Where the data comes from, how to check it, what stays private, what it costs, and how Sleuth works in Claude, ChatGPT, and Copilot.
Sleuth Insights is AI software and infrastructure for biopharma strategy, business development, and competitive intelligence teams: Sleuth builds analysis for questions like which indication to lead with or which pharma to pitch. Every plan includes three parts: analyses built for your questions, an explorable knowledge base, and searchable documents. Use them in Sleuth Studio, or in Claude, ChatGPT, or Copilot, or have Sleuth’s analysts do the work with Concierge. See the ways to work with Sleuth.
Sleuth builds a dataset for each question you ask: a table of the programs, companies, or trials that fit, with the columns you ask for and a source on every value. Then it builds what you’ll hand over on top: an interactive analysis, a deck, or a spreadsheet. Both stay in a project in your account, where you can reopen it, extend it, or keep it current. See what Sleuth Studio builds.
Sleuth Studio is the app where you work with Sleuth directly: describe your question, approve the plan, and watch Sleuth build the dataset and the analysis, then check the values Sleuth flags in Review mode and keep the project current with monitoring. Projects you run in Claude, ChatGPT, or Copilot are saved there too. Explore Sleuth Studio.
Sleuth is used by business development, strategy, and competitive intelligence teams at biotech and pharma companies, plus the investors and bankers who work with them. Customers range from one-person companies to five teams at a top-10 pharma. Read customer stories.
Sleuth gives ChatGPT and Claude the library they’re missing: a biopharma knowledge base, already matched and checked, with a source and a confidence score on every value. On their own, they’re strong analysts who rebuild each landscape from web searches and can’t reach sell-side research or China’s registries in Chinese. Some of what they find is wrong, and it’s difficult to tell where. Sleuth also works inside Claude, ChatGPT, and Copilot. See how Sleuth compares with ChatGPT and Claude.
Yes, for most biopharma strategy work. Every plan includes the same kinds of records, connected in Sleuth’s knowledge base: programs by target, modality, and phase, with catalysts and sales forecasts; trials from registries worldwide; and deals with upfronts, milestones, and royalties where they’re disclosed. Sleuth adds sell-side research and China coverage, and browsing is free and never uses credits. Then Sleuth does the narrowing down you’d otherwise do in Excel. Compare Sleuth with your database.
Yes. For biopharma documents, Sleuth is an alternative to AlphaSense: it searches the sell-side research, filings, earnings calls, press releases, and conference abstracts teams use AlphaSense for, and turns what’s in them into one table across every program. If your team relies on AlphaSense’s expert-call library, keep it for that. See how Sleuth compares with AlphaSense.
A Sleuth analysis takes longer than a chat answer, because Sleuth plans the project, builds and checks a dataset, and then builds the analysis. Larger landscapes take longer, and you can watch the project build. Against the work it replaces, it’s fast: one biotech screened 167 indications for its platform in a matter of days. See how a project runs in Sleuth Studio.
Sleuth’s data comes from the organizations that publish it: trial registries, regulators, patents, journals and medical meetings, companies, news, and sell-side research from 35+ providers, including Chinese sources read in Chinese. Every source is listed on the Data page.
Sleuth’s knowledge base is all of Sleuth’s data, connected: 150K+ drugs, 1M+ trials from 20 registries, 400K+ companies, 200K+ deals, and 150M+ documents, as of November 2026, each matched across its sources and linked to the documents behind it. Studio, Claude, ChatGPT, Copilot, and Sleuth’s analysts all work from the same knowledge base. See what’s in the knowledge base.
Yes. Sleuth’s knowledge base includes clinical trials from registries worldwide, including China’s, each linked to its drugs, sponsors, sites, and results. Trials are matched to the same drug records as programs and deals, so one question reaches all of them. See the trial registries Sleuth covers.
Yes. Sleuth’s knowledge base includes deals across licensing, M&A, collaborations, and financings, with upfronts, milestones, and royalties where they’re disclosed. Each deal is linked to its drugs, its companies, and the filings and press releases behind it. See Sleuth’s deal data.
Sleuth updates over 95% of its sources in real time, and runs every new document through the same matching and checks before it reaches the knowledge base. Turn on monitoring, and every project stays current too: Sleuth updates it as new trials, filings, and deals come in and shows you what changed. Monitoring only charges for what’s new. See how monitoring works.
Yes. Sleuth covers over 22,000 Chinese programs, as of November 2026, from China’s own registries, regulators, exchanges, company posts, and broker research. Each source is translated and run through the same matching and checks as every other source, so you see a Chinese company’s program in the same table as a US one, under the same drug record. See Sleuth’s China data.
