Undercharging is not a feeling. It is three numbers.

Probably — but "undercharging" is a comparison, and you are missing two of the three numbers it needs. Your headline rate is what appears on the proposal. Your effective hourly rate is everything you collected divided by every hour worked, billable or not. Your opportunity cost is the employment you gave up, loaded with the contributions an employer used to pay for you. Almost nobody has computed all three.

One number on its own says nothing. Three of them, computed together, produce an answer.

Why the question has no answer on its own

"Am I undercharging?" is a comparative claim with the comparison missing. Too low against what? Against the market, against your own cost of delivery, against the value to the client, against the salary you would earn doing the same work for somebody else? Those four benchmarks routinely produce four different answers, and two of them are rarely computed at all.

So the useful version of the question is: what is the highest of my floors, and where does my headline rate sit against it? That is answerable.

Number one: the headline rate, and what the market reports

The best available data on this market is the Freelancer-Kompass 2026 — freelancermap's survey of 5,412 freelancers across the DACH region, fielded between 17 November 2025 and 8 February 2026. Respondents average 45 years old with 18 years of experience, and 84% are men. It is weighted towards IT and consulting. Those are real limitations and they matter more than the headline numbers.

Average freelance hourly rate by sector, DACH. freelancermap, Freelancer-Kompass 2026, n=5,412.

Sector	Average hourly rate
Consulting and management	€121
Finance, accounting and legal	€114
SAP and ERP	€113
Data and analytics	€108
HR and coaching	€106
IT infrastructure	€100
Marketing and communications	€94
Engineering	€91
Software and web development	€91
Design, media and creative	€89
Research and analysis	€89

The overall average is €103 for 2026, against €104 for 2025 — the first decline since the study began. Now hold that next to the income figure from the same survey: average monthly project income fell from €8,432 to €6,653. That is €1,779, a fall of 21.1%.

The average rate fell by 1%. The average monthly project income fell by 21%. Whatever went wrong in that year, it was not the rate card.

There is a second reconciliation worth doing, because it looks like a contradiction and is not. €6,653 a month annualises to €79,836. Yet 32% of the same respondents report gross profit under €25,000 a year, and only 13% report over €125,000. The two figures sit together because monthly project income is earned in months when there is a project — and 24% of respondents had fewer than 50 project days in the past year. Fifty days is ten working weeks. A quarter of a professional population billed fewer than ten weeks of project time in twelve months.

Average monthly operating expenses in that sample: €783.

Number two: the effective hourly rate

This is what you earn per hour of your life, and it is the number that determines how you actually live. Total collected over twelve months, divided by every hour that went into the business — delivery, proposals, admin, travel, invoicing, the unbilled call, the drive.

The clearest published illustration of the gap comes from coaching. The ICF Global Coaching Study 2023 collected 14,591 valid responses from 157 countries. For Western Europe it reports an average hourly fee of $277, an average of 11.9 hours coached per week, and average annual revenue of $52,400.

Multiply the first two, across an assumed 48-week year:

$277 × 11.9 hours × 48 weeks = $158,222.

Reported average annual revenue: $52,400. That is 33.1% of the implied figure — a capture rate of about a third.

Multiplying three population averages is an illustration and not a measurement, and I would rather say so plainly: the average fee earner and the average hours earner are not the same person, and 48 weeks is my assumption rather than the study's. The gap is too large to be an artefact of that, though. Whatever accounts for it — discounting, unpaid or sponsored sessions, part-time practice, unbilled preparation, gaps between clients — none of the candidates is the rate card.

Which means every product sold into this market on the promise of raising your rate is addressing the smaller of the two problems. The larger one is the capture rate, and legal practice measures at 31% by an entirely different method, which is close enough to a third to be worth noticing.

The hours that go into the denominator are not modest, either. Eurostat's 2024 data, published on 21 August 2025, found 6.6% of employed people aged 20–64 in the EU worked 49 or more hours a week. Split by status, it is 27.5% of the self-employed against 3.4% of employees — a factor of 8.1.

Number three: what you gave up

The third floor is the one nobody computes, because it is a counterfactual and counterfactuals have no accounting entry.

The Bundesamt für Statistik publishes median gross annual earned income from the Schweizerische Arbeitskräfteerhebung, a survey of more than 100,000 respondents. The figures for reference year 2025 were published on 25 June 2026.

Median gross annual earned income, Switzerland, reference year 2025. Bundesamt für Statistik, published 25 June 2026.

Group	Median gross annual earned income
Employees, full-time (90%+ workload), men	CHF 92,000
Employees, full-time (90%+ workload), women	CHF 80,900
Employees, full-time, overall	CHF 87,000
Self-employed, men	CHF 85,700
Self-employed, women	CHF 67,500

CHF 85,700 is the median gross annual earned income of self-employed men in Switzerland — below the CHF 87,000 median for full-time employees, and below the CHF 92,000 median for full-time employed men.

The people carrying all of the risk are, at the median, earning less than the people carrying none of it.

