Most exam-prep apps describe their engine in language that means nothing you can verify: "adaptive," "AI-powered," "personalized." Here's what LicensePath's diagnostic actually does, described specifically enough that you could disagree with a design choice if you wanted to.
1. Every question is tagged to exactly one real blueprint category
Before any diagnostic logic runs, every practice question in LicensePath is assigned exactly one category id, matched directly to a category on your state's real exam content outline (the one published by the state regulator and the testing vendor, Pearson VUE or PSI, depending on your state). Texas's General Lines exam, for example, breaks into 13 categories with fixed weights: Life: Types of Policies at 15%, Health: Policy Provisions at 15%, Texas-specific rules for Life & Health at 14%, and so on down to the smallest category. That's not our invention; it's Pearson VUE's own published content outline for that exam.
This tagging is the foundation everything else depends on. If a question is miscategorized, the diagnostic reports a false picture of your readiness in that area. That's exactly why our internal review process treats a wrong category assignment as seriously as a wrong answer key (see why we cite every practice question).
2. The diagnostic maps coverage, not just a single score
A single overall percentage tells you almost nothing useful for exam prep, because it can hide a category you're solid on canceling out one you're genuinely weak in. LicensePath's diagnostic instead produces a per-category readiness picture: for every blueprint category on your state's exam, it tracks how many questions you've seen, how many you got right, and how recently.
Coverage
Have you seen enough questions in this category to say anything meaningful yet, or is it still an unknown?
Accuracy
The plain fraction of the ones you've seen that you actually got right, with no adjustment or curve applied.
Recency
A category you nailed three weeks ago and haven't touched since isn't the same as one you nailed yesterday.
3. Practice sessions weight toward your real exam's real weights
Daily practice doesn't pull questions uniformly at random, and it doesn't just chase your weakest category to the exclusion of everything else. It weights selection by two things at once: how weak you currently are in a category, and how much that category actually counts on your specific exam. A category worth 3% of your exam that you're struggling with gets attention, but not at the expense of a category worth 15% that you're only moderately behind on. On exam day, the 15% category is worth five times as much, regardless of how it feels to study.
This is the direct, practical payoff of the blueprint-mapping step above: without knowing the real weight of each category, there's no honest way to make this trade-off at all. An app that doesn't map to your state's real outline is guessing.
4. A missed question doesn't disappear once you get it right once
This is the part most exam-prep tools skip entirely, and it's the one we think matters most. Getting a previously-missed question right on your very next attempt could mean you learned it. It could also mean you got lucky on a guess, or half-remembered the specific wording rather than the underlying rule. LicensePath's practice-history model doesn't collapse those into the same outcome. A question you've missed moves through real states as you keep practicing:
| State | What it means |
|---|---|
| Missed once | You got it wrong the one time you've seen it since your last correct answer (or ever). |
| Repeated miss | You've missed it more than once with no correct answer in between. This is the strongest signal that it's a real gap rather than a slip. |
| In review | You got it right once since your last miss. That's promising, but one correct answer isn't proof, so it stays flagged. |
| Mastered | Two consecutive correct answers since your last miss. Only then does it stop being treated as a weak spot. |
Mastery isn't permanent immunity either. If a "mastered" question comes up again later and you miss it, it reopens as a fresh missed-once rather than a repeated-miss, because forgetting something you once knew is a different, more common situation than never having learned it. Treating it as harshly as a repeated miss would overstate how far behind you actually are.
Practice's "review missed questions" mode defaults to surfacing missed-once and repeated-miss items specifically, the two states that still represent real, unresolved gaps, while letting you filter to any specific state if you want to spot-check something you think you've already got.
5. The mock exam is the check on all of it
Everything above is practice: lower stakes, immediate feedback, explicitly designed to expose and close gaps. The full-length timed mock exam works differently by design, matching your state's real question count, real time limit, and real category proportions, with no hints and no explanations until it's over. It exists specifically so the diagnostic's read on your readiness gets tested against something that behaves like exam day, not just against itself.
See it against your own state's blueprint.
Join the launch list and pick your state. We'll tell you the moment it's open.