Analytics
Problems intel
Most-used problems, difficulty distribution, and tag frequency across your workspace's assessments.
Problems intel
The Problems intel view is a read-only summary of the problems your workspace uses. It is useful for spotting skew — one problem doing most of the work, or one tag dominating — and for picking the next problem to add to your shared library.
What it shows
| Section | What it is |
|---|---|
| Total unique problems | The number of distinct problems in use across your assessments |
| Traditional vs variation | A split between catalogue problems and AI-generated variations |
| Difficulty distribution | The split across Easy, Medium, Hard |
| Tag frequency | The most common problem tags, ranked by count |
| Top problems | The most-used problems, with the assessment count for each |
How to read it
A few patterns are worth knowing:
- One problem dominates. That is fine if it is intentional, but worth a look if it happened by accident. A single problem doing most of the work tells you nothing about how candidates compare across problems.
- Tag skew. If one tag is doing 60% of the work, your assessments are probably testing one skill. Add a complementary tag to broaden coverage.
- Variations are working. A high variation count means your team is using AI variations to keep screens fresh, which is a good signal.