Perplexity

Based on real interviews

Launch a trust feature for an answer engine

Choose how an answer engine should help readers check whether its citations support its claims.

Unlock with Full Access

The challenge

An answer engine already shows citations, but users struggle to find the passage supporting a particular claim. A claim-to-source aligner could improve the experience, although it misses some unsupported claims and adds latency and cost.

Choose one focused trust feature for a six-week build. Decide which users and queries to start with, what they see when support is incomplete, and how the team will measure whether the feature improves understanding without suggesting more certainty than the evidence supports.

What you'll practise

  • Product scope

    Choose the initial segment and the citation experience that addresses its problem.

  • Quality measurement

    Distinguish direct evidence of answer quality from clicks or reformulations.

  • Launch economics

    Use the supplied cost and latency constraints to defend the rollout.

Your writing workspacePreview
Your trust-feature launch memoYour response
Your recommendation

Write your reasoning and submit it for feedback.

The full brief and feedback are included with Full Access.