What are the alternatives to Criteria Corp for pre-employment testing in Australia and New Zealand?
Choosing between Criteria and AssessAll is a decision about whose fairness evidence you rely on. Criteria publishes an independent bias audit run to a New York City standard over 598,877 candidates. AssessAll publishes none, and ships a monitor that computes the same ratio on your own pipeline instead. One is evidence about the instrument, the other about your hiring.
What Criteria is genuinely good at
Read on their own pages on 24 September 2026. Where a figure is the vendor’s own self-report it is attributed rather than asserted — a competitor cannot verify another company’s internal numbers and should not write as though it could.
Criteria publishes a bias audit with a date, a method and a sample size on it, which almost nobody in this category does. Its legal-defensibility page states that an independent audit undertaken in June 2025 found no practically significant disparate impact against any gender, any US racial category, or any combination of the two, that the audit used the data and output standards required by NYC Local Law 144, and that it examined five years of candidate results encompassing over 598,877 candidates. AssessAll has commissioned no equivalent audit and publishes no comparable figure.
Source: criteriacorp.com — Legal Defensibility, read 24 September 2026It defines its fairness threshold in public rather than leaving it to the reader. The same page footnotes practical significance as the US EEOC four-fifths rule, which it states requires the selection rate for any one group to be within 80 per cent of the selection rate for the most selected group. Publishing the threshold alongside the result is what makes the result checkable, and it is the reason this concession is first on the page.
Source: criteriacorp.com — Legal Defensibility, read 24 September 2026It states an ongoing item-review practice, not only a one-off audit. Criteria's legal-defensibility flyer says the company reviews all test content for cultural sensitivity and for differential item functioning to ensure the test is free of content bias across cultural and gender groups. AssessAll has run no DIF analysis on its own aptitude or English forms and says so on its test pages.
Source: criteriacorp.com — Legal Defensibility of Pre-Employment Testing (PDF), read 24 September 2026It is a hiring suite rather than a measurement layer, and it has a genuine regional footprint. The plans page sells three products — Criteria Assess, Criteria Interview and Criteria Hire — with AI proctoring, an AI interview agent, a post-hire development tool and premium integrations as add-ons, and the site runs a separate Australia, Asia, Oceania and Middle East application instance at app-au.criteriacorp.com. AssessAll has no applicant tracking, no video interviewing and no post-hire development module.
Source: criteriacorp.com — Plans and pricing, read 24 September 2026Criteria states its own scale as 80M+ assessments completed, 6B+ talent data points, 20+ years of talent science and more than 4,000 organisations across 60 countries. These are the vendor's own figures, reproduced as stated; AssessAll cannot verify another company's internal numbers and does not assert them.
As stated on criteriacorp.com — home, read 24 September 2026Does Criteria publish a price?
No price is published. Read 24 September 2026: the plans page presents Criteria Assess, Criteria Interview and Criteria Hire, each behind a Request Pricing button, and says to talk to the team for pricing built around your headcount. A free trial is offered with no stated duration or contents. There is no currency figure, no per-candidate rate and no plan tier price anywhere on the page. Quoting on request is a normal enterprise model and is not a criticism, but it means an Australian or New Zealand buyer cannot run a cost comparison before entering a sales process. It also means any third-party page printing a Criteria price has estimated it.
Checked on criteriacorp.com — Plans and pricingAssessAll publishes its unit price: 1 credit = US$0.50. A Self-Insight Profile is 15 credits and an Applied Skill Assessment 26 credits, so a screening set runs about US$7.50 to US$13 per candidate. Volume packs discount 7 per cent at 1,000 credits and 13 per cent at 5,000, there is no seat licence and no annual minimum, and a new organisation starts with 250 free credits.
The full price sheetSide by side
An independent bias audit undertaken June 2025, run to NYC Local Law 144 data and output standards, over five years and more than 598,877 candidates, concluding no practically significant disparate impact by gender, by US racial category, or by their intersection.
