Hiring Assessments: A Small Business Guide to Pre-Hire Testing
What pre-hiring assessments are, 5 types with validity data, legal basics, how to choose, and what to do after you hire. Written for small businesses.
Every week, a small business owner asks me some version of the same question: "Should I be testing candidates before I hire them?" This guide to hiring assessments is my answer. I build an HR platform that starts with onboarding. FirstHR takes over once you make the offer, so pre-hire testing, candidate scoring, and psychometric evaluations come from other tools.
The short version: it depends on your hiring volume, the role, and whether you are willing to do it legally. Most small businesses do not need formal assessments for every hire. But for high-volume roles, safety-critical positions, or roles where a bad hire costs $50,000+, assessments add real value.
Below you will find the five main categories of pre-hiring assessments, their predictive validity (how well they actually predict job performance), the legal rules you cannot skip, and a decision framework for whether your business needs them. The guide ends with the post-hire step that connects assessment results to actual on-the-job success, because that step is the easiest one to drop.
- Sign the offer letter
- Sign the employee handbook
- Workplace Safety course
What Is a Hiring Assessment?
A hiring assessment (also called a pre-hiring assessment, pre-employment test, or pre-hire assessment) is a standardized evaluation given to job candidates to measure attributes relevant to job performance. The goal is to add objective data to the hiring decision beyond what a resume and interview provide.
The term covers a wide range of tools, from a 10-minute cognitive ability quiz taken on an online platform to a full-day assessment center with simulations, role-plays, and panel evaluations. For small businesses, the relevant tools sit at the simpler end: short online tests that are generally inexpensive to administer and produce a score or ranking that supplements your interview evaluation.
A few things assessments are not. They are not background checks (which verify factual history, not predict performance). They are not reference checks (which gather subjective feedback from prior employers). And they are not a substitute for structured interviews, which now rank as the single strongest predictor of job performance even without any assessment tool.
Why Small Businesses Use Assessments (And Why Many Do Not)
The case for assessments is strongest when hiring volume is high and a wrong hire is expensive. An estimate reported by the Society for Human Resource Management (SHRM) in 2017 puts the cost of recruiting, hiring, and onboarding a single employee as high as $240,000. At a small business, a hire that does not work out also disrupts the whole team.
Assessments reduce bad hires by adding objective measurement to a process that is otherwise highly subjective. The 2022 matrix from Sackett and colleagues, which ranks hiring methods by how well they predict job performance, shows what subjectivity costs: an unstructured interview comes in at 0.19 against 0.42 for a structured one. The format of the conversation alone moves prediction more than most buyers expect.
The Equal Employment Opportunity Commission (EEOC) treats properly validated assessments as a legitimate selection tool. The catch is that once adverse impact appears, meaning one protected group passes at a significantly lower rate than another, the employer carries the burden of showing the procedure is job-related and consistent with business necessity.
| When Assessments Add Value | When They Do Not |
|---|---|
| High-volume roles (hiring the same position 3+ times/year) | One-off senior hires where context and judgment matter more than testable skills |
| 50+ applications per posting (need to filter efficiently) | Small applicant pools (10-15 candidates) where manual review is feasible |
| Safety-critical roles (driving, equipment, patient care) | Creative or relationship-based roles where output is hard to standardize |
| Roles with clearly testable skills (coding, data entry, accounting) | Trust-based roles (executive assistant, chief of staff) where cultural fit matters most |
| High cost of bad hire ($30,000+ replacement cost) | Low-stakes temporary or seasonal positions |
The honest assessment for most small businesses: if you hire 5 to 10 people per year for different roles, a structured interview with consistent questions and a scoring rubric is usually enough. It is the top-ranked predictor in the 2022 matrix and costs nothing but interviewer time.
Formal assessments become clearly worthwhile when you hire the same role repeatedly (customer service, warehouse, sales) and need to screen a high volume of applicants consistently.
Nick Anisimov, FirstHR founder
The 5 Types of Pre-Hiring Assessments
This guide sorts the pre-hiring assessment market into five main categories: cognitive ability tests, personality assessments, job knowledge tests, work samples and simulations, and integrity tests. The assessment catalog at the Office of Personnel Management (OPM) lists more, such as situational judgment tests, physical ability tests, and assessment centers, but these five are the practical options for a small team.
Each category measures different things, has different predictive validity (how well it predicts actual job performance), and carries different legal risk. Understanding the categories helps you choose the right tool for your role without buying a vendor's marketing pitch about being "the only assessment you need."