That comparison needs care, because the two lines are not defined identically. The employee medians are restricted to a workload of 90% and more. The self-employed figures carry no such restriction, so some of the gap will be people who are self-employed part-time by choice. That is a genuine caveat and it is worth stating rather than hiding. What it cannot explain away is the direction, or the fact that the self-employed figure includes people working the 49-hour weeks in the Eurostat data above.

An opportunity-cost floor, with every assumption labelled

You can build a floor from the BFS figure. It is an estimate, not a measurement, and each step is marked.

Step 1, data. Median full-time employee earned income: CHF 87,000.

Step 2, assumption. Add roughly 15–20% for the employer-side social and pension contributions a self-employed person now funds themselves. That is an assumption, not a figure from the BFS release. It brings the total to CHF 100,050 at 15% and CHF 104,400 at 20%. Call it CHF 100,000.

Step 3, assumption. Divide by 1,280 billable hours in a full-time year. This is also an assumption — there is no rigorous published Swiss freelance-hours benchmark, and I will come back to that. CHF 100,000 ÷ 1,280 = CHF 78.13.

So roughly CHF 78 an hour, before a single franc of operating cost, simply to match the median employee.

Two things push it materially higher. Operating costs: the DACH average of €783 a month is €9,396 a year, which over 1,280 hours is €7.34 an hour — a different currency and a different country, so treat it as an indication of magnitude rather than a Swiss figure. And downtime: 1,280 billable hours is optimistic against everything measured in the utilisation literature. At 1,000 billable hours the floor becomes CHF 100 an hour. At 800, CHF 125.

None of that is a recommendation about what to charge. It is a floor — the level below which the arrangement is worse than the job you left, on the median, before any premium for carrying risk an employer used to carry for you. It is the counterfactual, and the counterfactual is what the word "undercharging" has been quietly assuming all along.

There is no honest Swiss freelance rate benchmark

I want to be plain about this, because it is unusual to admit and it changes how you should read every rate figure you have ever been quoted.

No rigorous published benchmark study of Swiss freelance or independent consultant rates exists. The Swiss franc figures in circulation — the CHF 150, the CHF 250, the "day rates in Zurich start at" — are practitioner estimates, agency price lists and forum consensus. They have no sample, no methodology and no publisher standing behind them.

The nearest thing to real data covering this market is the Freelancer-Kompass, which is DACH-wide, denominated in euros, 84% male and weighted towards IT and consulting. The BFS figures are Swiss and methodologically solid, but they are annual income rather than rates, and the self-employed line does not separate full-time from part-time. The ICF study is one profession and reports averages across a wide distribution.

That is the whole state of the evidence. Anyone quoting you a precise Swiss hourly benchmark is quoting a number that does not exist, and their willingness to do it is itself information about how they work.

What this means in practice

The three numbers are useful because of the order they resolve in, not because of any one of them.

If your effective hourly rate is roughly a third of your headline rate, the headline rate is not your binding constraint. Raising it multiplies a small fraction and leaves the fraction intact. The constraint is in the four terms between working and being paid.

If your effective rate is a healthy share of your headline rate — say two-thirds or better — and the headline rate sits below the sector figures and below your opportunity-cost floor, then the rate genuinely is the constraint, and the arithmetic of what a rise can cost you is more permissive than it feels.

And if your effective rate is below the CHF 78 floor, the question stops being about pricing. A practice earning less than the median employed alternative, per hour, while working the hours in the Eurostat data, is not underpriced. It is structurally unprofitable, and no rate change fixes a structure. The distance between how good you are and what you earn almost never closes at the price line alone.

What I usually find is that people can quote their headline rate instantly, guess their annual revenue within a few thousand, and have never once divided the second by their actual hours. The division takes twenty minutes. It is avoided because everyone already suspects what it will say.

What this cannot tell you from the outside

Each of these sources has a boundary and it is worth being exact about them. freelancermap is DACH, euro-denominated, 84% male and skewed to IT and consulting, and its averages conceal a distribution wide enough that 32% earn under €25,000 of gross profit while 13% earn over €125,000. ICF covers one profession, and the multiplication I did on it is illustrative. BFS is rigorous but reports income rather than rates, and its self-employed line is not restricted to full-time work. Eurostat is EU-wide, not Swiss.

None of them can tell you what you could charge, because that is not a property of your profession. It is a property of your position: who else the client could buy this from, what the decision in front of them is worth, whether your price is being compared to a competitor's price or to the cost of the problem, and how much of your revenue depends on two or three relationships that would be expensive to lose.

Those facts exist. They are in your proposals, your win-loss record and your client list, and they can be assembled in a couple of weeks. What is striking is that nobody produces them. Your fiduciary reports revenue collected. Your bank sees the balance. Neither of them computes hours, and neither computes a counterfactual, because a counterfactual has no ledger entry — which is exactly why the most decision-relevant number in an expert practice is the one nobody has ever been paid to calculate.