None. AssessAll has commissioned no independent bias audit and publishes no adverse-impact figure for its own forms. This row runs against AssessAll and it is the most important row on the page.
If published instrument-level fairness evidence is what you need, this row decides it and it decides it for them.
The categories a NYC Local Law 144 bias audit uses. The Department of Consumer and Worker Protection rule defines category as any Component 1 category required to be reported on the EEOC's EEO-1 report, and requires impact ratios calculated separately for sex categories, race/ethnicity categories and intersectional categories.
Gender and age band, and only on your own pipeline. AssessAll's monitor reads two self-declared dimensions and computes the same ratio over your shortlisted, offered and hired stages.
Neither set covers the attributes an Australian employer is actually asked about. The EEO-1 schedule has no Aboriginal and Torres Strait Islander category, and neither vendor's fairness output has a first-language or non-English-speaking-background dimension.
Not compared. AssessAll has not opened a Criteria customer-facing adverse-impact report, so nothing is claimed about whether one exists or what it shows.
An opt-in adverse-impact monitor over self-declared demographics, computing each group's selection rate across shortlisted, offered and hired, dividing by the highest group's rate, and flagging below 0.80. Groups smaller than five are displayed but never flagged. Aggregates only, and a flag is presented as a prompt to investigate rather than a verdict.
A vendor audit and your impact ratio are different numbers computed on different populations. The four-fifths exposure an employer carries is on their own selection rates, not on the vendor's.
Not published on the pages read. The science page states that Criteria validates across multiple validity measures and that validation studies are run by its I/O psychologists, and it carries no reliability coefficient, no validity coefficient, no standard error of measurement and no norm-group description. Read 24 September 2026.
The same absence. AssessAll publishes no reliability coefficient, no standard error of measurement and no norm group for its aptitude or English forms, and its CEFR band boundaries are documented in code as a launch calibration rather than as empirically derived cut scores.
This row indicts both vendors equally and it is the question to put to whoever you shortlist, in writing.
Not published on the product pages read. The Assess page groups the catalogue into cognitive aptitude, personality, emotional intelligence, risk, skills and game-based assessments with what each measures, and states no question count, no time limit and no score scale. Read 24 September 2026.
Published per form. Each aptitude product states its item count, its time limit and its competency list — for example a 25-item, 22-minute speed-calibrated numerical form against a 20-item, 25-minute form that rewards method over speed.
Seconds per item is the difference between a speed measure and a power measure, and a buyer cannot tell which they bought without both numbers.
None. Three plans, each behind a Request Pricing button, and an invitation to talk to the team for pricing built around your headcount. Read 24 September 2026.
US$0.50 per credit, published, and the same number the checkout charges.
Checked across three markets and six vendors, the providers who publish a price are the ones selling to small and mid-sized buyers. That pattern, not the country, is what predicts it.
A hiring suite: assessment, video interviewing, an AI interview agent, AI proctoring and a post-hire development product, with a separate Australia, Asia, Oceania and Middle East application instance.
A measurement platform with no applicant tracking, no video interviewing and no sourcing. A share link or QR code, no candidate account, and a ranked integrity-banded result you carry into whatever system you already run.
When to choose Criteria instead
These are the situations in which this page recommends the other vendor. They are here because a comparison with no such section is an advertisement, and because the fastest way to waste a quarter is to shortlist a platform that cannot do the thing you needed.
You need published, independent fairness evidence about the instrument itself
Criteria has an audit with a date, a standard and a sample size of more than 598,877 candidates. AssessAll has none. If your board, your legal team or a tender requires the vendor to show an independent audit rather than a monitoring feature, that requirement is met by them and not by this site.
You are replacing a hiring process rather than adding evidence to one
Criteria sells assessment, interviewing and post-hire development as one suite with its own integrations. AssessAll has no applicant tracking, no interview scheduling and no video interviewing, so it assumes you already have those and will keep them.
