Choose a sales assessment vendor on evidence, not on the demo: ask what the tool measures, what independent research shows it predicts in jobs like yours, how it treats different groups of candidates, and where it sits in your hiring process. In Ontario that last question matters more than most vendors admit, because personality-type tests shouldn't be given before a conditional offer, and in our own process every psychometric test waits until after the offer.
Why is this harder than it looks?
Here's the fair part. Assessment vendors solve a real problem. You're about to spend a lot of money on a sales hire, and a clean report with scores, colours and a recommendation feels like certainty you can buy.
Some of these tools are well built. The trouble is that from the outside, a well-built tool and a badly built one look the same. Both have a slick dashboard, a few customer logos and a validity number on the website.
What validity evidence should you ask for?
Ask who did the research. This is the single most useful question. In a large review of integrity test research, studies written by test publishers reported a corrected validity for job performance of .27, while studies by independent researchers reported .12 (Van Iddekinge et al., 2012). That's one kind of test, but the lesson travels: a vendor's own studies are where the numbers run high.
Ask what the number is compared to. The most recent large review of hiring methods cut the estimates for the top-ranked methods by .10 to .20 and put structured interviews at the top, at .42, with job knowledge tests at .40 and cognitive ability at .31 (Sackett et al., 2022). If a vendor quotes a single tool at .50 or higher, ask where that figure comes from and how old it is.
Ask whether it was tested on jobs like yours. A tool validated on call-centre staff tells you little about a technical salesperson selling to utilities. Ask for studies in sales roles, at your level, with real performance outcomes rather than manager opinion alone.
Ask for the technical manual. The professional standards for selection procedures, the SIOP Principles for the Validation and Use of Personnel Selection Procedures (Society for Industrial and Organizational Psychology [SIOP], 2018), set out the kind of evidence a selection tool should have. A serious vendor will have a manual written against standards like these. If they won't share one, that's your answer.
What should you ask about fairness and AI?
More assessment products now score candidates with algorithms trained on past hiring data. That can carry old patterns forward.
A study of vendors' public claims about algorithmic hiring found that data-driven assessments run "the risk of codifying inequalities" while offering "a veneer of objectivity" (Raghavan et al., 2020). A systematic review of 36 studies on algorithmic decisions in HR put it simply: "any predictive algorithmic decision-making tool built on historical data may inherit historical biases" (Köchling & Wehner, 2020).
Large language models are not exempt. In one study, language models asked to write hiring-decision emails favoured White-associated names over Hispanic-associated names in many settings, with results that shifted depending on how the prompt was written (An et al., 2024).
So ask:
Does the tool use AI or machine learning to score candidates? On what data was it trained?
What adverse impact testing has been done, by whom, and on which groups?
Can a person review and override every score?
There's an Ontario angle here too. Since January 1, 2026, employers with 25 or more employees must say in publicly advertised job postings whether they use artificial intelligence to screen, assess or select applicants (Osler, Hoskin & Harcourt LLP, 2026). If your vendor's tool uses AI, your posting may need to say so. Confirm the details with employment counsel.
What should you ask about timing and accessibility?
Ask where the tool belongs in your process. The Ontario Human Rights Commission's guidance says tests that assess "psychological or personality profiles" of applicants "should never be administered before a conditional offer of employment, and even then should be approached with caution" (Ontario Human Rights Commission [OHRC], 2008, section IV-6). Plenty of vendors sell their tools as an early screen. In Ontario, a personality-type tool used that way runs against the Commission's guidance. Our post on integrity tests covers one common example.
In our own process, personality testing comes after the offer, and we use it for coaching and onboarding, not to decide who gets hired. Our cognitive testing comes after the offer too.
Ask about accessibility. Ontario businesses with 50 or more employees must make their public websites and web content posted after January 1, 2012 meet WCAG 2.0 Level AA (with two exceptions), and the rule covers websites an organization controls directly or through a contractual relationship (Government of Ontario, 2026). If the vendor's candidate portal falls under that, ask them to show conformance. Separately, Ontario employers must tell applicants selected for an assessment that accommodation is available on request (Integrated Accessibility Standards, O. Reg. 191/11, s. 23).
A short checklist for any assessment provider
What exactly does it measure, and how does that link to the job?
Independent validity studies in sales roles, with real outcomes.
A technical manual you can read.
Adverse impact results, and whether AI is used to score.
Where in the process the vendor expects it to be used, and whether that fits Ontario guidance.
Accessibility and accommodation.
What the report tells you, in plain language. Our guide to how to read a sales assessment report covers what to look for.
Before the offer, the evidence that predicts best is still a structured interview, a realistic work sample such as a role play, and good references.
Two things that are both true
We'll own this one for HR: we've bought tools because the demo was impressive and the report looked like rigour. A good-looking report isn't the same as a valid one.
