Not All Assessments Are Created Equal
The assessment industry is largely unregulated. Any company can build a tool and sell it as a measure of personality, potential, or leadership fit. Often, assessments are bundled “free” with technology subscriptions. That makes selecting the right tool harder than it should be, and choosing one that lacks a foundation in personality psychology costs more than most organizations realize.
Selecting the right assessment can make or break any talent initiative. The science behind a tool, or lack thereof, has significant consequences.
Validity: Does it measure what it claims to?
Validity refers to the accuracy of an assessment. Basically, does it measure what it says it measures? When using an assessment for talent management initiatives, focus on predictive validity.
An assessment is valid if it predicts the outcomes it claims to predict: job performance, leadership effectiveness, retention, etc. Unfortunately, many tools on the market do not measure what they claim to measure.
A valid assessment has been tested against external criteria: job performance, peer ratings, leadership outcomes, in real organizations with real employees. That research tells you two things: how a person is likely to be perceived by others in the workplace, and how their scores compare to a broader population in ways that forecast behavior at work. Without that evidence, you are buying a description, not a prediction.
A valid assessment has documented evidence. This is not the same as marketing copy or testimonials, but research conducted with real people in real organizations demonstrates that the tool measures what they claim to measure. That evidence should be available from any provider worth using.
Reliability: Does it produce consistent results?
Reliability is consistency. A reliable assessment yields essentially the same result for the same person when they take it again.
Personality is relatively stable. How someone is at their core, how they handle pressure, how they relate to others, and how they approach problems do not change significantly from one month to the next. An assessment that yields different results for the same person over a short period does not measure personality.
This is the most visible failure point of type-based tools. Roughly half of the people who receive fixed categories: letters, colors, or archetypes, often change when retested just five weeks later. That inconsistency is a fundamental measurement problem that makes the results unreliable for any decision that matters.
What a well-built assessment does.
Proper assessment development is a rigorous process. It requires professionals with genuine backgrounds in psychology and research methods, tools built on validated science, and ongoing monitoring to ensure the instrument continues to measure what it is supposed to measure.
The most rigorously validated framework for doing this is the Five-Factor Model, which has more than a century of independent research behind it. Tools built on this foundation have a track record that newer instruments do not.
Beyond basic personality, the best assessments also surface behavioral tendencies that stay hidden in stable conditions but emerge when a person stops self-managing. These patterns derail capable people at exactly the wrong moment. They do not show up in interviews or performance reviews until after they have already caused problems.
A good assessment also produces output that answers a business question. Not a personality summary to file away, but insight you can act on: Is this person likely to succeed in this role? Where are the risks? What does development need to address?
Compliance is not optional.
Because the assessment industry is unregulated, publishers are not required to provide information regarding their compliance with employment guidelines. However, reputable assessment publishers provide information regarding the compliance of their tools with international guidelines. In the United States, employment assessments (traditional, AI-based, or software) must comply with guidelines established by the Equal Employment Opportunity Commission (EEOC).
Publishers that cannot demonstrate compliance are exposing their clients to legal risk.
Technical manuals are not marketing materials.
One of the most reliable ways to evaluate an assessment is to review the publisher's support materials. This means technical manuals and validation studies with documentation of the scientific processes used and the results of that research. This is distinct from a brochure, a case study, or marketing materials.
Be cautious of marketing claims, particularly around AI and other new technologies. AI is attractive to investors, and many companies use the term without applying it in any meaningful way. The claim is not the evidence.
Publishers who are unwilling to share technical data may lack evidence to support their claims.
What to ask before using any assessment.
Before deploying a tool, get clear answers to the following:
Does this assessment predict performance, and where is the evidence?
Has it been validated against external criteria, such as job performance and peer ratings in real organizations?
Does it produce consistent results over time?
Has it been reviewed by an independent body or appeared in peer-reviewed research?
Does it measure risk and derailers, or only strengths?
Does it comply with EEOC guidelines and the Uniform Guidelines on Employee Selection Procedures?
Are technical manuals and validation studies available?
The bottom line.
The best-known assessments are not always the best assessments. Familiarity and validity are different things, and in an unregulated market, tools spread because they are easy to market, not because they work.
At INDx, every engagement is built on assessments with documented validity, consistent measurement, and output designed to drive decisions. The science exists to improve talent decisions.
If you want to think through your current assessment approach, we are happy to have that conversation with you.
Contact us:hello@indxtalent.com