Every Sleuth plan includes sell-side research from 35+ providers, and you don’t need an account with each bank. The providers include global investment banks, US healthcare and mid-market banks, European brokers, independent and sponsored research, consulting and industry research, and historical archives. Ask us for the full list of providers.
Yes. Sleuth lets you read any sell-side report in full, search it, and see the exact passage behind a value, and you can download every dataset built from them. See how sell-side research works in Sleuth.
Yes. Sleuth covers preclinical programs and private companies: about 60% of the 150K+ drugs in its knowledge base are preclinical. Early programs come from company pipeline pages, conference abstracts and posters, and patents, including programs that so far appear only in patents. Private companies come with their pipelines, deals, and financings. See Sleuth’s sources for early-stage programs.
Yes. Sleuth tracks programs that stop, including ones dropped quietly: it keeps earlier versions of company pipeline pages, so a program removed without a press release shows up as a change. Where there’s a public reason, Sleuth records why the program stopped. See how Sleuth catches quiet changes.
Yes. Every Sleuth plan lets you start from any drug and browse its trials, deals, companies, and targets, the same records Sleuth uses, and send what you find to an analysis. Exploring the knowledge base and searching documents are free and unlimited on every plan. Explore the knowledge base in Studio.
Every value in a Sleuth dataset links to its source and has a confidence score, based on how good its best source is, whether independent sources agree, and, where it matters, how recent it is. When Sleuth can’t find a source for a value, it marks it unknown instead of filling it in, so you can see what to trust and what to check. See how Sleuth scores every value.
You don’t have to check every cell in a Sleuth dataset: values with strong sources that agree score high, and Review mode puts only the low scorers in front of you, with the reason. In one real dataset, Review mode flagged 21 of 3,600 cells for a person to check. See how Review mode works.
In Sleuth, click any value to see its source, the passage it came from, and a confidence score. Then open Review mode, which shows only the cells worth a second look, and mark each one reviewed or switch it to another source. See how to check a value.
Sleuth checks every value before you see it. Its engineering reconciles and normalizes every record, AI checks run at each step of the pipeline, disagreements between sources are flagged instead of smoothed over, and Sleuth’s data team red-teams the results. Every value is then labeled sourced, inferred, or unknown, so you can check any number in one click. See how Sleuth checks every value.
In Sleuth, sourced means a document Sleuth links to states the value, and inferred means Sleuth worked it out from its sources and shows how. Unknown means Sleuth couldn’t find a public source for the value, so it leaves the cell open instead of filling it. See examples on the Data quality page.
When two sources disagree, Sleuth keeps both, goes with the record it scores higher, and flags the cell so you can see the conflict. In Review mode you see both sources side by side and can switch to the other one. See how Sleuth handles conflicting sources.
Yes. On every plan, your questions, uploads, and results stay private to your account and are never shown to other customers. See how Sleuth uses your data.
Yes. Sleuth uses large language models from Anthropic, OpenAI, and Google, hosted by those companies. On Enterprise, you can choose models hosted on AWS Bedrock instead, so your data stays in Sleuth’s private network and isn’t sent to the model providers. See how Sleuth handles security.
Yes. Sleuth has a SOC 2 Type II report, so its controls are independently audited. Your data is encrypted in transit and at rest and stored in the United States. The SOC 2 report, controls, and policies are in Sleuth’s Trust Center. SSO, a single-tenant option, and a custom MSA come with Enterprise.
Sleuth costs $249 a month for Starter, for one person with 3,000 credits, and from $499 a month for Team, with 6,000 credits shared by any number of people. Credits pay only for the custom analysis Sleuth builds. You can start with a 14-day free trial, no card needed. See every plan on the Pricing page.
In Sleuth, one credit is one analytical cell: a judgment Sleuth makes about one item in your research, like an asset’s fit with your strategy. Browsing the knowledge base, searching documents, and asking about your results, charts, and exports never use credits. See how credits work.
A Sleuth project uses one credit for each item judged on each criterion, so scoring 150 assets on 3 criteria uses 450 credits. Facts from the knowledge base, like phase and sponsor for 100 drugs, use none. Sleuth shows you its plan before it runs, so you see the size of the job first. See a worked example.
Sleuth’s Team plan has no per-seat price: any number of people share one pool of credits, and one admin manages seats and spending. It costs $499 a month (6,000 credits), $799 (10,000), $1,899 (25,000), or $3,599 (50,000), with admin included. Starter is a one-person plan at $249 a month. Compare Starter and Team.
Sleuth starts at $249 a month with programs, trials, deals, and documents included, plus 3,000 credits of custom analysis, while a database subscription usually costs $20K+ a year for the data alone. With a database, the analysis is still on you. See what every plan includes.
