Tell Sleuth what you’re deciding. It works through every program, deal, and document, then builds the interactive analysis, deck, or spreadsheet, with a source on every number.
Sleuth builds what the question calls for: a partner map you can click through, slides for the meeting, or every row in a spreadsheet, all from the same knowledge base. Every number can be traced to its source, and you can hand the whole project to Claude or your team.
Click any company or program to see its row and its sources.
Slides for the meeting, built from the same rows.
Every row, plus the working data behind the analysis.
One of our own projects, three ways: our Q3 deep dive on T-cell engagers.
Every plan includes the knowledge base and documents Sleuth works from, so you don’t need a separate database or document search seat. Unlimited searching and exploring on all plans.
Everything your database has, connected. Browse and filter drugs, trials, deals, companies, and targets, the same records Sleuth uses. Browsing them is unlimited. Credits are used when Sleuth builds analysis on top.
Search 150M+ documents, including sell-side research from 35+ providers, filings, press releases, and conference abstracts, then read any of them in full. Upload your own files, and Sleuth reads those too.
Ask in your own words, or start from a workflow. Either way, Sleuth plans the project, builds the datasets, reads the filings and earnings calls, and checks its own work before it builds the deliverable.
Type the decision you’re making. Sleuth answers with a plan you can edit, approve, or run straight through with auto-approve.
Pick a common job and answer a few questions. Sleuth runs a plan built for that job, with the same checks.
Sleuth starts from the fields every database has, then adds the data needed to answer your question. It reads abstracts, posters, filings, and earnings calls to fill it in, with a source on every value.
Every value has its source and a confidence score, and is marked sourced, inferred, or unknown. Sleuth checks its own work first, then Review mode shows you the few cells it couldn’t settle, and why.
Turn on monitoring, and Sleuth updates the data behind your analysis as trials, filings, and deals come in. You see what changed and where it came from, and can go back to any earlier version. Leave it on, and the analysis becomes a dashboard your team checks every week.
On its own, Claude rebuilds each landscape from the open web and misses much of it. Connect Sleuth, and Claude, ChatGPT, Copilot, or other AI tools work from the same knowledge base as Studio. Every project they run is saved to your Studio account.
63%
more of the landscape found with Sleuth connected, on average across 15 tests
+55%
Claude
+27%
ChatGPT
+108%
Perplexity

A nine-person biotech screened 167 autoimmune indications in Studio, narrowed them to one lead basket, and refined the work in place when its board pushed back.

Juvena’s CSO covers a first-in-class pipeline in Studio, from mapping who else is working on each new target to weighing which drug formats could win.
Five teams at a top-10 pharma do their research in Studio, from choosing which exhibitors to meet at a partnering event to checking a partner’s asking price.