> For the complete documentation index, see [llms.txt](https://docs.tickertrends.io/tickertrends/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.tickertrends.io/tickertrends/documentation-v2/4.-features-and-workflows/ai-search.md). # AI Search Use natural-language search to start investment research workflows. ### Search investment research workflows **AI Search** lets users search TickerTrends in plain English. Instead of starting from a ticker or manually setting filters, users can type a natural-language question and let TickerTrends route them to the most relevant page, screener, or dataset. Use it to start a research workflow from a ticker, term, or question. ### How it works The search bar accepts: * ticker and company searches * keyword and term searches * natural-language questions The router may direct the request to a relevant destination. Review the resulting scope and filters. ### Example queries Users can ask questions such as: * `what fashion brands are trending on TikTok?` * `viral food` * `what's exploding on TikTok right now?` * `find private AI companies with accelerating consumer interest` * `how is Nvidia trending on social media` These prompts help users discover: * available categories * relevant entities * changing keywords * the right screener tab for deeper work ### What the router does Depending on the query, TickerTrends may route users to: * **Exploding Trends** for keyword discovery * **Trackers** for directional company indicators * **KPI Tracking** for KPI estimates, revisions, and consensus context * **Data Source KPIs** for source-level signal analysis * **Catalysts** for catalyst-oriented research * **Upcoming Earnings** for near-term earnings research * company pages when the query is ticker-specific * a relevant screener mode with the **Private** universe selected for private-company research Results and routing may vary by query, coverage, and entitlement.
### Saved screener views Save a screener view to return to the same research workflow later. Saved views retain: * the active screener mode * the selected universe or scope * applied filters * the search query * sorting * selected data type Open a saved view directly from a query to restore its saved configuration. Availability depends on the user's access level. For more on screeners, see [Unified Screener](/tickertrends/documentation-v2/3.-unified-screener-kpi-forecasts-trackers-data-source-kpis-exploding-trends-upcoming-earnings.md). ### Search suggestions and recent results When users click into the search bar, TickerTrends can surface: * recent tickers * recent private companies * direct company matches * natural-language guidance for what users can ask This makes the search bar useful both as a navigation tool and as a discovery surface. ### Best practices AI Search works best when the prompt is phrased like a search goal. Good examples: * `consumer companies with accelerating awareness` * `what fashion brands are trending on TikTok?` * `viral food` Less useful prompts are overly long or conversational. Short, direct intent works best. ### Why it matters AI Search makes TickerTrends easier to use when users do not yet know: * which ticker to start with * which dataset to search * which screener tab fits the question AI Search is a navigation and research aid. It does not validate a conclusion or recommendation.