The validity numbers above come from the current meta-analytic matrix that Sackett and colleagues (2022) published in the Journal of Applied Psychology, not from the widely quoted Schmidt and Hunter figures of 1998. Structured interviews rank first at 0.42, with job knowledge tests second at 0.40.
The revision matters most where it moved numbers down. The 1998 analysis put work samples at 0.54 and cognitive ability at 0.51. The 2022 analysis found the older studies over-adjusted when correcting for range restriction, and the corrected figures read 0.33 and 0.31. No single tool is sufficient, and the 2022 authors argue for a selection system with several components rather than one test.
Work samples are still the format a small business can act on immediately, even at the revised 0.33. The Office of Personnel Management describes work sample tasks as carrying a high degree of content validity, with performance on the test relating highly to performance on the job. That is a property of the format, not of any vendor.
Aptitude Tests, Skills Tests, and Other Names for the Same Five
Vendors and job boards rarely use the five category names above, so the mapping is worth having. A pre-employment aptitude test is a cognitive ability test. A pre-employment skills assessment test is either a job knowledge test or a work sample, depending on whether it asks what the candidate knows or asks them to do it.
| What the vendor or the listing calls it | What it usually measures | Which of the five it is |
|---|---|---|
| Aptitude test, cognitive test, reasoning test | Numerical, verbal, and logical reasoning, usually under a time limit | Cognitive ability |
| Skills assessment test, proficiency test, technical test | What the candidate already knows about the craft: syntax, rules, product, regulations | Job knowledge |
| Job simulation, take-home task, practical exercise | The candidate producing an actual piece of the work | Work sample |
| Behavioral assessment, work style profile, personality inventory | Traits and preferences rather than ability | Personality |
| Honesty test, reliability test, dependability screen | Likelihood of counterproductive behavior on the job | Integrity |
One caution on aptitude tests in particular. The label travels with a lot of unvalidated products, and a timed reasoning quiz sold under that name carries exactly the same disparate impact exposure as any other cognitive measure. Ask for the validation evidence before the familiar word does the reassuring.
The Legal Part You Cannot Skip
Pre-hiring assessments are legal in the United States. But "legal" comes with conditions. The EEOC and the Uniform Guidelines on Employee Selection Procedures (1978) establish three requirements that every covered employer using assessments must meet.
Whether federal law covers you depends on headcount. Title VII of the Civil Rights Act and the ADA cover employers with 15 or more employees, the Age Discrimination in Employment Act (ADEA) covers those with 20 or more, and state or local laws may also apply.
One definition is worth knowing before the requirements themselves: the Uniform Guidelines define a selection procedure as any measure, combination of measures, or procedure used as a basis for any employment decision. That definition covers an algorithmic screening model as squarely as a paper test, so "we just use software" is not an exit from any of this.
The first requirement is job-relatedness. The assessment must measure something that is actually required to perform the job. A math test for an accounting role is job-related. A math test for a receptionist who never uses math is not. Every test must connect to a documented essential function of the position.
The second is consistent administration. Every candidate for the same position must take the same test under the same conditions, apart from disability accommodations. You cannot give an assessment to some candidates and not others, or give different versions to different demographic groups. The process must be standardized.
The third is the absence of unjustified disparate impact. If your assessment produces significantly different pass rates for different protected groups (race, sex, age, disability status), you must be able to show the test is job-related, which in practice means validation evidence for the job.
Validation does not settle the question on its own. A challenger can still point to an equally effective alternative with less adverse impact, and the Uniform Guidelines expect you to look for one.
Cognitive ability and job knowledge tests carry the most exposure here. OPM notes that cognitive tests and job knowledge tests typically produce larger racial and ethnic score differences than other valid predictors, while personality and integrity tests generally show few, if any.
The arithmetic in the Uniform Guidelines is the four-fifths rule. Under 29 CFR 1607.4(D), a selection rate for any race, sex, or ethnic group that falls below four-fifths of the rate for the group with the highest rate is generally regarded as evidence of adverse impact. Run that ratio on your own pass rates before anyone else does.
The Uniform Guidelines themselves may not stay on the books. The EEOC's regulatory agenda lists their rescission at the final rule stage, with final action projected for November 2026. The duty underneath stays: Title VII itself, at 42 U.S.C. 2000e-2(k), still makes a test with disparate impact unlawful unless it is job-related and consistent with business necessity.
5 Red Flags the EEOC Looks For
The practical rule for small businesses: use job-related skills tests or work samples that directly mirror job duties. Because they are inherently job-related, they are the easiest to defend if pass rates do differ between groups. Avoid generic personality tests unless the vendor provides validation data specific to your type of role.