When a Sleuth plan runs out of credits, a run that’s already going always finishes. You get alerts before that, at 75% and 90%. Past your plan’s credits, you can keep going at a fixed price per credit, up to a monthly limit you set. The limit starts at $0, so nothing extra is charged unless you raise it. See how extra usage works.
Yes. Unused Sleuth credits roll over: on monthly plans, your balance can hold up to two months of credits, so 6,000 on Starter. Annual plans can hold up to two extra months on top of the year’s credits. Bonus and trial credits never expire and don’t count toward either limit. See the rollover rules.
Yes. Sleuth’s annual Team plans cost 10 times the monthly price, so you get two months free. All the credits arrive upfront. Invoices are available on annual Team plans of 10,000 credits a month or more, and on Enterprise. See annual pricing.
Yes. You can cancel or downgrade a Sleuth plan at any time, and the change takes effect at the end of your billing period. Upgrades take effect right away, and the extra credits arrive with them. See how plan changes work.
Sleuth Enterprise has no list price: Sleuth prices it for the teams, systems, and security requirements involved. See what Enterprise includes.
Yes. Sleuth works in Claude, ChatGPT, and Copilot on every plan: add Sleuth to the AI tool your company already uses, sign in with your Sleuth account, and ask there. Sleuth builds the analysis from the same knowledge base and documents as Studio. Every project is saved to your account. Set up Sleuth in your AI tool.
Yes. In 15 tests, Claude, ChatGPT, and Perplexity each found more of a biopharma landscape with Sleuth connected: 63% more on average. The method and every test are on the Benchmarks page.
Sleuth works in Microsoft 365 Copilot through Copilot Studio: someone with access adds Sleuth to an agent there, then publishes that agent to Copilot or Teams. Each person signs in with their Sleuth account the first time, then asks questions in Copilot. See the Copilot setup steps.
Yes. Sleuth has an MCP server at https://mcp.sleuthapi.com/mcp, so anything that supports MCP, the standard AI tools use to connect to outside data, connects to Sleuth on any plan, including Gemini and agents your team builds. See how to connect Sleuth over MCP.
No. Using Sleuth in Claude, ChatGPT, or Copilot costs nothing extra: connecting is included in every plan, and analysis costs the same there as in Sleuth Studio. See how credits work in AI tools.
Yes. Sleuth works with your own documents on any plan: upload your files, and Sleuth reads them alongside everything else, or start a project from your own spreadsheet. On Enterprise, Sleuth also connects to Drive, SharePoint, Box, or your data warehouse and matches your internal program names to the same records. Talk to us about connecting your systems.
Sleuth exports PowerPoint decks, interactive reports, charts, and every dataset as a download. You can also hand a project to Claude or another AI tool and keep working there. See the outputs in Sleuth Studio.
Yes, with Enterprise. Sleuth’s API is for applications and systems that call Sleuth directly. Agents that support MCP don’t need it: they connect on any plan, the same way Claude does. To build Sleuth data into a product you sell, you’ll need Enterprise too. See Sleuth for Enterprise.
Sleuth Concierge is Sleuth’s analyst service: Sleuth’s analysts take your question and deliver a finished report or deck. The project behind it is saved to your Studio account. They start the day you send it, and you see the first findings within days. See how Concierge works.
Sleuth Concierge has no list price: Sleuth quotes each question, for a fraction of what a consulting project costs. Start with one project, or keep the analysts on retainer and send new questions as they come up. Ask for a Concierge quote.
A Sleuth custom dataset is a dataset Sleuth builds to your spec and keeps current, like a competitor tracker or a target list your team updates by hand today. You set the rows, the columns, and the rules, and Sleuth delivers it on your schedule, into your systems. See how custom datasets work.
Sleuth checks every Concierge and custom dataset delivery before it reaches you: an analyst checks Concierge work, and quality checks built for each dataset run on every custom dataset delivery. In Studio, you review the few cells Sleuth flags yourself. Compare how each way is checked.
To start with Sleuth, click Start free, choose Sleuth Studio or your AI tool, and sign up. Every new account is approved by hand, and your invite comes by email.
Yes. Sleuth’s free trial gives you 14 days of everything in Starter, with 2,500 credits for analysis and no card. When you choose a plan, any trial credits you haven’t used come with you. See the trial details.
Sleuth for Startups gives companies that have raised under $10 million a free year of Sleuth for the whole team, with 3,000 shared credits a month for analysis. Investors and incubators that back early companies can apply too. Apply for Sleuth for Startups.
Start from a Sleuth workflow built for a common job: indication selection, pharma partnering, standard of care evolution, asset scouting, asset assessment, or graveyard analysis (why programs in your space stopped). Answer a few questions about your program, and Sleuth runs a plan built for that job. Browse the use cases.