How many job openings are there in Switzerland?

What our job index contains — as of 7 September 2026

Counted over 22,518 published adverts in OUR index — this is not the number of open positions in Switzerland. The national figures come from SECO and the Federal Statistical Office. This table says what the adverts we see look like.

The two official counts, and what each one actually measures

The larger number comes from the Federal Statistical Office and belongs to a quarterly survey called the employment barometer. It samples around eighteen thousand enterprises covering sixty-five thousand establishments in the secondary sector (industry and construction) and the tertiary sector (services). A vacancy, in its definition, is a newly created, unoccupied or soon-to-be-vacant position that the enterprise intends to fill immediately or in the near future and for which active recruitment steps have already been taken. In the 2nd quarter of 2026 that came to just under ninety-nine thousand positions, 2.7% more than in the same quarter a year earlier.

The smaller number comes from SECO, the State Secretariat for Economic Affairs, and counts something narrower: not what is open, but what was reported. In July 2026 around forty-five thousand vacancies were registered with the RAV, the regional employment centres — 4.4% fewer than in June and 13.4% more than a year earlier. SECO writes the decisive sentence into its own definitions: the registered vacancies cover only part of all vacancies in Switzerland.

Set the two against each other and you get about 46%. Fewer than half of the country's open positions are known to the public employment service, and since roughly two thirds of the registered ones are there because the Stellenmeldepflicht obliges the employer, the voluntarily reported remainder amounts to a little under a sixth of all vacancies. One caveat: one figure is a July monthly stock, the other a quarterly average, and the two survey frames do not coincide. Treat the ratio as an order of magnitude, not a decimal.

There is no third official number. What a job board prints is a count of adverts, and how large it is depends on how widely that board collects and how quickly it removes expired posts. Comparing two home pages is therefore not comparing two labour markets.

If you are reading this from outside Switzerland

Two features of the Swiss count have no equivalent in most other countries, and both change what the number means for someone arriving.

The first is the Stellenmeldepflicht. Occupations with an unemployment rate at or above 5% must be notified to the public employment service before they may be advertised anywhere else, and registered job seekers get a five-working-day head start on them. That rule is why 65.9% of the registered vacancies are registered at all. It also means the RAV channel is systematically richer in exactly the occupations with the most competition — and that a newly arrived candidate who is not registered will not see those posts first.

The second is that the official count is a count of positions at Swiss establishments, regardless of who fills them or where they live. Cross-border commuters, holders of a B or C permit and people relocating are all counted the same way, because the statistic asks the employer, not the employee. So the national figure is not a measure of how many posts are open to you personally. That depends on your permit situation, your language profile and your recognised qualifications — none of which appears anywhere in the ninety-nine thousand.

One more practical point about sources: the monthly SECO report is published in German, French and Italian, and the amstat.ch data portal carries no English version. The quarterly FSO release does exist in English. If you want the current national figure in English, that release is the one to read.

What the official count leaves out, and why it is exactly the adverts

The methodological annex of the employment barometer contains a line almost nobody quotes: results from NOGA division 78, 'Employment activities' — recruitment, placement and staffing agencies — are not included. That is methodologically clean, because otherwise the same post would be counted twice, once at the client company and once at the agency. But it means the official figure excludes precisely the type of advertiser that is most visible on job boards.

How visible can be measured. In our own index of open Swiss adverts, 10.0% come from firms that identify themselves in their own company name as a recruitment, placement or staffing business. That is a lower bound, since only those that say so are caught. The single largest advertiser in our index is one such firm: it alone accounts for 6.1% of all adverts, roughly one in sixteen.

Then there is what is not a position at all in the official sense. 6.4% of adverts carry an apprenticeship, internship or trainee programme in the title — these advertise a qualification, not a post to be filled. A further 0.1% are agricultural, and agriculture is outside the survey frame entirely, which covers only sectors II and III.

Put the three exclusions together and 16.4% of our advert stock falls outside the official definition — roughly one advert in six, before a single position has been counted. That accounts for a good part of the distance between a result list and the federal figure. It does not account for all of it: in the other direction, every post an employer fills internally, through contacts or without advertising is missing from the list while still counting as a vacancy in the survey.

One advert is not one job

Every result count carries a silent assumption: one line, one job. It fails in both directions. Upwards, because a single advert can be hiring several people — and it almost never says so. In our index, 0.01% of adverts name a number of people wanted; they are literally isolated cases, mostly public-sector training posts that write 'eight positions' into the title. Everything else leaves open whether one person is being hired or five.

Downwards, because the same post can appear in the list several times: across several portals, as a republished advert, or as a branch listing from a chain filling the same role in twenty locations. Which patterns produce those repeats, and what gives them away, is covered by our answer on why the same job appears several times online, where the figures sit.

The practical consequence is that a result count is neither an upper nor a lower bound on the number of jobs. It is a statement about adverts. It is useful for comparisons inside the same list — this occupation against that one, this week against last — and useless for the question of how many posts are open in the country.

