Benchmark candidate application formsStaffing (France)
Form score
Technical UX audit of the candidate application forms of ten of the largest staffing brands in France, measured with the Exatom form-audit engine on 4 November 2026. Scores run 0–10, higher is better. In a multi-step flow the score is taken on the page where the candidate first fills in personal details; where an account is required, that is the sign-up page.
| Brand | Overall ↓ | Accessibility | Performance | Style | Usability | Web standards | Validation | Privacy |
|---|
The benchmark in numbers
November 20266.4Average score out of 10
10Brands scored
8Median fields asked
Category weights
Accessibility×6
Usability×6
Web standards×6
Performance×5
Style×4
Validation×4
Privacy×2
AI readiness×2
The ranking is tight: six brands sit between 5.9 and 7.0. Triangle Intérim has a short, clean page with linked labels, the right input types and a perfect style score; Randstad is the most consistent, with no scored category below 7.3 apart from AI readiness. At the bottom, Manpower combines the longest form with the weakest usability, and Hays, despite one of the shortest forms, loses its whole web-standards score on labels and input types. AI readiness is 0 for all ten brands and is left out of the table. Triangle Intérim's page speed could not be measured, so performance is left out of its overall score. This engine version added the Privacy and AI readiness categories: the scores cannot be compared with the German car insurance benchmark.
Universal failures – what most brands get wrong
10 of 10 are not ready for AI agentsNo form exposes its fields through WebMCP, so an AI assistant cannot fill in or submit the application
8 of 10 have no inline validationErrors only appear after the candidate presses the button
7 of 10 have field borders with low contrastInputs are hard to locate, above all for low-vision users
6 of 10 create an account when the candidate appliesPassword, confirmation or log-in screens come before the first job-related question
5 of 7 set labels too close to the fieldsLabel and input blur together, so the form is harder to scan
4 of 10 give no mobile keyboard for phone or postcodeCandidates on a phone get a full keyboard to type digits
Accessibility
Measures how usable the form is for all users: label associations, contrast of labels, input text and field borders, font sizes and the ability to submit the form. Weight in the overall score: ×6.
Market average: 7.0 / 10. Proman (9.1), Randstad (8.7) and Hays (8.4) lead. Gi Group is lowest at 2.8: small input text, a pale label colour, borderless fields and labels set tight against the inputs. Start People, Hays and Manpower use the placeholder as the only label, so the label disappears as soon as the candidate starts typing.
Benchmark ranking – Accessibility
Score ranking – Accessibility
6.5–10 Good
5.0–6.4 Fair
0–4.9 Needs work
Performance
Covers page speed: how fast the page with the form loads and becomes usable. Weight in the overall score: ×5.
Market average: 7.1 / 10 over nine measured pages. Six brands score 8.1. Gi Group and Manpower follow at 6.4, and Hays is slowest at 2.5. The page speed of Triangle Intérim could not be measured.
Benchmark ranking – Performance
Score ranking – Performance
6.5–10 Good
5.0–6.4 Fair
0–4.9 Needs work
Style
Evaluates visual design: the position of the form on the page, label placement, field alignment, spacing and font consistency. Weight in the overall score: ×4.
Market average: 7.3 / 10. Triangle Intérim scores 10, Proman 9.1 and Start People 8.8. Adecco is lowest at 3.6: labels sit beside rather than above the inputs, and the form is pushed down and stretched by a long page. Manpower (4.7) mixes fonts inside the form and spreads 14 fields over a large area.
Benchmark ranking – Style
Score ranking – Style
6.5–10 Good
5.0–6.4 Fair
0–4.9 Needs work
Usability
Covers interaction quality: autocomplete, the prominence of the call-to-action, keyboard order, mobile keyboards and the number of fields. Weight in the overall score: ×6.
Market average: 7.5 / 10. Triangle Intérim (9.5) and Adecco (9.4) lead. Manpower is lowest at 3.3: autocomplete is switched off, the submit button barely stands out, passwords cannot be shown and pasting them is restricted, and the field count is the highest in the benchmark.
Benchmark ranking – Usability
Score ranking – Usability
6.5–10 Good
5.0–6.4 Fair
0–4.9 Needs work
Web standards
Checks whether fields are real form fields: programmatically linked labels, HTML5 input types and semantic structure. Directly affects autofill and assistive technology. Weight in the overall score: ×6.
