Here is the problem with every argument about which personality assessment is "more accurate": nobody has the answer key. You can compare two DISC results for a real person, but you cannot check either one against the truth, because the truth about a real person is exactly the thing in dispute. So a few weeks ago I did the only thing that gets around that. I wrote the truth first.
Meet Morgan Reed
Morgan does not exist. Before either assessment was opened, I wrote a fixed behavioral profile for a fictional person: highly conscientious, meaningfully steady, socially reserved, evidence-driven, supportive without being passive, and willing to become direct when the facts justify it. In DISC terms, that is a high C with a real S, a low I, and a lower D that shows up situationally rather than as a default.
Two details matter later. Morgan accepts minor imperfection when a deadline requires it, and Morgan supports change when the rationale is sound. Neither is typical of the cartoon version of a high C, and I put both in deliberately.
The rules
Then I answered two very different DISC experiences exactly as Morgan would.
- The benchmark was locked first. Morgan existed before either result.
- Same simulated person for every answer. Each choice was made against the same written behavioral rules.
- No retakes. Results are recorded as they occur, including the inconvenient ones.
- Each system is judged as designed. The traditional test is built to finish in minutes. Ours is built to be longitudinal, so it gets five story assessments before the full comparison.
The traditional side is a free 24-item DISC Snapshot from Assessments 24x7. The story side is The Assessment Library. As I write this, the Snapshot is complete and two of the five planned story assessments are finished, so everything below is interim, and I will say so more than once.
Where a questionnaire runs out of room
Both instruments landed in the same C/S neighborhood, which is the minimum you would hope for. The difference is in what happened past that point. Asked to describe Morgan from 24 forced choices, the questionnaire generalized: it leaned Morgan toward perfectionism and toward relying on old methods, two things I had specifically written out of the personality. Remember the details: Morgan tolerates imperfection under deadline pressure and supports well-reasoned change. The stories never made that mistake, because they never asked Morgan to pick an adjective. They watched Morgan let a good-enough plan ship and get behind a change with a sound rationale. When the person is written down in advance, that is the kind of precision you can actually check, and after two stories the story-based profile is the one that has it.
What two stories found
After two completed story assessments, The Assessment Library reads Morgan as C 60%, S 25%, D 10%, I 5%, labeled a C + S blend. The dashboard also shows 4% confidence, which sounds alarming until you know what it measures. Our Confidence Meter tracks how much evidence sits behind a profile, not how accurate the profile is. It reaches 100% at fifty completed assessments, so two stories read as 4% by design.
What I did not expect after only two stories was the shape. High C, meaningful S, low I, lower D. That is the relative order I wrote into Morgan before testing, and the proportions are sensible rather than lopsided. The one trait still unproven is the situational assertiveness: Morgan will challenge a bad plan when the evidence is strong, and the current written summary has not expressed that yet. Stories three through five are where it has to show up, or the story method has missed something the benchmark says is there.
The bigger difference was not the numbers. It was the context. The Snapshot asked Morgan to choose between descriptors. The stories put Morgan inside situations and recorded what Morgan did when the situation changed:
- An uncertain restructuring: gather legitimate facts before acting.
- The evidence becomes strong: present the case instead of staying passive.
- A coworker needs support: listen and help practically without breaking a confidence.
- An employee makes a costly mistake: stay calm, fix the problem, skip the unnecessary confrontation.
- A plan is good enough: stop analyzing, assign ownership, execute.
- A personal commitment matters: build it into the plan instead of hoping time appears.
A story-based assessment can watch the same underlying traits express themselves differently as the problem changes. A questionnaire has to ask you to describe those traits directly. That is the whole argument for the format, and it is the part of this experiment I already feel most confident about.
The interim scorecard
I rated both experiences across seven categories, with personality accuracy carrying half the weight, because everything else matters less if an assessment describes the person incorrectly. The interim weighted score is The Assessment Library 8.8 out of 10, Snapshot 7.9. The full table, category by category, lives on the case-study page, which is the version I will keep updating.
After two stories, The Assessment Library leads on personality accuracy, on nuance and contextual understanding, on engagement, on the naturalness of the questions, on resistance to obvious gaming, and on information gained per decision, where it rates 9.2 to the questionnaire's 7.2. That last row is the one I would point a skeptic to. A story asks for more of your time, and it spends that time well: one story choice can reveal planning, empathy, delegation, assertiveness, follow-through, and whether the person knows when to stop analyzing, all at once. Raw speed and information efficiency are different things, and stories are built for the second one.
Keep the word interim attached to that 8.8. It reflects two of five planned stories, it will be recalculated from scratch after the fifth, and it could move in either direction.
The answer does not look like the trait
One more difference showed up that I had not set out to test. In an adjective format, anyone who knows DISC can see the answer key: aggressive is D, outgoing is I, steady is S, accurate is C. In the stories, Morgan had to choose an action, not a label, and the action often carried more than one trait. On top of that, the answer positions were shuffled when I reopened an assessment, so there was no stable A/B/C/D pattern to lean on.
I want to be careful here. This does not make our assessments impossible to game, and this experiment did not measure formal faking resistance. What I can say is that in this test, the intended DISC signal was less obvious, because the respondent was choosing behavior inside a situation rather than recognizing a word.
Why I keep going to five
The temptation after two favorable stories is to publish the victory lap. I am not going to, for a reason built into the product: the Compare tool, which places a traditional DISC report next to a story-revealed profile, does not unlock until a person has finished five assessments, because the revealed side needs real evidence behind it. The experiment follows the same rule. Assessment five is the checkpoint where I will re-rate every category, run Morgan's Snapshot through the Compare tool, and update the numbers.
Here is what would change my mind. If the stories never surface Morgan's situational assertiveness, the story method missed a trait the benchmark says is there, and I will write that down. If they do, the extra time will have bought a more precise person than a questionnaire can describe in four minutes, and I will write that down too.
Follow along, or run your own
The full interim case study has every table, the situations Morgan faced, and the matched-versus-still-to-prove checklist against the benchmark. It will update when the five-assessment sequence is complete. And if you would rather test the method on the one person you cannot fake, your first story assessment is free. Read it, choose, and see what your choices say.
One last disclosure, because credibility is the whole point: Morgan is fictional, this is one controlled case study, and it is not a scientific validation of either instrument. It is a careful, documented, repeatable look at what two very different ways of asking the same question found when the answer was written down in advance.