Then document everything: what test you used, why you chose it, how you set the pass score, and how you administered it. This documentation is your defense if a hiring decision is ever challenged.
One page per role, filled in before the first candidate takes the test. The section that does the most work is the first one: if you cannot name the essential function a test question maps to, that question does not belong in the test.
If you want to read the rules themselves, two documents cover almost everything a small employer needs. The EEOC fact sheet on employment tests and selection procedures sets out the legal framework, and the ADA regulation at 29 CFR 1630.14 sets out when a job applicant can be given a medical examination, which is only after a conditional offer.
Personality tests can land on either side of that line. The EEOC's guidance on preemployment medical examinations treats a psychological test as medical when it can point to a mental disorder or impairment; a test that measures only things such as honesty, tastes, and habits does not count as one.
Do You Actually Need a Pre-Hiring Assessment?
The honest answer for most small businesses: probably not for every hire. Assessments make sense in specific situations, and the five questions below help you decide whether your next hire warrants a formal assessment or whether a structured interview with a work sample is enough.
If you answered "no" to most of these questions, a structured interview process (consistent questions, scoring rubric, multiple interviewers when possible) gives you strong predictive validity without the cost and complexity of formal assessments.
If you answered "yes" to 3 or more, consider adding an assessment to your hiring process. Start with a work sample (free, and it outpredicts any personality test) before investing in a vendor platform. Add a vendor-hosted cognitive or skills test when your hiring volume justifies the per-candidate cost.
How to Choose an Assessment Without Getting Sold
To choose an assessment without getting sold, work through four steps in order: define what the job requires, ask for validation evidence, ask for adverse impact data, and take the test yourself before any candidate does.
Starting from the job matters because the pre-employment assessment market is crowded with vendors making strong claims about predicting job performance, reducing turnover, and improving quality of hire. Some of those claims are supported by evidence. Many are not. Here is how to evaluate assessment tools as a small business buyer without getting oversold.
Step 1: Start With Job Analysis, Not Tool Shopping
Before looking at any vendor, define what the job actually requires. What skills does the person use daily? What knowledge is a prerequisite versus trainable? What behaviors predict success in this role at your company? Write these down. Then find a test that measures those specific things. Do not start with a tool and work backward to justify it.
Step 2: Ask for Validation Evidence
Validation means the vendor has data showing the test predicts performance for your type of role. Ask for criterion-related validity studies (do people who score high on the test actually perform better on the job?). A vendor who cannot provide this data is selling a survey, not a validated assessment. Look for sample sizes of 100+ and correlation coefficients above 0.20.
Step 3: Ask About Adverse Impact Data
Does the test produce significantly different pass rates for different demographic groups? Reputable vendors measure this and provide the data. If a vendor says "our test is fair" but cannot show you adverse impact statistics, they either have not measured it or the numbers are unfavorable.
Step 4: Check the Candidate Experience
Take the test yourself. If it takes longer than 30 minutes, produces confusing instructions, or feels irrelevant to the job, your candidates will feel the same way. Candidate drop-off from long or poorly designed assessments is a real cost.
| Question to Ask the Vendor | Good Answer | Red Flag |
|---|---|---|
| Can you provide validation data for this role type? | Shares criterion-related validity study with sample size and correlation | Says 'our test is validated' without providing data |
| What is the adverse impact ratio by protected group? | Shares actual adverse impact data and 4/5ths rule analysis | Says 'our test is fair' without data |
| Who designed the test? | Names I/O psychologists or psychometricians with credentials | Vague reference to 'our science team' |
| How long does the test take candidates? | 15-30 minutes with specific time data | Says 'it varies' or 'candidates love it' |
| What accommodations do you support? | Specific ADA accommodations (extra time, alternative format, screen reader) | Says 'we comply with ADA' without specifics |
How the Score Enters Your Candidate Evaluation
A score is one input, and the fastest way to waste it is to let it become the decision. Put every candidate for the opening on a single sheet: for each one, the three or four criteria from the job description down one side, and a column each for the interview score, the assessment result, and the work sample.
Score each criterion the same way for everyone, on a scale with a written anchor describing what a 3 and a 5 look like, and fill it in immediately after each step rather than reconstructing it at the end. Read down the columns before you read across the rows, because that is where an inconsistent scorer shows up.
Never rank on the assessment column alone. Use it to settle a question the interview left open, and where two candidates sit within a point of each other, let the work sample break the tie rather than the personality profile. An interview evaluation form keeps the scoring consistent when more than one person is interviewing.
