Search & Discovery Features

Search people can trust and browsing that makes sense: query log baselines, quality metrics and judgement lists, weekly zero-result reviews, relevance tuning, facets, autocomplete and the engine decisions behind them.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Then open the Marketplace in the app, search for Search & Discovery Features, and click Install. It goes into your own files, and works offline from then on.

What it installs.

1

Pillar

1

Area

50

Projects

200

Starter tasks

🧩 Product Management

Search & Discovery Features

Improving how users find content, products or answers inside the product through search, filters, sorting and browsing paths.

  1. Search and discovery audit of the current product
  2. Top queries and zero-result baseline
  3. Findability task study with eight users
  4. Search ownership and decision rights
  5. Catalogue and content metadata quality inventory
  6. Search quality metric definitions
  7. Search event tracking for queries, impressions and clicks
  8. Search engine approach decision
  9. Graded judgement list of 200 queries
  10. Weekly zero-result query review
  11. Synonym and spelling dictionary upkeep
  12. Monthly relevance tuning cycle
  13. Search quality dashboard and monthly report
  14. Pinned results and merchandising rules governance
  15. Index freshness and sync monitoring
  16. Search feedback triage
  17. Offline relevance regression suite
  18. Quarterly facet and category taxonomy review
  19. Information retrieval fundamentals for the product team
  20. Relevance judging calibration workshop
  21. Reading ranking metrics like NDCG, MRR and precision
  22. Query understanding basics
  23. Search and browse interface pattern study
  24. Semantic and vector search primer
  25. Online ranking experiments and interleaving
  26. Autocomplete and query suggestions redesign
  27. Facets and filters for the main results page
  28. Default sort order and sort options decision
  29. Zero-result page that recovers the search
  30. Typo tolerance and fuzzy matching settings
  31. Hybrid keyword and semantic retrieval trial
  32. Browse navigation restructure with card sorting and tree testing
  33. Personalised ranking signals decision
  34. Search latency budget for results and autocomplete
  35. Search engine migration cutover
  36. Full reindex and schema change rehearsal
  37. Peak season search readiness
  38. Search relaunch support briefing and help content
  39. Adding a new content type to search
  40. Ecommerce product search for a catalogue team
  41. Permission-aware search in a B2B app
  42. Help centre search for support deflection
  43. Minimum workable search for a small team
  44. Search on-call runbook for engineers
  45. More like this and related items on detail pages
  46. Learning to rank from click data
  47. Query intent classification and routing
  48. Multilingual search analysers and per-language relevance
  49. Search log privacy and retention rules
  50. Federated search across several content sources

Where it starts.

Each project arrives with its purpose, milestones and first tasks written out. Here is how the first three read.

Search and discovery audit of the current product

This is a team plan, best installed into a shared workspace so everyone with a role in it can see and own their tasks. Before changing ranking or adding filters, someone needs to write down how people actually find things in the product today: the search box, category pages, filters, sort menus, saved views and links from elsewhere.

First tasks

  • Walk every screen and list each place a user can search, filter, sort or browse
  • Ask engineering which engine or query powers each search box
  • Pull the share of sessions that use search versus category browsing
  • Write up the top ten findability problems with a screenshot of each

Top queries and zero-result baseline

Your search log is the most honest record of what users want and cannot find, typed in their own words. Exporting ninety days of queries and sorting them by volume shows the short head that drives most traffic, the long tail, and the share of searches that return nothing. That zero-result rate and the top 100 queries become the numbers every later change is judged against.

First tasks

  • Ask engineering where search queries are logged and how far back they go
  • Export ninety days of queries with result counts and clicks
  • Group the 50 most common zero-result queries by cause: typo, synonym, missing content
  • Record the zero-result rate and top 100 queries in a dated baseline sheet

Findability task study with eight users

Logs show what people typed, not what they gave up on before typing. Watching eight users try ten realistic find tasks, half by search and half by browsing, reveals vocabulary mismatches, filters people ignore and categories that make sense only to the team that named them. Run it early so the backlog is built on observed failure rather than opinion.

First tasks

  • Pick ten find tasks from recent support tickets and top queries
  • Recruit eight users across at least two customer segments
  • Run each session and note success, time and the words users tried
  • Repeat the study with fresh users to measure progress @recurring(yearly)

Make it yours.

Install Search & Discovery Features, then tell Aurum the date you're aiming for. It checks whether the plan fits the weeks you actually have.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Blueprints are general information to help you organise your own work. They are not medical, financial, legal or other professional advice. See the Terms.