Best pre-employment assessment tools in 2026, AI features compared

As of November 2026, the best pre-employment assessment tools are TestGorilla for small and mid-size teams that want a large self-serve test library with a free plan and published prices, Criteria for cognitive, personality and emotional intelligence tests with AI proctoring and a published bias audit result, Vervoe for job-specific work tasks generated from your job description and graded by AI, Harver, which now includes pymetrics, for high-volume hiring, and SHL for large employers testing in many countries and languages. Hogan suits leadership and professional selection built on personality, Wonderlic suits teams that want one combined score without paid certification, and iMocha suits technical and English-language skills testing at scale. Whichever you choose, the employer, not the vendor, has to show that the test is job related and fair to every group it screens.

This page compares the platforms. For what each kind of test measures, start with our guide to types of pre-employment assessments; for coding tests, see technical interview platforms.

How we chose these tools

• It runs a pre-employment test itself: cognitive, personality, skills, situational judgment, language or a job simulation, taken by the candidate before a hiring decision. Tools that only interview are in our AI interviewer roundup.
• It is sold today under that name. We left out Plum because plum.io now redirects to Phenom's April 28, 2026 announcement that it acquired Plum, so we could not confirm a standalone product. We left out Indeed Assessments because Indeed's own support center carries an article titled "Product Update: Indeed is Discontinuing Assessments".
• It says something checkable about validation, fairness or accommodations, so a buyer has a document or a process to ask about.
• It serves employers or agencies, from a ten-person company to enterprise volume hiring.
• Every fact comes from the vendor's own site or help center, checked in November 2026 and linked below. We did not score or rank the products.

How do the pre-employment assessment tools compare?

As published by each vendor, as of November 2026. "Not published" means we could not find it on the vendor's own site. Every tool here is taken by the candidate through a link in a browser.

Sources, checked November 2026: TestGorilla pricing, test library, assessments technology and ATS integrations; Criteria plans, AI Proctoring, AI Scoring, integrations and free trial; Vervoe pricing, AI Assessment Builder, AI scoring, anti-cheating and integrations; Harver assessments, pymetrics, responsible AI and integrations; SHL cognitive, language, video interviews, ATS integrations and SHL Online; Hogan FAQ and Convogo announcement; Wonderlic Select and test taker hub; iMocha skills assessment and integrations.

What validation and accessibility evidence does each vendor publish?

This is what each vendor says on its own site; none of it is our finding, and none of it replaces your own adverse impact check on your own applicants.

Sources, checked November 2026: TestGorilla technology and accommodations; Criteria legal defensibility and time adjustments; Vervoe and its accessibility article; Harver responsible AI; SHL inclusion and accessibility and candidate accommodations; Hogan FAQ and accessibility; Wonderlic testing accommodations; iMocha.

Which tools suit small and mid-size teams?

TestGorilla

TestGorilla now sells Gorilla Assessments and a separate sourcing product. Assessments has a Free plan with 10 credits a month and unlimited use of five essential tests (problem solving, communication, time management, Big 5 and motivation); Core starts at $215 a month per account billed annually with the full library; Plus starts at $520 a month and adds the AI interview library, job simulations, custom tests, ID verification and the ATS integrations (pricing). AI scoring is optional: it scores open-ended answers against rubrics, shows the reasoning, and every score can be overridden; integrity signals are presented as flags for review, "never an automatic disqualification" (TestGorilla). Trade-off: job simulations and integrations sit on the top plan. Best for: teams that want to start free and pay as hiring grows.

Criteria

Criteria sells three packages, Criteria Assess, Criteria Interview and Criteria Hire, all priced on request, with AI upgrades and add-ons on top; Criteria Assess includes every test, non-AI proctoring, standard ATS integrations and SSO (Criteria plans). AI Proctoring captures the candidate's image before and during the test, samples the screen and audio, and gives your team a report of flagged incidents to review (AI Proctoring). Tell candidates about the camera and audio sampling before they start. Best for: employers that want well-documented aptitude and personality tests with video interviewing from one vendor, after a 14-day trial.

