The threshold at which the employee believes in it
A percentage means nothing until it has been converted into euros for the person concerned.
Take an SME valued at four million euros, with forty employees. Opening 5% of the capital to the whole workforce represents €200,000 to share out, or €5,000 per person on average. Over a five-year horizon, if the value grows by half, the individual gain reaches €2,500. That is a thirteenth month's salary spread over five years.
Nobody changes their behaviour at work for that. The scheme will be received as a complicated bonus, and it will produce exactly the effect of a complicated bonus.
The same effort concentrated on six senior managers gives a different equation: around €33,000 per person on entry, a potential gain of €16,000 in the same scenario. At that level, the person follows the value of their company, asks questions about the margin and behaves differently towards a badly priced quote.
The choice is therefore not only "how much", but "how much, for how many people". Diluting widely and diluting deeply are two distinct projects, and you have to choose.
The legal thresholds and what they change
A few company-law reference points, for an SAS whose articles of association have not laid down specific rules.
| Threshold | What it opens up |
|---|---|
| Under 10% | No significant collective rights. The employee receives dividends and follows the value. |
| Over 33% | Blocking minority on extraordinary decisions. The owner-manager can no longer amend the articles or decide on a merger alone. |
| Over 50% | Control of ordinary decisions. Power changes hands. |
These legal thresholds matter less than people think, because the essentials are settled in the shareholders' agreement. A well-drafted agreement can give employees holding 8% of the capital enhanced information rights, an observer seat on the strategy committee and a veto over three or four structural decisions. Conversely, a badly built agreement can leave 25% of employee shareholders without any real visibility.
The question to ask is therefore not "what percentage gives them power", but "what power am I prepared to share, and how I write it down".
Three calibrations commonly encountered
5 to 10%, widely spread. The aim is recognition and value sharing. Works well when the owner-manager is not considering a business transfer in the short term and wants to build a lasting bond with the whole team. To be combined with a company savings plan (PEE) and regular communication on the value, otherwise the scheme is forgotten.
15 to 25%, concentrated on the management team. The aim is retention and preparation. This is the typical calibration of an owner-manager who is considering selling in five to eight years and wants the team to be in a position to take over. The shareholders' agreement becomes central: cross-promises, exit clauses, valuation mechanism.
Over 50%. This is no longer an employee ownership policy, it is a business transfer. It requires structured financing, an investor who carries part of the capital, and a redesigned governance. It is the scheme we put in place when a team takes over its company.
How much is your SME worth?
An estimate from your last three balance sheets, the starting point of any employee ownership plan.
Value my SME →Opening up in stages rather than all at once
The approach we see working best consists of setting a seven-year target and reaching it in annual steps.
A first tranche of 3% in the first year makes it possible to test the mechanics: valuation, internal communication, the team's reactions, the questions asked. The following tranches are adjusted according to what has been learned. The scheme keeps an allocation capacity for future hires, which matters a great deal if you plan to recruit managers.
Opening up all at once has a serious drawback: it freezes the distribution at a given moment. A manager recruited two years later discovers that they arrived after the distribution, which creates exactly the sense of injustice the scheme was meant to avoid.
The most costly calibration mistakes
Giving too little to too many people. It is the most frequent one. The owner-manager wants to be fair, spreads widely, and nobody perceives the effect. The administrative cost remains, the benefit disappears.
Opening up without planning the exit. An employee shareholder who does not know how they will get their money back ends up regarding their shares as paper. Worse, an employee who leaves and remains a shareholder against their will becomes a passive and sometimes hostile partner.
Aligning the distribution on the organisation chart alone. Hierarchical weight is not always the best criterion. In an industrial company, the methods manager or the workshop foreman often weighs more on the real value than certain support roles placed higher in the organisation chart.
Not setting aside a reserve. Using up the whole envelope in the first allocation deprives the company of a recruitment argument for the following years. A reserve of around a third of the target envelope is a useful precaution.
Key takeaways
- Below an amount the employee genuinely perceives, the scheme is received as a bonus.
- The thresholds that matter legally are those of the shareholders' agreement, not those of the commercial code.
- Better to open up 10% with an organised exit than 30% without a liquidity mechanism.
Sources and references
The figures in this article come from the deals we study and the cases we handle. They are field observations, not published statistics. Legal references are cited with their article number.
- French Commercial Code, articles L.227-1 et seq. Rules governing the simplified joint-stock company (SAS), and the freedom its articles of association allow on majorities and exit clauses.
- French Commercial Code, articles L.225-197-1 et seq. Rules governing free share awards.
- French Civil Code, article 1843-4. Valuation of shares by an expert in the event of a dispute between the parties.
This article is for general information purposes. It constitutes neither legal advice, nor tax advice, nor an investment recommendation. Sources last checked: September 2026.
