Search Basics
Go to Cases and use the field labelled Search cases above the list. Three ways to find something:- Keyword search – full-text matching, on every plan
- Semantic search – meaning-based search, Team plan and above
- Tag filter – click a tag under Popular tags:
Keyword Search
Type words or phrases from case titles, observations, root causes and resolutions. Examples:bearing failure→ returns all cases mentioning “bearing” and “failure”thermal imaging→ cases discussing thermal methodsMarch 2026 incident→ cases mentioning specific dates or incidents
Semantic Search
Semantic search understands meaning, not just keywords. Describe what you’re looking for in natural language, and FTS returns cases with similar concepts. Why it’s powerful:
Semantic search works by converting your query into a vector (a mathematical representation of meaning) and comparing it to all cases in your workspace. It’s forgiving of phrasing and finds conceptually similar cases even if they use different terminology.
Semantic search is part of the Team plan and above; on Starter, search is keyword-only. There is no per-workspace switch for it — the plan decides. No case data is used to train AI models.
Tag-Based Filtering
If your workspace uses a taxonomy, you can filter by tags without typing:- Click one of the tags under Popular tags: above the case list
- The list narrows to cases carrying that tag; Clear tag filter removes it again
Advanced Search Operators
What the search box actually does: it looks for your words in the case title, the observation, the root cause and the resolution — and in nothing else. The text that automatic text recognition read out of your evidence files is not part of the search index, on any plan. Results are ordered by the Sort order drop-down described below; the state of a case does not move it up or down. An earlier version of this page said Published and Verified cases were shown first — they are not.Saved Searches
Search Results
Results display:- Case Title – Click to open
- Preview – the opening of the observation text, cut off after two lines. It is a visual cut, not a fixed number of characters, so how much you see depends on the width of your window.
- Tags – Visual badges
- State – Draft, Verified, Published, etc. Click any result to open the full case.
No relevance percentage is shown. Sorting is by the criteria above, not by a match score, and there is no keyboard navigation between results.
Search Tips & Best Practices
Use natural language in semantic search
Use natural language in semantic search
Instead of
motor overheat, try what should I do if the motor gets too hot?Combine search methods
Combine search methods
Start with semantic search to find related cases, then filter by tag to focus on your department.
Filenames are NOT searchable
Filenames are NOT searchable
An earlier version of this page recommended meaningful filenames “because filenames are what search reads”. That is not true: the search index holds the case title, observation, root cause and resolution — the filename is in none of them, and neither is the text extracted by OCR. A clear filename helps you read the evidence list; it does not help you find the case. If you want to find a case by one of its files, name the file in the case text — that is what search reads. Captions cannot be entered anywhere in the product today (open item #98).
Read the Case History
Read the Case History
It lists what happened and when — creation, edits, state changes, reviews and evidence uploads — for the most recent 200 entries. It does not repeat the case text, and it does not repeat a reviewer’s comment: comments live in the Reviews panel below it. Search matches the case text, not the history entries.
Search Privacy
Semantic search queries are processed immediately and not retained after the search completes, and no queries or results are used to train AI models. There is no per-workspace AI switch — on the Starter plan semantic search is simply not part of the plan.Taxonomy
Learn how tags and categories are organized.
AI Assistance
Discover other AI-powered features in FTS.