Writing the demand section of the study

The demand section is written as a short demonstration: the area and why it was chosen, the counting unit, the number of households or establishments with its source and reference year, the proportion that has the need with its survey, then the resulting potential customer base. Five sentences can be enough.

It helps to add what would make the result vary. An area widened to twenty-five minutes instead of twenty, a proportion measured in another district: the reader sees that the number kept is not the most flattering one possible, but the most defensible. Such honesty rarely costs anything; its absence often does.

Regulation and suppliers

The environment includes regulation. The guidance asks whether any specific legislation constrains or changes your strategy. The first question is the nature of the activity, commercial, craft, liberal or agricultural, because it determines who you deal with and which criteria apply.

The official guidance on the nature of an activity states that to be a craftsperson you must meet three criteria: a trade on the official list of craft trades, fewer than 11 employees at creation, and specific know-how proven by a diploma (such as a CAP or BEP) or a professional qualification. It lists bicycle repairers among its examples.

These criteria say whether the business falls within crafts; the same guidance then names the regional chamber of trades and crafts as the point of contact for craftspeople. The study states where Inès stands against each criterion and the formalities to plan for: their timing can delay the opening, and therefore push back first-year turnover.

On the supplier side, the guidance asks about constraints: whether fluctuating raw material prices affect your production, and whether suppliers in the sector hold a monopoly. Written, dated quotes attached as an appendix are the best answer: they set your variable costs.

Keeping a log of what you collected

Each time a source is consulted, it is logged at that moment, not when the study is written. Three weeks later, nobody remembers the exact page or the filter applied. A simple spreadsheet is enough; what matters is that it is filled in at every search, including those that produced nothing.

• The date of consultation and the exact address of the page or file
• The name of the source, its reference year and its geographical scope
• The filter applied: municipalities, activity codes, age band
• The number noted, copied exactly as it appears, without rounding
• A remark on what the figure does not say

This log becomes the appendix of the study and the raw material of the assumptions register. It is also useful six months later, when the forecast has to be updated: the same searches are run again with the same filters, and the gap between two readings becomes information in its own right about how the area is changing.