1.7% sounds small — in your branch it is a different number

The most useful figure in the official statistic is not the total but the rate: vacancies divided by labour demand, which is occupied plus vacant posts. In the 2nd quarter of 2026 it stood at 1.7%. Put differently, about one job in sixty in Switzerland is currently being filled. Keep that in mind whenever a headline shouts about a hundred thousand vacancies: the same fact is large or small depending on what you measure it against.

Between branches the rate spreads widely. At the top are IT and other information services at 2.8%, the manufacture of computers, electronics and watches at 2.5%, information and communication overall at 2.3%, and machinery manufacturing and financial and insurance services at around 2.1%. Construction sits at 2.1%, health and social work at 2.0%, trade at 1.9%, hotels and gastronomy at 1.7%, transport and storage at 1.6%. At the bottom are public administration at 1.4% and education at 0.7% — a factor of almost four between education and IT services.

Composition tells a different story from the rate. Three quarters of all vacancies sit in the service sector and a quarter in industry and construction. The single largest economic division is health and social work, which alone carries about 18% of all vacancies in the country. So translating the national figure into your own situation takes two filters, not one: the branch, and then the occupation inside it.

The number has fallen, but not everywhere

The FSO series runs back to the 2nd quarter of 2003 and puts today's figure in order. The peak is the 2nd quarter of 2022; against it, the 2nd quarter of 2026 is 21.9% lower. Against the pandemic quarter of spring 2020 it is about 62% higher, and against the start of the series in 2003 it is a good two and a half times as large. Anyone telling you the Swiss labour market has been emptied out, or has collapsed, is quoting one of those comparisons and not the others.

More interesting than the total is how differently the branches moved. Taking 2021 as one hundred, the total stands at about 110 in the 2nd quarter of 2026. Underneath it, construction is at 141, transport and storage at 139, health and social work at 137, public administration at 126 and education at 122 — while information and communication is at 83, IT services at 83, hotels and gastronomy at 87 and machinery manufacturing at 90.

That is the real story behind the national figure. The total has moved by about a tenth since 2021; its composition has moved by forty per cent upwards and a sixth downwards. Almost any statement about 'the Swiss job market' is in fact a statement about a branch that was not named.

How many people there are per opening

The other half of the question is usually: and how many people are competing for them? There are official numbers here too, though no direct one. In July 2026 SECO registered around one hundred and thirty-nine thousand unemployed and around two hundred and twenty-seven thousand job seekers — the second group also includes people who have a job or are not immediately placeable. That works out at roughly three unemployed and five job seekers per registered vacancy. Measured against the larger FSO count instead, it is about 1.4 unemployed per open position.

These are crude ratios, not application counts. Registered unemployed people do not apply to every post, and most applications come from people who already have a job. Still, the distance between three and 1.4 is informative in itself: it measures the gap between what the public employment service sees and what exists.

That both things can be true at once is shown by the barometer's third indicator. Weighted by number of employees, 33.9% of enterprises reported difficulty recruiting skilled staff in the 2nd quarter of 2026 — an increase of 0.4 percentage points on the previous quarter. Weighted by number of establishments it is only 9.7%: large employers have the problem, the many small firms report it less often. By branch the range runs from machinery manufacturing at 58.8% and health and social work at 50.1% down to education at 15.2%. A market with 1.4 unemployed per opening and, at the same time, one job in two in health and social work sitting at an employer that reports recruitment difficulty is not a contradiction. What is scarce is not work; it is particular qualifications.

Why even the official number keeps moving

A figure from a sample survey is an estimate, and estimates get corrected. In the 1st quarter of 2026 the FSO renewed the barometer's sample and consequently revised the time series from the 2nd quarter of 2018 to the 1st quarter of 2026. That note appears on every sheet of the current release. So a number quoted from an older article may no longer exist in that form — including for quarters that closed years ago.

Then there is the lag. The barometer is quarterly and the 2nd-quarter 2026 figures were published on 27 August 2026. The monthly SECO report follows a calendar announced in advance: the August 2026 report on 7 September, the September one on 6 October. Both sources are careful, and from a job seeker's point of view both are old.

In practice: take the official figures for the order of magnitude and a live list for today's state, and do not confuse the two.

The number that actually applies to you

The national figure answers no question you actually have while searching. It becomes useful only once you narrow it three times: to the branch, to the occupation, and to the region you can realistically work in. Just under ninety-nine thousand then becomes a few hundred to a few thousand — and a headline becomes a quantity you can work through in a week.

Two things follow. First, if your list is short, that is more often your filters than the market; the rate in your branch tells you whether you are searching in a field at 2.8% or one at 0.7%. Second, because fewer than half of open positions are known to the RAV and roughly one advert in six is not a position in the official sense, one source is never enough — the public employment service for the notifiable occupations, a broad search for volume, and employers' own career pages for the rest.