Wonderlic

Wonderlic Select combines cognitive ability, personality and motivation into one score against job profiles, with interview guides built from the results; Wonderlic says no paid certification is needed and that price depends on the package and your employee count (Wonderlic Select). The cognitive section is short and timed, often around 12 minutes; the personality and motivation parts are untimed (test taker hub). Trade-off: no AI features or published validation document on the product page. Best for: hiring managers who want one clear number per role.

Which tools suit skills-first and technical hiring?

Vervoe

Vervoe's AI Assessment Builder reads a pasted job description and builds up to four skill groups with up to three questions each, 9 to 12 questions in total, which you preview and edit before sending (AI Assessment Builder). Answers can be video pitches, written plans or role-plays; machine learning grades them and ranks candidates, with scores broken out by skill group (AI scoring). Its anti-cheating relies on randomized questions, geolocation checks and plagiarism detection that now includes ChatGPT and web sources (anti-cheating). Pricing is by request. Best for: roles where a realistic task says more than a multiple-choice test.

iMocha

iMocha pairs 10,000+ skills tests across 300+ job roles with coding assessments on 30+ compilers, AI-LogicBox for logic without code, AI-EnglishPro, a CEFR-aligned English test scored by AI for grammar, fluency and tone, and a Smart Proctoring Suite that monitors webcam, screen and tab switching and logs violations with timestamps (iMocha). It is positioned as a skills intelligence platform for workforce planning as well as hiring, and its pricing page leads to a demo booking. Best for: IT services, BPO and enterprise teams testing technical and English skills at volume.

Which tools suit enterprise and high-volume hiring?

Harver, now including pymetrics

Harver lists more than 450 assessment options in dozens of languages, from situational judgment and cognitive tests to contact center agent simulations and spoken language tests (Harver assessments). Its pymetrics page says "pymetrics, now part of Harver" and describes 12+ games measuring traits such as attention, effort and risk tolerance, about 25 minutes long, with immediate feedback for every candidate (pymetrics). Harver says it uses AI for language assessment, video interview analysis, job-success models built on incumbent data and fraud detection, with human oversight (responsible AI). Pricing is not published. Best for: contact centers, BPOs and retailers hiring hundreds a month; see also AI tools for high-volume hiring.

SHL

SHL's catalog runs from Verify cognitive tests, which SHL says are automated and remotely proctored, to the OPQ personality questionnaire, situational judgment tests, call center and coding simulations, and language tests a candidate can take by mobile, desktop or phone (SHL cognitive, language). It integrates with more than 80 applicant tracking systems. Enterprise pricing is by quote; SHL Online is a self-service store aimed at smaller employers (SHL Online). SHL also offers assessment and development centers; see what is an assessment center. Best for: multinational employers that need the same tests in many countries.

Which tool suits leadership and personality-based selection?

Hogan Assessments

Hogan's core inventories cover everyday personality (HPI), derailers under pressure (HDS) and values (MVPI), each 15 to 20 minutes, with the HBRI reasoning test at about 35 minutes; Hogan says the tools are available in over 40 languages, pricing varies by client, and most products require certification before you can use them (Hogan FAQ). In July 2026 Hogan acquired Convogo, an AI platform that helps coaches and talent professionals combine Hogan results with other information (Hogan). Trade-off: it measures personality and reasoning rather than job skills, and we found no ATS integrations listed on its site. Best for: leadership hiring where certified users interpret the results.

Which assessment type predicts what?

The tool matters less than the test type you pick for the job. The U.S. Office of Personnel Management's assessment pages, checked November 2026, summarize each type; our types of pre-employment assessments guide adds the meta-analytic numbers.

The usual design is a short, low-effort test early, a structured interview in the middle, and a work sample for finalists. For game-based tests such as pymetrics and Criteria's, ask the vendor what each game measures and what validity evidence it has for your kind of job.

How do you choose a pre-employment assessment tool?

• Small teams hiring a few people a quarter: start with TestGorilla's free plan or Criteria's trial, and pick two or three tests per role rather than a full battery.
• Hundreds of hires a month in one job family: Harver or SHL, where simulations, language tests and ATS automation matter more than library size.
• Professional and technical roles: Vervoe for realistic tasks, iMocha or our technical platforms for code.
• Leadership roles: Hogan or SHL's OPQ, interpreted by someone trained on the instrument.
• Agencies: confirm whether the client or the agency owns the test, the results and the adverse impact records before you send a single invite.

