BlogCraft
Employee referral bonus: how much, and why it is not enough
A referral bonus averages €774 in France and is taxed as salary. What makes employees refer has less to do with the amount than with being involved.
Contents
- How much should a referral bonus be?
- Is a referral bonus subject to social security contributions?
- Why does raising the bonus not bring more referrals?
- What actually makes a consultant refer someone?
- Are referred hires better?
- Do referrals threaten diversity?
- What can be delegated in a referral programme?
An employee referral bonus averages around €774 in France, according to an IntoTheMinds survey of 500 HR managers, and it almost never keeps a programme alive on its own. IT services firms hiring for scarce profiles often pay more, and find that referrals do not rise at the same pace as the cheque. The reason is documented: what makes a consultant recommend someone has far more to do with how they are treated during the hiring process than with the amount promised at the end.
The amount is still a decision to make, with its own payroll rules and payment schedule. But it can be settled in ten minutes, and it is everything else that decides whether the referral policy produces hires or becomes a document nobody opens again after the offsite where it was presented.
How much should a referral bonus be?
The average observed in France is €774, and most programmes pay it in two or three instalments rather than all at once. The same IntoTheMinds survey describes a typical combination of about €1,000 per referred hire: one part at hiring, another at the end of the probationary period, and an extra amount when the profile is scarce. The split is not an accounting detail; it tells the referrer that what is being rewarded is a hire that lasts, not a CV forwarded.
That figure has limits that need stating before it is used. It is a self-reported survey across all sectors, and an IT services firm looking for a cloud architect is not in the same market as a retailer hiring shop staff. The right benchmark is internal: what a hire through another route costs you today, whether a job ad, an agency or paid sourcing. A bonus that represents a reasonable fraction of that is defensible, whatever its absolute amount.
The same survey gives a figure that puts the question in perspective: 31% of the companies surveyed have a referral programme, and 11% had one and dropped it. Roughly one programme in four that was launched has therefore been stopped. The survey does not say why, but the experiment described below suggests that the size of the bonus has little to do with it.
Is a referral bonus subject to social security contributions?
Yes: paid to an employee, a referral bonus is part of their pay, subject to social security contributions and income tax like any other bonus. Service-Public lists “indemnités, primes, gratifications et tous autres avantages en argent” among the sums counted when contributions are calculated, and the referral bonus has no exemption scheme of its own.
Two practical consequences follow, and they regularly surprise referrers and employers alike. The first is that the €1,000 announced gross arrives noticeably smaller in the employee’s account, while the company pays its own employer contributions on top: the real cost is therefore higher than the amount written in the policy. It is better to state it as a gross figure, clearly, than to let someone discover the difference on their payslip.
The second is legal. A bonus set out in a unilateral commitment by the employer, which is what a referral policy circulated to staff amounts to, becomes mandatory as soon as its conditions are met. Everything that conditions payment must therefore be written down before the first case: how long the referred hire has to stay, what happens to the bonus if the referrer leaves in the meantime, and who is excluded from the scheme, typically recruiters and managers hiring into their own team.
Why does raising the bonus not bring more referrals?
Because a larger bonus brings in more referrals of lower quality, and what actually makes a programme work lies elsewhere. That is the finding of a rarely cited and yet decisive experiment: Friebel, Heinz, Hoffman and Zubanov introduced referral programmes at random across the stores of a retail chain, with different amounts, and measured what happened.
Three results matter for an IT services firm. Larger bonuses increase the number of referrals but lower their quality, which means that past a certain point you are paying for recommendations made as a favour. The increase in referrals itself is modest. And yet the overall effect of the programme is substantial: attrition falls by 15% in the stores concerned, including among employees who were not referred.
The mechanism the authors consider best supported is that employees value being involved in hiring. It is not the bonus that keeps people, it is being asked for their view on who will come and work next to them. In an IT services firm, where a consultant often spends more time at the client’s site than at the office, that involvement is all the rarer, and all the more valuable when it happens.
What actually makes a consultant refer someone?
Knowing what became of their recommendation, and feeling that it was taken seriously. This is where most programmes fail, and it costs nothing.
The scenario is familiar. A consultant forwards the CV of a former colleague, the recruiter thanks them, and then nothing for three weeks. The referrer hears from their friend that there was an interview, then that they were not selected, without knowing why. They will not do it again, and they will say so to others. The silence employers are blamed for towards candidates, which affected 53% of them in 2026, comes from an overloaded queue rather than from negligence, and it hits the referrer in exactly the same way.
