Saudi Arabia's Skill Verification Program: An Employer's Guide to What the Certificate Does — and Doesn't — Tell You

The short answer

The Skill Verification Program (SVP) — also called Professional Verification — is a Saudi government programme run by the Ministry of Human Resources and Social Development (HRSD) with the Ministry of Foreign Affairs and the Technical and Vocational Training Corporation. It requires expatriate workers in covered professions to pass a theoretical and practical examination in their trade — at an approved centre in their home country before arrival, or at a certified centre inside the Kingdom — before a work visa is issued or renewed. It launched in 2021 with core trades such as plumbing, electrical work, welding, automobile mechanics and carpentry, and by early 2025 HRSD had expanded the service to 160 labour-exporting countries and 1,007 professions, including engineering and health fields. Employers interact with it through the official portal at svp.qiwa.sa.

That is the government's half of the capability question: a pass/fail gate on trade competence at the border. This guide is about the employer's half — the questions the SVP certificate was never designed to answer. Which of forty certified electricians should you actually hire? Can the supervisor you are promoting communicate in working English? And is your Saudisation development programme building capability you can prove, or just attendance? To be explicit up front: AssessAll has no affiliation with HRSD, Takamol, TVTC or Qiwa, does not administer SVP examinations, and an AssessAll assessment is not an SVP certificate — only approved centres issue those.

What the SVP certificate proves — and what it deliberately does not

The SVP answers one question well: does this worker meet the minimum verified standard for this profession? It is a threshold, by design. A pass tells you the welder can weld to the examined standard; it does not tell you whether they are the best of the forty welders who applied to you, how they respond when a job goes wrong, whether they can read a method statement or brief a mixed-language crew, or how they will develop over the next two years. Two candidates with identical certificates can be very different hires.

It is also, by design, a programme about expatriate entry into covered professions. It does not measure your existing workforce's current capability, it does not measure Saudi nationals in your development programmes, and it re-examines nobody as roles, technology and standards change. None of this is a criticism — a national entry-gate examination cannot and should not try to be an employer's selection and development instrument. It simply means the certificate is where an employer's evidence starts, not where it ends.

The selection gap: choosing among candidates who all hold the same certificate

As verification coverage expands, a new problem appears: every serious applicant for a covered role arrives with the same pass. The certificate stops differentiating precisely because it works. What still differentiates is measurement the gate does not attempt: role-scenario judgement (what does this technician do first when a client reports a fault the manual does not cover?), working English where the role demands it — written comprehension and AI-scored spoken responses from the candidate's own recorded voice — and reliability under pressure, measured through timed situational-judgement scenarios covering ownership, escalation and safety-adjacent decisions.

Practically, that is one assessment sitting per candidate, composed for the role from a plain-language brief and delivered through a share link or QR code: candidates create no account, sit it on any phone or laptop, and every attempt is proctored — an identity-baseline photo at the start, face-presence monitoring throughout, tab switches and screen-capture attempts logged, and a High/Medium/Low integrity rating with the evidence attached rather than summarised away. Results arrive scored and ranked, so a hiring manager screening a volume intake reads a ranked list with evidence, not a stack of identical certificates.

The other half: Saudisation programmes need measured capability, not certificates of attendance

The SVP covers expatriate entry; nationalisation targets cover headcount. Between them sits the question that decides whether a Saudisation or Emiratisation programme actually works: what can the national cohort do now, and what changed because of what you spent? Completion certificates cannot answer it — a cohort can complete a programme having learned little, or be put through training it never needed because nobody measured its starting point.

The employer-side fix is the baseline–develop–prove loop: measure the cohort's actual capability before training is booked, target development at the gaps the baseline exposed, then re-measure the same competencies afterwards so the gain is a delta in a board pack rather than an attendance record. We cover that loop end to end — including why it is the evidence that separates a substantive nationalisation programme from a paper one — in the guide to measuring national-workforce capability for Saudisation and Emiratisation, and the tooling behind it (TNA Studio baselines with individual and group reports, Learning Journeys for before/during/after measurement) on the Gulf workforce development page.

Who you would buy that evidence from (added 1 September 2026)

A certificate is issued by a programme; the capability evidence beside it is bought from a vendor, and the Gulf shortlist is shorter than the market makes it look. There are four shapes on offer: global enterprise assessment vendors with large validated libraries and published norm data; regional MENA platforms built bilingual from the interface down; subscription skills-test marketplaces; and pay-per-use measurement platforms. Nothing about holding an SVP certificate changes which of the four fits you — volume, language and what you need to prove do.

The single most useful thing to establish early is what each one will actually cost per person at your volume, which is harder than it should be. Checked at each provider's own published pages on 1 September 2026, SHL's product catalogue and Mercer Mettl's pricing page both route a buyer to a quote or a demo without stating a figure, and Evalufy names three tiers without publishing prices. TestGorilla publishes plan prices. AssessAll publishes a per-credit price of US$0.50, which is where the per-candidate figures in the next section come from.

Two questions decide most Gulf shortlists and are usually asked too late. First, which parts of the product are available in Arabic — the candidate interface, the item content, the AI scoring and the report can each be answered differently by the same vendor, and only the middle two change what a candidate is really being asked. Second, where attempt data and proctoring imagery are stored and on what basis they leave that location. AssessAll is English-first and publishes no in-region hosting option, so if either is a hard requirement it should not be on your list. The four options are set side by side, with the case for each alternative, in Assessment platforms compared for the Gulf.