You are hiring in New York City, or for a US entity that is
Local Law 144 obliges the employer to have a bias audit no more than a year old for an automated employment decision tool, and to publish a summary of it. Criteria at least describes an audit run to that standard; AssessAll has nothing to bring to that obligation. Check the date of whatever a vendor shows you against your own use date, because the one-year clock sits on the employer.
You want game-based assessment specifically
Criteria acquired Revelian, the Australian game-based assessment brand, and lists game-based assessments as a product category. AssessAll has no game-based instruments at all.
You want a single vendor with an established local presence and a local support line
The Australian brand a lot of shortlists still carry, Revelian, now redirects to criteriacorp.com, and Criteria runs a separate regional application instance for Australia, Asia, Oceania and the Middle East. AssessAll has no Australian office.
The four-fifths number you are actually liable for, at the volume you actually hire
Criteria's audit is stable because it is computed on 598,877 candidates. The ratio you are exposed to is computed on your own pipeline, and at the volumes most Australian and New Zealand employers hire at, the same formula behaves completely differently. Here is the arithmetic on a single round of hiring, using the impact ratio exactly as the New York rule defines it: the selection rate for a category divided by the selection rate of the most selected category.
Take 60 applicants and 8 hires, with 38 men and 22 women in the pool. Six of the men are hired and two of the women. The men's selection rate is 15.79 per cent, the women's is 9.09 per cent, and the impact ratio is 0.58 — well under the 0.80 threshold, and flagged.
Now move a single hire. Five men and three women, same pool. The men's rate is 13.16 per cent, the women's 13.64 per cent, and the impact ratio is 0.96. One hiring decision has moved the ratio by 0.38 and turned a flagged pipeline into a clean one. Nothing about the test changed.
That is not an argument that the four-fifths rule is wrong. It is the reason AssessAll's monitor refuses to flag any group smaller than five, and it is the reason a vendor's audit cannot answer the question an employer is asked. The audit tells you whether the instrument disadvantaged a group across 598,877 people. It cannot tell you whether your eight hires did.
Run it on your own last hiring round before you shortlist either vendor. If the answer is that you need an independent, published audit of the instrument — because a tender asks for one, because you hire into New York City, or because your own volumes are large enough that the instrument rather than the round is what moves the number — then Criteria has that evidence and AssessAll does not, and you should buy from them.
What a Local Law 144 bias audit actually measures, and why the scope matters here
Criteria's claim is unusually specific, which is what makes it worth reading carefully rather than just crediting. The audit used the data and output standards required by New York City Local Law 144. The Department of Consumer and Worker Protection's final rule defines the impact ratio as the selection rate for a category divided by the selection rate of the most selected category, or the scoring rate for a category divided by the scoring rate of the highest scoring category, and requires that it be calculated separately for sex categories, race/ethnicity categories and intersectional categories.
The rule then defines category. It means any Component 1 category required to be reported by employers under 42 U.S.C. section 2000e-8(c), as specified in 29 C.F.R. part 1602.7, as designated on the EEOC's Employer Information Report EEO-1. The rule notes that these calculations are consistent with section 1607.4 of the Uniform Guidelines on Employee Selection Procedures.
So the categories are a schedule set by a United States federal reporting form. That is the right schedule for the jurisdiction the law belongs to, and Criteria says so plainly — its finding is stated as no practically significant disparate impact against any gender, any US racial category, or any combination of the two. The wording is accurate and it is not overclaimed.
It is also, read from Australia, a scope statement. The EEO-1 schedule has no Aboriginal and Torres Strait Islander category. It has no category for candidates from a non-English-speaking background. An Australian employer asking whether a cognitive screen disadvantages those groups is asking a question this audit did not ask, and no vendor in this category — including AssessAll — has published an answer to it.