The right assessment, used at the right time, can genuinely help a new salesperson and their manager. And the wrong one, or the right one used too early, adds risk instead of removing it. For owners in Guelph and across Ontario, the questions above take an hour and save a lot more.
In short
Choose a sales assessment vendor on independent evidence, not on the demo. Publisher-written studies of integrity tests reported more than double the validity of independent ones, .27 against .12 (Van Iddekinge et al., 2012), and the top validity estimates were revised down in 2022, with structured interviews on top at .42 (Sackett et al., 2022). Ask about AI scoring and bias (Köchling & Wehner, 2020; Raghavan et al., 2020), accessibility, and timing: in Ontario, personality-type tests should never be given before a conditional offer (OHRC, 2008).
Key takeaways
Publisher-written integrity test studies reported a corrected validity of .27 against .12 for independent studies (Van Iddekinge et al., 2012).
Structured interviews (.42) and job knowledge tests (.40) top the revised estimates, with cognitive ability fifth at .31 (Sackett et al., 2022).
Algorithmic tools built on historical data "may inherit historical biases" (Köchling & Wehner, 2020).
Since January 1, 2026, Ontario employers with 25 or more employees must disclose AI use in publicly advertised job postings (Osler, Hoskin & Harcourt LLP, 2026).
Ontario's Human Rights Commission says personality-type tests should never be given before a conditional offer (OHRC, 2008).
Frequently asked questions
What should I ask a sales assessment vendor?
Ask what it measures, for independent validity studies in sales roles, for the technical manual, for adverse impact results, whether AI is used to score, and where in the hiring process the vendor expects it to be used.
How do I know if a sales assessment is valid?
Look for independent studies with real job outcomes in roles like yours, not only the vendor's own research. Publisher-written studies tend to report higher validity (Van Iddekinge et al., 2012).
Can I use an assessment to screen sales candidates early in Ontario?
Not in our process. The Ontario Human Rights Commission says personality-type tests should never be given before a conditional offer (OHRC, 2008), and we keep all psychometric testing, cognitive included, until after the offer. This is general information, not legal advice; confirm with employment counsel.
Do I have to tell candidates if an assessment uses AI?
Since January 1, 2026, Ontario employers with 25 or more employees must disclose in publicly advertised job postings whether they use AI to screen, assess or select applicants (Osler, Hoskin & Harcourt LLP, 2026). Confirm the details with employment counsel.
Work with us
Down to your last few candidates? Our Final Candidate Review tests your finalists against your own best people before you make the offer. If it doesn't help, you don't pay.
This post is general information, not legal advice. Confirm employment decisions with employment counsel.
References
An, H., Acquaye, C., Wang, C., Li, Z., & Rudinger, R. (2024). Do large language models discriminate in hiring decisions on the basis of race, ethnicity, and gender? In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) (pp. 386-397). https://doi.org/10.18653/v1/2024.acl-short.37
Government of Ontario. (2026, September 21). How to make websites accessible. https://www.ontario.ca/page/how-make-websites-accessible
Integrated Accessibility Standards, O. Reg. 191/11, s. 23. https://www.ontario.ca/laws/regulation/110191
Köchling, A., & Wehner, M. C. (2020). Discriminated by an algorithm: A systematic review of discrimination and fairness by algorithmic decision-making in the context of HR recruitment and HR development. Business Research, 13, 795-848. https://doi.org/10.1007/s40685-020-00134-w
Ontario Human Rights Commission. (2008). Human rights at work 2008 (3rd ed.), section IV-6: Requesting job-related sensitive information. https://www.ohrc.on.ca/en/iv-human-rights-issues-all-stages-employment/6-requesting-job-related-sensitive-information
Osler, Hoskin & Harcourt LLP. (2026, February 3). AI in hiring: Ontario employers grappling with new job posting disclosure requirement. https://www.osler.com/en/insights/blogs/employment-and-labour-law-blog/ai-in-hiring-ontario-employers-grappling-with-new-job-posting-disclosure-requirement/
Raghavan, M., Barocas, S., Kleinberg, J., & Levy, K. (2020). Mitigating bias in algorithmic hiring: Evaluating claims and practices. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency (pp. 469-481). https://doi.org/10.1145/3351095.3372828
Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022). Revisiting meta-analytic estimates of validity in personnel selection: Addressing systematic overcorrection for restriction of range. Journal of Applied Psychology, 107(11), 2040-2068. https://doi.org/10.1037/apl0000994
Society for Industrial and Organizational Psychology. (2018). Principles for the validation and use of personnel selection procedures (5th ed.). Industrial and Organizational Psychology, 11(S1), 1-97. https://doi.org/10.1017/iop.2018.195
Van Iddekinge, C. H., Roth, P. L., Raymark, P. H., & Odle-Dusseau, H. N. (2012). The criterion-related validity of integrity tests: An updated meta-analysis. Journal of Applied Psychology, 97(3), 499-530. https://doi.org/10.1037/a0021196