What employer-side capability evidence costs

Assessment on AssessAll is metered per person in US dollars: 1 credit = US$0.50, a typical screening assessment set runs 15–26 credits (US$7.50–13) per candidate, and fuller diagnostics run higher — proctoring and voice-scored sections add 5 credits each. Credit packs discount volume by up to 13%, there are no seat licences, platform fees or annual contracts, and new organisations receive 250 free credits — enough to screen a real shortlist or baseline a pilot cohort before paying anything. For a volume intake, ranking an entire certified applicant pool typically costs less than one interview round's worth of management time.

One scope note worth repeating because it protects you as much as us: capability assessment complements the SVP — it does not replace it, substitute for any government examination, licence or visa requirement, or certify compliance with any nationalisation quota. Treat any vendor claim to the contrary as a red flag.

If what you need is the specification of an entry-level screen rather than a programme design, the published frontline customer-service blueprint for the UAE, Saudi Arabia and Qatar sets out five live modules row by row, with item counts, time limits, competencies, weights and the cost per candidate in US dollars.

One screen, two languages, two cohorts: the check almost nobody runs (added 15 September 2026)

A Saudi employer running capability assessment at any scale ends up doing something the measurement literature treats as a specialised problem and most hiring programmes treat as an administrative detail: putting the same instrument in front of populations that differ by language. An Arabic version and an English version of one screen are two instruments until someone demonstrates otherwise, and a nationalisation programme compounds it — the cohorts being compared are Saudi nationals and expatriate staff, who differ in schooling language, in prior test exposure and often in the register of workplace Arabic or English the items are written in.

The ITC Guidelines for Translating and Adapting Tests, second edition — published by the International Test Commission and freely available at intestcom.org — are the standard here, and three of them bear directly on this. Guideline C-2 asks you to provide statistical evidence of construct, method and item equivalence for all intended populations. Guideline C-1 sets the sample the evidence needs: studies to identify potentially biased test items require a minimum of 200 persons per version. And Guideline SSI-2 draws the conclusion most programmes skip: “Only compare scores across populations when the level of invariance has been established on the scale on which scores are reported.” A ranked shortlist drawn from an Arabic sitting and an English sitting of the same screen is exactly such a comparison.

The item-level statistic that answers it is differential item functioning: whether candidates from two groups who have the same level of the ability being measured answer a given item correctly at different rates. It is not the same as the raw score gap between the cohorts, which may be entirely real, and it is not the same as bias, which is a judgement about why that no statistic can reach. The differential item functioning calculator runs the check from response counts by ability band, and applies the threshold that keeps it honest: ETS flags operationally at a minimum of 200 people in the smaller group and 500 in total, so this is a check you run after pooling several hiring waves, not on one intake.

The practical version for a nationalisation programme is shorter than the theory. If Arabic and English versions of a screen exist, do not report them on one scale until someone has looked at the items; if a single language version is used with cohorts whose first language differs, the same caution applies with less force but does not disappear; and if a shortlist is ranked across cohorts, that is the point at which the question stops being academic. None of this is required by Saudi law today and none of it is what the SVP certificate covers — it is what makes an employer’s own capability evidence defensible, which is the entire reason for gathering it.

What the ministry’s own announcements say — and the three things they do not (added 24 September 2026)

Both numbers in this guide now have a primary source with a date on it, and reading those sources turned up structure the guide had been missing. The Saudi Press Agency item of 20 January 2025 states that the service expanded to 160 labour-exporting countries and 1,007 professions, and — the part nobody repeats — that those are “professions classified within groups 1-3 under the Saudi Unified Classification of Professions, including engineering and health professions”.

That is a more useful sentence than the headline count, because it tells an employer how to look up their own exposure instead of guessing at it. Groups 1–3 of a classification of this family are the managerial, professional and technician-and-associate-professional bands. The programme most people picture — plumbers, welders, carpenters — is the 2021 starting point; the 2025 expansion moved upward into higher-skilled occupations, which is the opposite of what a shortlist assembled from the launch coverage would assume.

HRSD’s own launch announcement, dated 7 March 2021, adds the two structural facts this guide had summarised away. The programme covers “more than 1000 specialized professions belonging to 23 specialty fields as per the Saudi Standard Classification of Occupations” — so whether a role is covered is a classification lookup, not a judgement call. And it runs on two tracks: examinations at selected international centres before a worker arrives, and examinations of workers already in the Kingdom at certified local centres, with enforcement for existing workers “gradually enforced — based on the size of the establishment — starting July 2021”. Establishment size deciding your commencement date is the single most actionable fact in either announcement and it appears in neither of the consultancy explainers that rank for this query.

Now the absence, and it is the reason this guide exists. Neither announcement publishes a pass mark, a re-sit rule or a validity period. The programme is described as a threshold and the location of the threshold is not stated in the material that created it. An employer building a hiring process on top of a pass therefore knows that a standard was met and does not know what the standard was, how close the candidate came to it, or whether the certificate in front of them expires. Those are three questions to put to a certified centre in writing, and they are the exact questions a ranked, scored assessment answers for the candidates in front of you — which is the argument this whole guide makes, now with the gap sourced rather than asserted.

Stated again because it matters: AssessAll has no affiliation with HRSD, Takamol, TVTC or Qiwa, administers no SVP examination, and an AssessAll result is not an SVP certificate. This section reports what two Saudi government sources say, read on 24 September 2026, and nothing on this page is legal or immigration advice.