Resources & guides

Vetiu editorial · 3 min read

Why support knowledge gaps are hard to see, and how to check for them

Completion, confidence and correct decisions are different signals. Use realistic quizzes and search feedback to find where support is needed.

Illustration of topic results revealing areas for further learning

Silence is not proof of understanding

An agent who stops asking questions may understand the policy, or they may have settled on an incorrect interpretation. A completed course shows that someone worked through the learning path. Confidence shows how they feel. Neither, on its own, proves that they will make the right decision in an unfamiliar case.

Knowledge gaps often live in exceptions: a damaged item outside the normal return window, an account recovery request with incomplete verification or a refund requiring approval. These situations may appear too rarely for a manager to spot a pattern through observation alone.

Ask about the decision, not the wording

A useful quiz question describes a situation an agent could actually face and asks for the next action. The correct answer must be supported by an approved article. Wrong answers should reflect plausible misunderstandings, not trick wording or obviously absurd choices.

Keep checks focused on a small number of important decisions. If a question is ambiguous or the policy is incomplete, the result tells you about the content as much as the person. Review those possibilities before treating a low score as a coaching issue.

Group results by a consistent topic

An overall quiz score can hide a weak subject. A person might answer most product questions correctly and still misunderstand every refund exception. Consistent topic tags make it possible to see that distinction across people and over time.

In Vetiu, knowledge-gap reporting uses results from separately assigned quizzes and check-ups, grouped by question topic. Course quizzes remain part of course progress instead. This separates an immediate learning check from later evidence of retention.

Read the evidence with context

Vetiu’s question statistics also let managers review accuracy and answer counts for the latest 100 answers and sort questions by accuracy. That helps prioritise investigation; it does not establish why an answer was wrong.

  • Check participation: results from a few people do not necessarily represent the whole team.
  • Inspect weak questions: a confusing answer option can lower accuracy even when the policy is understood.
  • Review the source: if agents cannot find the rule in the article, improve the article first.
  • Compare like with like: use comparable audiences, topics and periods when reviewing a trend.

Close the loop with content and coaching

A weak topic across many people may call for a clearer article and a focused team update. A gap affecting one person may call for individual coaching. Search-related issues in AI Insights provide another signal: missing or unclear content may prevent agents from finding a reliable answer at all.

After making a change, assign a focused follow-up check-up and compare the next results. Quizzes cannot uncover every skill or behaviour, so combine them with ticket review and observation. The goal is to make uncertainty visible early enough to help, not to label people by a score.

Put your support knowledge to work

Related reading