What keeps a programme alive therefore comes down to a few regular gestures. You acknowledge the referral the same day, tell the referrer when their contact has an interview, explain a rejection in one honest sentence, and announce the hire before they hear it at the coffee machine. None of this needs dedicated software: it needs someone to remember, at the right moment, for every referral in progress.
The other lever is to make referrals specific. “If you know anyone, do not hesitate” produces nothing, because nobody mentally scrolls through their contacts on such a vague invitation. “We are looking for a Java developer who has worked in a banking environment, available in January” brings a name to mind. Open roles should therefore be shared with consultants the way they are shared with recruiters, with the same level of detail.
Are referred hires better?
They stay longer, and that is where most of their value lies. The study by Burks, Cowgill, Hoffman and Housman, based on nine large firms in call centres, trucking and high tech, shows that referred hires are 10 to 30% less likely to quit than other hires, even though they resemble them on most measures of skill.
For an IT services firm, that figure carries weight. With a median turnover of 18% in the sector, a figure we covered when looking at how long it takes to place a consultant, and around 30% of new joiners leaving within the first six months, a hire who stays is often more profitable than a brilliant one who leaves before the first renewal. That is also the window in which the consultant nobody calls drifts away, and the referrer, who knows the referred hire, is often the first to see a departure coming.
The second result is more uncomfortable and more useful. The authors find that the advantage of referrals comes mainly from those made by the most productive employees, and that people recommend others who resemble them, in behaviour as much as in profile. They draw a direct recommendation from it: ask your best people first.
Do referrals threaten diversity?
Yes, as soon as they become the main channel, because a personal network reproduces the make-up of whoever owns it. Payscale surveyed 53,000 American workers in 2017: out of a hundred referred people, 44 are white men, and women of colour are 35% less likely to have been referred than comparable white men.
These are American figures and do not transfer as they stand, but the mechanism transfers perfectly. An IT services firm whose technical team is overwhelmingly male and drawn from three schools will receive referrals that look like it, and each successful hire will strengthen the pattern. The paradox is that the IntoTheMinds survey shows half of HR managers seeing referrals as an inclusion tool: intention and outcome can diverge for a long time without anyone measuring it.
Measuring it is simple, though. Every quarter, compare the make-up of referred hires with that of hires from other channels, and look at the share of hires that come through referrals. As long as they remain one source among several, alongside the talent pool you search before sourcing and job ads, the risk stays contained; once they pass half of all hires, it becomes structural.
What can be delegated in a referral programme?
The follow-up, entirely, and the decision, never. What makes most programmes fail is a string of small omissions: the acknowledgement never sent, the referrer never told, the end-of-probation payment forgotten because nobody noted the date.
All of that stays inside the company and can be handed over without difficulty. That is how Balt runs a referral programme: he logs each recommendation, updates the referrer at every stage and reminds payroll when the balance falls due, while the message to the referred candidate, which leaves the company, is still written or reviewed by a recruiter before it goes.
The question of the amount therefore deserves ten minutes and one clear line in the policy. The one that deserves time is this: how many of last year’s referrers know today what became of their recommendation?
Frequently asked questions
What is the average employee referral bonus?
The IntoTheMinds survey of 500 HR managers in France gives an average of €774. IT services firms hiring for scarce profiles often go higher, but the amount matters less than people think: past a certain point it brings in more referrals of lower quality.
Is a referral bonus subject to social security contributions?
Yes. Paid to an employee, it is a bonus under French social security law and enters the base for both employee and employer contributions, as well as taxable income. It appears on the payslip for the month in which it is paid.
When should the referral bonus be paid?
The most common practice splits it: one part at hiring, the balance at the end of the probationary period, sometimes an extra amount for a scarce profile. The conditions must be written into the policy before the first payment, because a unilateral commitment by the employer makes the bonus mandatory as soon as its conditions are met.
Do employee referrals hurt hiring diversity?
They can, because everyone recommends people from their own network, who tend to resemble them. The risk is measured by comparing the make-up of referred hires with that of other channels, and it is contained by keeping referrals as one source among several rather than the main channel.
Sources
- IntoTheMinds, Recrutement par cooptation : état du marché français, survey of 500 HR managers (2023)intotheminds.com
- Service-Public.fr, Salaire, primes et avantages (fact sheet F2301)service-public.gouv.fr
- Burks, Cowgill, Hoffman and Housman, Understanding the Value of Hiring through Referrals, IZA DP 7382 (2013), published in the Quarterly Journal of Economics (2015)docs.iza.org
- Friebel, Heinz, Hoffman and Zubanov, What Do Employee Referral Programs Do?, NBER Working Paper 25920nber.org
- Payscale, Why Relying on Referrals Could Hurt Your Efforts to Hire a Diverse Workforce (2017)payscale.com