What Happens After the Assessment: The Step Every Guide Skips
What happens after the assessment is onboarding, and onboarding decides whether a good pick turns into a good hire. Every article about hiring assessments ends at the same point: the candidate passes the test, you make the hiring decision, and the article is done. But the assessment predicted fit. It did not deliver it.
The person still needs to show up on Day 1, learn the role, build relationships, and produce results. That is where a hire succeeds or fails, and passing a cognitive test does not take care of any of it.
The bridge between "passed the assessment" and "productive at 90 days" is structured onboarding. Here is what it looks like in practice.
The assessment data should inform the onboarding plan. If a candidate scored exceptionally high on product knowledge, they need less product training and more cross-functional exposure. If they scored average on communication but high on technical skills, the 30-day plan should include more shadowing and feedback on client interactions. When assessment results never reach the onboarding plan, the company has paid for data it does not use.
FirstHR runs the post-hire workflow: e-signatures on offer letters and policy acknowledgments, training modules assigned with deadlines, and task workflows that put each check-in on a due date.
Nick Anisimov, FirstHR founderCommon Mistakes Small Businesses Make With Assessments
The most common mistakes range from letting the score make the hiring decision to never using the results once the person starts work. Here are seven, with why each one happens and how to fix it.
| Mistake | Why It Happens | The Fix |
|---|---|---|
| Using assessments as the sole decision-maker | The score feels objective and defensible | Assessments supplement interviews, they do not replace them. Use both. |
| Testing skills that can be trained | Vendors sell tests for everything | Test for abilities the person needs on Day 1. Train the rest during onboarding. |
| Applying the same test to every role | Simplicity and vendor pricing incentives | Match the test type to the role. Cognitive for complex roles, work samples for hands-on roles. |
| Skipping ADA accommodations | Not aware of the requirement | Ask every candidate before the test whether they need accommodations. Document the process. |
| Not connecting assessment results to onboarding | Assessment and onboarding are managed by different people (or nobody) | Share relevant results with the manager. Use them to customize the 30-day training plan. |
| Buying enterprise tools for SMB problems | Vendor marketing creates urgency | At under 15 hires/year, a free work sample + structured interview outperforms a paid platform test used alone. |
| Ghosting candidates who fail | Feels awkward to deliver bad news | Send a brief, respectful rejection within one week. You are building employer reputation with every candidate. |
The underlying mistake: treating assessments as a replacement for human judgment rather than a supplement to it. The assessment produces data. The hiring manager interprets that data in context. The onboarding process validates whether the assessment predicted correctly. Skip any of these steps and you are using a fraction of the system.
Frequently Asked Questions
What is a pre-hiring assessment?
A pre-hiring assessment is a standardized test or exercise that job candidates complete before you decide whom to hire. Each one targets something tied to performance in the role: cognitive ability, personality traits, job-specific knowledge, hands-on skill, or integrity. It adds objective evidence, comparable across candidates, to what a resume and an interview already tell you. Options run from a free work-sample exercise you write yourself to a paid test hosted on a vendor platform.
What are the three most common types of pre-hiring assessments?
There is no official ranking by usage, but the three types usually listed first are cognitive ability tests (measuring problem-solving and reasoning), personality assessments (measuring work style and behavioral tendencies), and job knowledge or skills tests (measuring technical competence specific to the role). Of these three, job knowledge tests predict job performance best in the current meta-analytic matrix published by Sackett and colleagues (2022), with an operational validity of 0.40. Cognitive ability tests follow at 0.31, and broad personality measures sit lower still, with conscientiousness at 0.19. That ordering reverses the 1998 figures that still circulate, in which cognitive ability led at 0.51. The practical read for a small business is that a test tied to the specific job beats a test of general traits.
Are pre-hiring assessments legal in the US?
Yes. US employers can lawfully test job candidates. The federal anti-discrimination laws apply by size: Title VII and the ADA from 15 employees, the ADEA from 20. Once those laws cover you, the tests must not discriminate, and a test that screens out a group protected by Title VII or the ADA has to be job-related and consistent with business necessity. The Uniform Guidelines on Employee Selection Procedures still set out how adverse impact and validation are judged, even though the EEOC has listed them for rescission, and the statutory standard holds either way. A personality test that works as a medical examination can only be given after a conditional offer. State law can add more: Massachusetts, for example, restricts written honesty tests. The EEOC publishes guidance on testing practices that is worth reading before you choose a test.
How long should a pre-hiring assessment take?