The attributes Australian law protects, and the two either vendor can see
The Fair Work Ombudsman lists the attributes protected from workplace discrimination under the Fair Work Act as race, colour, sex, sexual orientation, age, physical or mental disability, marital status, family or carer's responsibilities, pregnancy, religion, political opinion, national extraction, social origin, breastfeeding, gender identity, intersex status, and experiencing family and domestic violence. It states that the protections cover someone applying for a job as an employee, and that adverse action includes not hiring someone.
Seventeen attributes, and they apply at the point of screening, not only after a person is employed. Two of them — national extraction and social origin — have no counterpart in the EEO-1 schedule at all, and they are exactly the attributes a screen loaded with idiom, local context or academic register risks sorting on without anybody intending it.
Against that list, AssessAll's adverse-impact monitor reads two dimensions: gender and age band. Criteria's published audit reads sex and US race or ethnicity and their intersection. Nobody in this comparison is measuring national extraction or social origin, and this page is not going to pretend otherwise. What a buyer can do is ask for the analysis by name, at what sample size, and how many items came back flagged — and treat a vendor that has never been asked the question as a vendor that does not have the answer.
This is not legal advice and AssessAll is not a law firm. The Fair Work Act, the Racial Discrimination Act and the state and territory acts interact in ways that need an Australian employment lawyer, and nothing on this page certifies compliance with any of them.
What neither vendor publishes, stated once and applying to both
Criteria's science page says the company validates its assessments across multiple validity measures and that validation studies are conducted by its team of I/O psychologists, with hundreds of additional local validation studies run with client organisations. Read on 24 September 2026, it carries no reliability coefficient, no validity coefficient, no standard error of measurement and no norm-group description. The legal-defensibility flyer states that all test content is reviewed for cultural sensitivity and for differential item functioning, and gives no sample size, no date and no item-level result.
AssessAll's position is not better. It publishes no reliability coefficient, no standard error of measurement and no norm group for its aptitude or English forms, it has run no DIF analysis, and its CEFR band boundaries are documented in the code as a launch calibration rather than as empirically derived cut scores. Where AssessAll publishes more is item count, time limit and the competency list per form, which is the arithmetic of seconds per item that separates a speed measure from a power measure.
The useful version of this section is not a scoreboard. It is a list to send whoever you shortlist: what is the internal consistency of this form and on what sample; what is the standard error at my cut score; who is the norm group; when was the last DIF run, by group, at what N, and how many items came back in the highest category. Any vendor that answers those four questions in writing has told you more than either of the pages compared here.
Why this page exists: what the rest of the lane says about Criteria
Two vendor-written Criteria alternatives pages were opened on 24 September 2026 before this one was written. Compono's guide compares Criteria against TestGorilla, the Predictive Index, SHL and itself on use case, target market, assessment type, candidate experience, scientific rigour, platform complexity and integrations. It publishes no reliability coefficient, no standard error, no norm group, no item count, no time limit, no cut-score guidance, no price and no adverse-impact data.
TestGorilla's page names TestGorilla, Criteria, the Predictive Index, Wonderlic, Mercer Mettl, iMocha, Harver, TestDome and Vervoe, and compares them on library size, validation, integrations, pricing model, free plan, anti-cheating, video interviewing, interface and support. It prints prices for five of them — from $142 a month for TestGorilla, from $4,950 a year for the Predictive Index, from $200 a month for iMocha, from $100 per five candidates for TestDome, from $19 a month for Vervoe — and no price for Criteria, which matches what Criteria's own plans page shows. It publishes no psychometric quantity either.
So the lane is contested in the sense that vendor-written pages already exist, and empty in the sense that not one of them answers a measurement question. The thing Criteria actually publishes that almost nobody else does — a dated, independent, 598,877-candidate bias audit — appears on neither rival page. A comparison that omits a competitor's strongest published evidence is not a comparison, and this page opens on it for that reason.
Questions people ask
Does Criteria Corp publish its pricing?
No. Read on 24 September 2026, the plans and pricing page presents three products — Criteria Assess, Criteria Interview and Criteria Hire — each behind a Request Pricing button, and invites you to talk to the team for pricing built around your headcount. A free trial is offered without a stated duration or contents. There is no currency figure anywhere on the page, so any third-party page quoting a Criteria per-candidate price has estimated it.
