# 80% careers in Switzerland: advertised ranges and leadership roles

In Panta’s 4 November snapshot, 3,365 listings had an 80%-compatible percentage signal without conflicting fields. Only 170 stated exactly 80%; 3,195 advertised a range containing 80%. We also found 325 compatible listings among 937 career records with leadership-title signals. These are observed advert signals, not national estimates or evidence about promotion prospects.

By Panta. Published 2026-11-04.
Canonical: https://withpanta.com/insights/eighty-percent-careers-switzerland/

## How the adverts include 80%

Among 3'365 listings with an interpretable workload that explicitly supports 80%. Frozen 4 November 2026; title/employment fields, not confirmed individual working arrangements.

| Evidence category | Listings |
| --- | ---: |
| Exact 80% | 170 |
| Range including 80% | 3195 |
| Discrete 80% alternative | 0 |

## A fixed 80% job and an 80–100% range answer different questions

Searching only for the words “part-time” can hide the arrangement you actually need. An exact 80% position, a 60–80% range and an 80–100% range all contain 80, but leave different questions for an applicant to ask.

We examined titles and employment fields for **14,536 fresh-backed listings classified as physical Swiss workplaces**, observed on 4 November 2026 across 710 employers. Of the 3,365 records with an interpretable percentage containing 80, **170 stated exactly 80% and 3,195 gave a continuous range**. Thus 94.9% of these detected signals were ranges. We found no unconflicted discrete 80% alternative in the corpus.

This describes recorded wording. A range is a lead to discuss, rather than a guarantee that an employer will accept a particular percentage, compressed week or distribution of working days. A 60% job also does not automatically offer 80%. Records with explicit numeric disagreement between title and employment field remain unresolved.

## What changes when we focus on careers?

The full inventory includes training opportunities. For career comparisons we removed **3,412 records with training labels** and **44 further records with obvious spontaneous-application or talent-pool titles**. The resulting cohort contains **11,080 records from 698 employers**. Contract permanence was not inferred from the absence of a training label.

In this cohort, 3,271 records had an 80%-compatible signal: 167 exact values and 3,104 ranges. That is **29.5% of all cohort records**, or **67.8% of the 4,822 with interpretable numeric workloads**. Both denominators matter. A further 6,212 records had no detected percentage, and 46 were conflicting or ambiguous. We do not assign those records to 100%.

The family breakdown illustrates where signals appeared in this index. These are ordered, broad title matches; they are not employer industries or a complete occupation taxonomy.

| Title family | 80%-compatible / all records | No percentage detected | Employers |
| --- | ---: | ---: | ---: |
| Healthcare | 928 / 1,990 | 823 | 76 |
| Sales and customer advice | 582 / 1,644 | 838 | 131 |
| Technology and engineering | 252 / 1,257 | 895 | 278 |
| Finance and insurance | 55 / 354 | 252 | 135 |

For example, 252 of 1,257 technology/engineering records contained an 80% signal, but 895 had no percentage in the parsed fields. Among its 362 interpretable records, 252 contained 80%. These are different descriptions of the same data, not competing estimates of actual flexibility.

The downloadable pack includes education/research/social, trades/operations and the 4,669 other/unclassified records. All displayed families meet our gate of at least 100 records and ten employers. Nevertheless, incomplete title rules and sharply different information availability prevent a reliable league table of Swiss careers. Some recognizable roles, including SAP and social-work titles, remain unclassified.

## Eighty percent appears in leadership-labelled adverts

We separately identified explicit leadership-title words, including heads of functions, department/team leaders and deputies. Generic “manager” alone was insufficient: an account or project manager is not necessarily a people manager. Assistant/secretary-to-leadership and hospitality chef-de-rang/departie titles were excluded.

There were **937 leadership-labelled records from 217 employers** in the career cohort. **325 contained 80%**: 12 exact values and 313 ranges. Another 471 had no percentage; 138 stated other interpretable percentages and three remained unresolved.

Reviewed examples included a head of plant engineering, a stroke-unit team leader, an engineering department head and a social-work team lead. Their numeric ranges included 80%. The sample also included jobsharing wording, which is a separate arrangement from one person working 80%.

This supports a useful search conclusion: applicants looking for an 80% workload can examine leadership-labelled opportunities alongside other roles. It does not show how often employers hire at 80%, whether every matched title involves staff management, or what happens to subsequent promotion or pay.

## Employer coverage changes the apparent picture

Coop-Gruppe contributes 1,464 career-cohort records. Removing that group changes the overall 80%-signal fraction from **3,271/11,080 (29.5%) to 2,478/9,616 (25.8%)**. Within the leadership-labelled group, it changes from **325/937 (34.7%) to 188/749 (25.1%)**.

The effect is particularly large in sales/customer advice: 582/1,644 records contain 80% in the full cohort, compared with 66/732 after removing the largest employer group. This is a warning about concentration and source-field availability, not evidence that other employers reject part-time work. Employer IDs can group several brands, and removing one group cannot make the remaining inventory nationally representative.

## Missing fields can conceal the answer

We reviewed 88 deterministically selected records across exact values, ranges, other percentages, conflicts, ambiguity, missing fields, leadership and unclassified titles. This was AI-assisted review with iterative rule corrections, not a held-out accuracy study or independent human double coding.

For the **12 missing-field cases**, we read the full stored descriptions. One explicitly stated 80–100%, another stated 100%, and a third offered a minimum of 40% without an upper endpoint. Nine had no numeric workload statement. Benefit discounts in those descriptions were not counted as workload. Weekly hours were also left unconverted because their full-time basis was unknown.

These cases explain why the main analysis stays narrow: missing title/field percentages are not full-time jobs, and scanning every percentage in a description would manufacture false matches. They also show why a jobseeker should read the original advert rather than rely on one index field.

## How this fits other Swiss research

[FunkyJobs’ September 2026 analysis](https://www.funkyjobs.ch/journal/teilzeit-inserate-schweiz) already examines reduced workloads across job fields using 29,650 direct-employer adverts. Its method counts percentages below 100 and reports readable-workload coverage of 97–100% in the displayed fields. Our question is specifically whether 80 is contained in the recorded value or range, with a much larger unknown group. Different cutoffs, selection rules and classifications mean the raw percentages cannot be treated as provider performance or market-change comparisons.

The [Federal Statistical Office’s part-time series](https://statistik.admin.ch/de/asset/36326371) concerns employed people and defines part-time as below 90%. Our adverts measure neither actual employee schedules nor the official part-time share. The [Advance/HSG Gender Intelligence Report 2025](https://www.advance-hsg-report.ch/en/) uses employee HR data to study workplace equality; advert availability cannot establish its career-outcome findings.

For an applicant, the next questions are concrete: does the employer accept 80% for this role, which days or coverage are expected, and how are responsibilities allocated? The dated [audit data](/insights/data/eighty-percent-careers-2026-11-04.json) and [methodology](/insights/methodology/) let you inspect what the report measured before applying it to your own search.