There is no legal or standard length, so treat it as a design choice and keep it proportional to the role. A screening test of 15 to 30 minutes is a sensible target for most roles, and a take-home work sample should come with a firm time limit that fits the seniority of the job. Before sending any assessment, sit it yourself: if it runs longer than you expected or the instructions confuse you, candidates will have the same experience, and some will quit partway. Shorter is generally better for small businesses competing for candidates against faster-moving employers, because every extra step is another point where a strong applicant can drop out.
Can a small business skip the assessment entirely?
Yes. Plenty of small businesses make good hires with structured interviews, reference checks, and simple work samples, and never buy a formal assessment. A formal test starts to pay off when you fill the same role 3 or more times a year, when a posting draws 50 or more applications, or when the job depends on a skill that is hard to judge in an interview, such as coding, data analysis, or operating equipment. If you hire 5 to 10 people a year across different roles, asking every candidate the same questions and scoring the answers against a rubric may get you better results than paying for a per-candidate testing tool.
How much do pre-hiring assessments cost?
It ranges from nothing to whatever a vendor charges, and there is no standard price list. A work sample you write yourself, such as a writing prompt, a spreadsheet exercise, or a role-play scenario, costs only the time it takes to build and score. For tests you buy, the federal Office of Personnel Management (OPM) describes cognitive, personality, integrity, and job knowledge tests as generally cheap to administer, and says an off-the-shelf test usually costs less than one developed from scratch. Formal simulations are the expensive end, because OPM points out they can cost a lot to develop and take a long time to run. Vendors set their own prices, so get a quote from each one at your real candidate volume and compare it with what a free work sample and a structured interview already give you.
What is the difference between a hiring assessment and a background check?
A hiring assessment measures a candidate's ability, personality, or skills to predict job performance. A background check verifies factual information about a candidate's history: criminal records, employment history, education credentials, credit history, and driving records. They serve different purposes at different stages. Assessments happen during the screening phase to narrow the candidate pool. Background checks usually come later, often after a conditional offer, to verify that the selected candidate's claims are accurate. The legal frameworks differ as well: assessments are judged mainly under the federal anti-discrimination laws the EEOC enforces, while a background check run through a screening company also falls under the Fair Credit Reporting Act.
Do pre-hiring assessments actually predict job performance?
The evidence is mixed by assessment type. According to Sackett and colleagues (2022), whose meta-analytic matrix in the Journal of Applied Psychology replaced the 1998 Schmidt and Hunter estimates, structured interviews predict best at an operational validity of 0.42, followed by job knowledge tests at 0.40 and empirically keyed biodata at 0.38. Work samples read 0.33 and cognitive ability tests 0.31, both lower than the 0.54 and 0.51 the 1998 analysis reported, because the authors found a systematic overcorrection for range restriction in the older numbers. Integrity tests sit at 0.31 and conscientiousness at 0.19. Unstructured interviews also fall to 0.19, which is why how you run the interview matters as much as whether you test at all. No single assessment predicts job performance perfectly, and the authors advise building the hiring process from several methods instead of leaning on any single test, partly because combining predictors also narrows group score differences.
What should I do if a candidate fails the assessment?
Document the result and the decision criteria. If the candidate is otherwise strong (strong interview, relevant experience, good references), consider whether the assessment is testing a trainable skill versus an innate ability. A candidate who scores low on a product knowledge test can be trained. A candidate who scores very low on a cognitive ability test for a role requiring complex reasoning presents a different situation. Never share specific assessment scores with candidates. You can say the role has been filled or that you are moving forward with other candidates. On records, EEOC regulations require covered employers to keep every personnel and employment record for one year from the date the record was made or the personnel action was taken, whichever is later, and anything relevant to a filed charge stays until the charge is finally resolved.
How do I know if an assessment vendor is legitimate?
Ask four questions. First, does the vendor provide validation evidence showing the test predicts performance for your specific type of role? Validity data should reference sample sizes, job families, and statistical measures. Second, does the vendor provide adverse impact data showing the test does not disproportionately exclude protected groups? Third, is the test designed by an industrial-organizational psychologist or psychometrician? Fourth, does the vendor offer compliance support (ADA accommodations, documentation for EEOC inquiries)? If the vendor cannot answer these questions clearly, the product is likely a generic personality quiz marketed as a hiring tool, not a validated assessment.
Make HR the easy part of running your business
Onboarding, e-signatures, training and core HR in one place. Up and running in 5 minutes.
- 7-day free trial
- No credit card required
- Cancel anytime