Is Revelian still a separate product in Australia?
No. Checked 9 September 2026 and unchanged at 24 September 2026: revelian.com redirects to criteriacorp.com. Criteria runs a separate application instance for Australia, Asia, Oceania and the Middle East at app-au.criteriacorp.com, and lists game-based assessments — the category Revelian was known for — among its products. If your shortlist is a few years old, check which brands on it still exist as separate products before spending a call on each.
Has Criteria Corp had a bias audit?
Criteria's legal-defensibility page states that an independent audit of its assessments undertaken in June 2025 found no practically significant disparate impact against any gender, any US racial category, or any combination of gender and US racial category, that the audit used the data and output standards required by NYC Local Law 144, and that it examined five years of results encompassing over 598,877 candidates. That is the vendor's description of a third party's finding; the underlying impact ratios are not published on that page. AssessAll has commissioned no equivalent audit.
Does a vendor's bias audit cover my hiring?
No. A Local Law 144 bias audit is computed on the tool's candidate population and tells you about the instrument. The four-fifths exposure an employer carries is computed on their own selection rates, at their own cut score, in their own applicant pool. The two are different numbers on different populations, and at small hiring volumes the second one moves enormously on single decisions — in the worked example on this page, one hire out of eight moves the impact ratio by 0.38.
Does the Criteria audit answer an Australian fairness question?
Only partly, and Criteria does not claim otherwise. The NYC rule defines the categories as those designated on the EEOC's EEO-1 report, so the audit is calculated over US race or ethnicity categories, sex categories and their intersections. The EEO-1 schedule has no Aboriginal and Torres Strait Islander category and no non-English-speaking-background category. The Fair Work Ombudsman's list of protected attributes includes national extraction and social origin, which have no EEO-1 counterpart at all. This is a scope observation, not a criticism, and it applies with more force to AssessAll, which publishes no audit of any kind.
What does AssessAll do about adverse impact?
It ships an opt-in monitor, not an audit. Where candidates volunteer demographic bands, the platform computes each group's selection rate across the shortlisted, offered and hired stages, divides by the highest group's rate, and flags anything below 0.80. Groups smaller than five are displayed but never flagged, because a rate computed on three people is noise. It works on aggregates only, never surfaces an individual's declared demographics, and presents a flag as a prompt to investigate rather than a verdict. It is not an independent bias audit and it is not a Local Law 144 audit — those require an independent auditor, and no platform can audit itself.
How much does AssessAll cost for an Australian small business?
One credit is US$0.50. A Self-Insight Profile is 15 credits and an Applied Skill Assessment 26, so a screening set is about US$7.50 to US$13 per candidate. Volume packs discount 7 per cent at 1,000 credits and 13 per cent at 5,000. There is no seat licence and no annual minimum, and a new organisation starts with 250 free credits. Because Criteria publishes no figure, a like-for-like cost comparison is not possible until you are in their sales process.
Every page this comparison was built from
Each of these was opened on . Nothing on this page is claimed about a product page that was not opened — where a row would have required one, the row says so instead of inferring it.
"An independent audit of Criteria's assessments undertaken in June 2025 found no practically significant disparate impact against any gender, any US racial category, or against any combination of gender and US racial category. The 2025 bias audit used the data and output standards required by NYC Local Law 144, and examined the results obtained by job candidates who completed Criteria assessments over the previous five years (encompassing over 598,877 candidates)." Footnote: "Practical significance refers to the US EEOC 4/5ths rule, which requires that the selection rate for any one group must be within 80% (or 4/5ths) of the selection rate for the most selected group." No impact ratios, no reliability coefficient and no norm group on the page. Read 24 September 2026.
General validation claims — validating across multiple validity measures, validation studies by a team of I/O psychologists — with no reliability coefficient, no validity coefficient, no standard error of measurement and no norm-group description. Read 24 September 2026.