Market average: 7.1 / 10. Triangle Intérim and Actual score 10, Randstad 9.2. Hays scores 0: its e-mail field is a plain text input and none of its fields has a programmatically linked label. Proman (3.4) also uses a text input for e-mail, and Start People (5.3) has no linked labels at all.
Benchmark ranking – Web standards
Score ranking – Web standards
6.5–10 Good
5.0–6.4 Fair
0–4.9 Needs work
Design analysis
How each application journey is built, recorded by walking it by hand from a job ad on the French site to the first page that asks for personal details, with a fictitious candidate. Eight of the ten brands collect the application on a single form page, but only Hays and Actual open the form straight from the job ad. Five brands require an account and Synergie creates one with the application.
Form structure and number of fields – all brands
| Brand | Structure | Account | Screens | Application | Personal | Consent | Total |
|---|
The table counts the inputs a candidate meets from the job ad up to and including the scored page; screens run from the apply button to that page, and an e-mail or password typed twice counts as two fields. Totals run from 6 to 14, with a median of 8. The shortest forms, Actual and Triangle Intérim (6) and Hays (7), ask for name, e-mail and sometimes phone, plus the CV. The longest, Manpower (14), Adecco (13) and Synergie (12), combine account creation with the application. The four guest forms average 6.5, the six account flows 6.4: the cost of the account shows up in screens and fields rather than in the score.
Which brands are in the benchmark
The ten largest staffing groups in Europe, on their French candidate site
| Rank | Group (HQ) | French brand | EU share | Outcome |
|---|---|---|---|---|
| 1 | The Adecco Group (CH) | Adecco | 7.1% | Scored |
| 2 | Randstad (NL) | Randstad | 6.6% | Scored |
| 3 | ManpowerGroup (US) | Manpower | 4.2% | Scored |
| 4 | Hays (UK) | Hays | 2.3% | Scored |
| 5 | Proman (FR) | Proman | 1.9% | Scored |
| 6 | Gi Group (IT) | Gi Group | 1.6% | Scored |
| 7 | House of HR (BE) | StaffMe | 1.5% | Not researchable: app-only |
| 8 | Synergie (FR) | Synergie | 1.4% | Scored |
| 9 | Groupe Crit (FR) | Crit | 1.2% | Not researchable: account before personal details |
| 10 | Recruit (JP) | Start People | 1.2% | Scored |
| – | Groupe Actual (FR) | Actual | <1% | Replacement 1, scored |
| – | Partnaire (FR) | Partnaire | <1% | Skipped: terms + bot check before any field |
| – | Robert Half (US) | Robert Half | <1% | Skipped: shadow DOM, not scoreable |
| – | Triangle Intérim (FR) | Triangle Intérim | <1% | Replacement 2, scored |
The starting list is the ten largest staffing groups in Europe by market share (SIA). Where a brand could not be researched, the next largest staffing group with a French candidate site took its place: Actual and Triangle Intérim. SIA publishes ranks beyond ten only to members; the groups outside the top ten each hold less than 1% of the European market.
Brands that could not be scored
Two of the top 10 and two replacements are not in the ranking
| Brand | EU share | Reason |
|---|---|---|
| House of HR (StaffMe) | 1.5% | House of HR's French staffing activity runs through StaffMe. Candidates can only sign up and apply in the StaffMe mobile app; the website links to the App Store and Google Play and has no web application form. |
| Groupe Crit | 1.2% | No guest application. "Je postule" leads to an e-mail check and then a password-only account creation page (password, confirmation, terms, opt-in). Personal details are only asked after the account is created and verified, which was not done. For reference, the sign-up step scores 7.8. |
| Partnaire | n/a | Account required. Before any personal field appears the visitor must accept the terms and pass an automated security check on a separate screen, which was not bypassed. |
| Robert Half | n/a | CV-first flow: step 1 uploads a CV (or LinkedIn), step 2 shows the parsed contact details for confirmation (first and last name, phone, e-mail, postcode, 2 checkboxes). The fields are web components in shadow DOM, so the audit engine only found the cookie-banner inputs and returned no valid score. |