Sourced analysis

How much does a hire cost in France?

Nobody measures it. No French public statistic publishes a cost per hire, or a cost per application. The figure everyone quotes, “€5,000 to €8,000”, comes from a French blog post dated 26 April 2018, whose author already wrote that no French source existed on the subject.

What is actually published in France is the volume of planned hires, reported difficulty, shortages by occupation and the price of labour. This page brings those sources together, dated and linked, shows why no average can serve as a benchmark, and gives the method for calculating your own cost per application instead of borrowing someone else's.

In France, no public statistic publishes an average cost per hire: an SME with 50 to 250 employees therefore has to calculate its own. At Seeqle, a talent acquisition solution is available from €500 per month. To put that amount in context, divide the total cost incurred over a period by the number of applications received over the same period. This cost per application is the only benchmark that holds, because it can be counted and verified, and because it is calculated on your own volumes rather than a borrowed average.

2.28 million

Planned hires announced in France for 2026, down 6.5% year on year. France Travail, BMO 2026 survey, 21 April 2026.

43.8%

Share of companies expecting difficult hires in 2026, versus 50.1% in 2025. France Travail, BMO 2026.

6 in 10

Occupations facing strong or very strong shortages in 2024, versus 7 in 10 in 2023. DARES, Dares Résultats no. 5, 4 February 2026.

The canonical figure

What an order of magnitude produced in 2018 is worth today

The €5,000 to €8,000 per hire range is now repeated by hundreds of French pages, often without a source. It always traces back to the same text. Four reasons to stop quoting it as is.

It is eight years old

Published on 26 April 2018. Since then, INSEE was still measuring a 2.0% year-on-year rise in hourly labour costs in the second quarter of 2025. An order of magnitude from 2018 cannot be reused as is eight years later.

Its scope is not defined

Full cost or direct cost? Managerial or non-managerial? Large company or SME? With or without internal time? Each of these parameters changes the result several times over, and none is specified.

Its method is not published

No sample, no observation period, no definition, no primary source. The figure was copied from page to page until it became self-evident, without the chain of citations ever leading back to a measurement.

Its author said so himself

The article states that no French data existed on the subject at the time. The figure was originally presented as an order-of-magnitude estimate. It is now quoted as a measurement.

Break it down

A hire is five costs, not one

Talking about the cost of a hire makes no sense until you say which of these five items you are adding up. Three can be measured, one can be calculated, the last one you simply bear.

Media cost

The media budget spent to attract candidates, divided by the number of applications received. This is the cost per application. It can be read directly in the distribution tools, campaign by campaign, and it is the only item that moves within a week if you decide to act on it.

Screening cost

Reading, discarding, sorting. Applications received multiplied by the time taken to process one application, multiplied by the fully loaded hourly cost of the person screening. A hundred applications at four minutes each comes to nearly seven hours before the first interview, paid whether or not the role gets filled.

Recruiter time cost

Interviews, write-ups, coordination with managers, follow-ups, reference checks. No invoice makes it visible, so almost nobody tracks it. It is calculated as hours multiplied by fully loaded hourly cost, including managers' time.

Vacancy cost

What the role costs while it is unfilled: output not produced, overtime, temporary staff, longer lead times for customers. It depends on the price of labour, which INSEE tracks every quarter, and on the length of the vacancy, which is published nowhere at national level. You can calculate it in-house. You will not be able to compare it.

Cost of a failed hire

A departure during the probation period resets the four previous items to zero and extends the vacancy accordingly. It is the only one of the five that can be tackled upstream, by qualifying better before the interview.

Key takeaway

Only one of these five items can be measured in a reproducible way that is comparable from one campaign to the next: the media cost, in other words the cost per application. A claim about the cost of a hire in France that does not say which of the five it adds up cannot be verified.

French public sources

What INSEE, DARES, France Travail and APEC actually measure

Four institutions publish dated, verifiable data on recruitment in France. None of them publishes a cost.