Criteria Assess, Criteria Interview and Criteria Hire, each with a Request Pricing button; "Talk to our team for pricing built around your headcount"; add-ons listed as AI Proctoring, AI features, AI Interview Agent, Criteria Develop and Premium Integrations; a free trial offered with no stated duration or contents. No currency figure anywhere. Read 24 September 2026.
Six catalogue categories — cognitive aptitude, personality, emotional intelligence, risk, skills, game-based — each with what it measures. No question count, no time limit, no score scale and no norm group on the page. Read 24 September 2026.
Self-reported figures: "80M+ Assessments completed", "6B+ Talent data points", "20+ Years of talent science", "4000+ Organizations worldwide", "serving customers in over 4,000 organizations across 60 countries". A separate login for Australia, Asia, Oceania and the Middle East at app-au.criteriacorp.com. Read 24 September 2026.
"We do this by reviewing all test content for cultural sensitivity and for differential item functioning to ensure the test is free of content bias across cultural and gender groups." One academic citation, Terpstra, Mohamed and Kethley (1999). No sample sizes, no impact ratios, no coefficients and no audit dates in the document. Read 24 September 2026.
"Impact ratio" means either (1) the selection rate for a category divided by the selection rate of the most selected category or (2) the scoring rate for a category divided by the scoring rate for the highest scoring category. "Category" means any component 1 category required to be reported by employers pursuant to subsection (c) of section 2000e-8 of title 42 of the United States Code as specified in part 1602.7 of title 29 of the Code of Federal Regulations, as designated on the Equal Employment Opportunity Commission Employer Information Report EEO-1. The rule requires the impact ratio to be calculated separately for sex categories, race/ethnicity categories and intersectional categories, states that the calculations are consistent with 29 C.F.R. section 1607.4, and clarifies that an AEDT may not be used if its most recent bias audit is more than a year old. Read 24 September 2026.
Protected attributes listed as "race, colour, sex, sexual orientation, age, physical or mental disability, marital status, family or carer's responsibilities, pregnancy, religion, political opinion, national extraction, social origin, breastfeeding, gender identity, intersex status, experiencing family and domestic violence." Protections stated to cover "someone applying for a job as an employee" and "a new employee who hasn't started work"; adverse action stated to include "not hiring someone". Read 24 September 2026.
Compares Criteria, TestGorilla, the Predictive Index, SHL and Compono on use case, target market, assessment type, candidate experience, scientific rigour, platform complexity and integrations. No reliability coefficient, no standard error, no norm group, no item count, no time limit, no cut-score guidance, no price and no adverse-impact data. Read 24 September 2026.
Names TestGorilla, Criteria, the Predictive Index, Wonderlic, Mercer Mettl, iMocha, Harver, TestDome and Vervoe. Prints starting prices for five — TestGorilla "From $142/mo (billed annually)", the Predictive Index "From $4,950/yr", iMocha "From $200/mo", TestDome "From $100/5 candidates", Vervoe "From $19/mo (Pay & Go)" — and no Criteria price. No reliability coefficient, no standard error, no norm group, no item count, no time limit, no cut-score guidance and no adverse-impact data. Read 24 September 2026.
Related
- Assessment platforms compared — Australia & New Zealand
- Pre-employment testing for a small business in Australia and NZ
- Adverse impact ratio — the four-fifths threshold
- TestGorilla alternatives
- Pricing — the published unit price
Founder of AssessAll and of Bodhih Training Solutions, a corporate training company in Bangalore. Works on assessment design, scoring and reporting across hiring, L&D and certification programmes.
Last reviewed
AssessAll is not affiliated with Criteria, and product and company names on this page belong to their respective owners. This comparison is published by a competing vendor: read it alongside the other vendor’s own materials, and check the linked pages rather than taking either side’s summary of them. Rows are re-verified monthly and deleted rather than left standing when they cannot be.