What should AI do in an assessment, and what should it not?

AI is useful for drafting tests from a job description, scoring open answers against a rubric and flagging possible cheating. It should not make the hiring decision. Ask every vendor three things: can a person see why the AI gave each score, can a person override it, and does a cheating flag ever reject a candidate automatically. TestGorilla says every AI score shows its reasoning and Vervoe says every score is traceable to the evidence behind it; Criteria says its video scoring reads only the transcript. A test the AI wrote from your job description still needs a person to check every question against the job; the Uniform Guidelines say a procedure's name or label is no evidence of validity.

For candidates, keep total test time short, say what is monitored (camera, screen, audio), say where AI is used, and put the accommodation route in the invitation. Some vendors handle requests themselves and others leave the decision to you, as the table above shows. Our guide to ADA accommodations in interviews covers wording.

How do you use pre-employment assessments lawfully?

• Title VII disparate impact. A neutral test that screens out a protected group at a higher rate creates liability unless the employer shows it is job related for the position and consistent with business necessity (42 U.S.C. 2000e-2(k)). The same section allows professionally developed ability tests that are not designed, intended or used to discriminate (subsection (h)), and bars adjusting scores or using different cutoffs by race, color, religion, sex or national origin (subsection (l)). Our Title VII and AI hiring page covers what changed in federal enforcement.
• The Uniform Guidelines. 29 CFR Part 1607, current on eCFR as of November 2026, treats a selection rate below four-fifths of the highest group's rate as evidence of adverse impact (1607.4D), asks you to prefer an equally valid procedure with less adverse impact (1607.3B), lets you rely on a publisher's validity study only if your job is shown to be similar, while leaving you responsible (1607.7), and rules out promotional literature as evidence of validity (1607.9). Keep the impact records it describes (1607.15).
• ADA and timed tests. The EEOC's regulation requires tests to measure the skill, not a sensory, manual or speaking impairment, unless that skill is what the test measures (29 CFR 1630.11). The EEOC's Job Applicants and the ADA says an employer may have to adjust the time allowed for a written test, and may tell every applicant the process includes a timed test and ask who needs an accommodation.
• NYC Local Law 144. If an assessment's automated score helps decide who advances for a New York City job, check whether it counts as an automated employment decision tool. If it does, the tool needs a bias audit within one year of use, a public summary of the results, and notices to candidates, which DCWP says must go out 10 business days before use (NYC DCWP). See our Local Law 144 bias audit guide, and the state-by-state overview for laws such as Illinois' AI Video Interview Act when a vendor analyzes video.

The EEOC's 2023 technical assistance on AI and Title VII was removed from its website in January 2025, so do not rely on vendor slides that still quote it. This is general information as of November 2026, not legal advice.

Assessment plus interview: how do you combine them?

Decide before the first candidate how the two will be combined: a pass mark that gates the interview, a weighted score, or the assessment as one input per competency. Use the assessment for what it measures well, such as a skill or reasoning, and the structured interview for behavior, judgment and motivation, so the two do not double-count the same thing. When they disagree, check what each measured before you trust either; our guide to when interview and assessment results disagree walks through it, and the candidate assessment summary template puts both on one page for the hiring manager.

For the interview that follows the test, Interview Signal, a desktop app for the person running the interview, gives a live question guide and a scorecard built from the candidate's own quotes with no bot on the call, but it does not run assessments or tests. We make Interview Signal; everything about the other products on this page comes from their own sites, linked and dated.

Setup checklist

1. Write the job's day-one must-haves, and separate them from what you will train.
2. Pick one test per must-have, using the most job-like measure you can afford.
3. Ask the vendor for its validation evidence and fact sheets for those tests, and compare the jobs studied with yours.
4. Ask whether any AI scores, ranks or flags candidates, how a person reviews it, and for any bias audit summary.
5. Set cut scores and the interview-combination rule before inviting anyone.
6. Put the time required, what is monitored, where AI is used and how to request an accommodation in the invitation.
7. Pilot on one requisition and check pass rates by group with the four-fifths calculation.
8. Review pass rates, complaints and accommodation requests at least once a year, and keep the records.