France Travail

2026 labour-needs survey (BMO), published on 21 April 2026

2.28 million planned hires announced for 2026, down 6.5% on 2025. 43.8% of companies expect difficult hires, versus 50.1% the previous year. 23% of establishments plan to hire, versus 24% in 2025. The highest volumes are expected in health and social care (322,000 planned hires), hospitality and food service (319,000), retail (264,000), industry (211,000) and agriculture (193,000). View the source

DARES

Labour market shortages in 2024, Dares Résultats no. 5, 4 February 2026

Six occupations in ten face strong or very strong shortages in 2024, versus seven in ten the previous year, representing a little over half of total employment. Shortages remain very high in construction, mechanics and metalworking, maintenance and healthcare. View the source

APEC

Company sourcing practices, published on 17 July 2026, 2024 data

54% of companies that hired at least one manager ran into difficulties. Posting a job offer and activating their network enabled 81% of companies to complete their hires. Multiposting of job offers fell to 71% (-2 points), direct approach to 66% (-3 points) and use of intermediaries to 47% (-3 points). View the source

INSEE

Informations rapides no. 228, published on 16 September 2025

In the second quarter of 2025, hourly wages and hourly labour costs rose by 2.0% year on year: +2.2% in industry, +2.6% in construction, +1.9% in services. It is the only public data that makes it possible to update a vacancy cost calculated in-house. View the source

THE FINDING

None of these four sources publishes a cost per hire or a cost per application. They measure the volume of planned hires, perceived difficulty, shortages by occupation and the price of labour. The price of acquiring a candidate is not a public statistic in France.

Gaps

Why a French average would be of no use to you

Suppose the figure existed. It would add together labour markets that have nothing to do with one another. French publications do not give the cost, but they measure the gaps, and the gaps are enough to disqualify the average.

The gap between sectors

France Travail counts 322,000 planned hires in health and social care for 2026, and 193,000 in agriculture. Two volumes, two talent pool densities, two levels of competition for candidates' attention. The same budget does not produce the same number of applications in each.

The gap between occupations

DARES classifies six occupations in ten as facing strong or very strong shortages, a little over half of total employment. The other half is not. An average that mixes a role filled in three weeks with a role open for six months tells you nothing about either.

The gap from one year to the next

43.8% of companies expect difficult hires in 2026, versus 50.1% a year earlier. INSEE, for its part, measured a 2.0% year-on-year rise in hourly labour costs in the second quarter of 2025. A cost benchmark that is a year old describes a market that has already moved.

What this implies

A figure borrowed from an average, a foreign study or an old article does not come close to yours: it describes another labour market. The only value you can base a decision on is the one your own campaign produces, for your role and your local labour market.

Method

Calculate your cost per application, then work back to cost per hire

None of this requires a study. The figures you need are in your distribution tool and in your ATS.

Calculate by family of roles

A forklift driver in the Marne and a management controller in Lyon have neither the same talent pool nor the same price. A company-wide average mixes the two and does not help you decide anything.

Cost per application = budget spent ÷ applications received

The budget actually consumed, not the budget planned. And over a period that goes beyond the learning phase of the distribution algorithms: judged after fifteen days, a campaign always looks more expensive than it is.

Work back to the hire with your own rates

Cost per hire = cost per application ÷ (qualified application rate × interview rate × hire rate). These three rates are in your ATS. If they are not, that is the first thing to fix, before any budget discussion.

Add the time

Applications received multiplied by the unit screening time, multiplied by the fully loaded hourly cost. On many roles, this figure exceeds the media budget. That is what makes a cheap but off-target application more expensive than a costly, qualified one.

Redo the calculation for every campaign

A cost per application is not a company constant. It is the price of a local market at a given moment, and it moves with the season, competition in the local labour market and the wording of the job offer.

Keep the tool line separate from the media line

Two budget lines, two logics. At Seeqle, the platform is available from €500 per month. See pricing

The figure you cannot borrow

There is no French national cost per application. Neither INSEE, nor DARES, nor France Travail, nor APEC publishes one. Yours can be measured in one campaign, and any value found elsewhere will be either foreign, outdated or without a method.

Before setting a budget, it is still useful to know what you already attract and where you lose it. That is what the Talent Attraction Diagnostic does, in two minutes and with no data to prepare.

Sources

All sources cited, dated and linked

France Travail

2026 labour-needs survey (BMO), national press release of 21 April 2026. francetravail.org

DARES

Labour market shortages in 2024, Dares Résultats no. 5, 4 February 2026. dares.travail-emploi.gouv.fr

APEC

Recruiters less aggressive on the managerial job market, published on 17 July 2026, 2024 data from the managerial recruitment practices study. apec.fr