Frequently asked questions

Getting started

Is Tabulous a good fit for us?: Tabulous fits R&D teams whose real record of their science ultimately lives in a shared drive, and who want to start adding structure to it without a custom engineering project. It shines when your workflows keep changing and you want a system that changes with them. It’s probably not the right fit if your files live somewhere other than Google Drive, SharePoint, or Box, or if you’d rather adopt a predefined data model than shape your own.
What does setup involve?: Connect your drive, set up your first tables, and start building. Describe what you need, and our AI or yours drafts it. Most labs are working in an afternoon.
Do we have to migrate our data?: Minimally. Most labs start with a new Google shared drive, SharePoint site, or Box folder, and Tabulous creates folders that match your tables. Drag in the spreadsheets you already keep, and move older files over once you like how things are organized.
Which storage providers do you support?: Google Drive, SharePoint, and Box.
What happens if we stop using Tabulous?: Your files stay in your own drive, already organized by record. Your tables aren’t locked in a proprietary format. Export them to a spreadsheet, or pull them into another database through the API. We earn our seat by being useful, not by holding your data hostage.

Which AI can we use?: Our built-in assistant, or your own. Any AI tool that supports MCP can connect to Tabulous, including Claude, ChatGPT, and Cursor, and it knows how Tabulous works the moment it’s connected. ChatGPT and Cursor have plugins available now, and Tabulous will be in Claude’s connector directory soon. Until then, you can add it to Claude as a custom connector in a couple of minutes.
Can AI change our data without us knowing?: No. Most AI tools either write to your data directly or are locked to read-only. In Tabulous, AI can do the work, but every edit arrives as a proposal, and nothing is saved until a person on your team approves it. Agents can never approve their own changes.
Can we build our own tools on Tabulous?: Yes. Ask your AI for a sample dashboard or a plate-map parser, and Tabulous hosts it as an App or Script in your workspace, with your team’s permissions already in place. Everything you build reaches both your records and the files behind them through one API, so there are no servers or credentials to manage.

How is this different from pointing Claude at our drive?: You keep Claude; Tabulous gives it more to work with. On a bare drive, Claude has to infer how everything connects from file names and folders, and what it works out stays in one person’s chat. Connected to Tabulous, it starts from records that already say which samples went into which run, and its work becomes shared tables and tools the whole lab can use.Ask Claude what it could do with Tabulous
Is Tabulous an ELN or a LIMS?: It’s the foundation you build them on. Teams use Tabulous for notebooks, sample tracking, scheduling, and analysis, shaped around how their lab works.
How is this different from Airtable or Notion?: Airtable and Notion are flexible general-purpose tools, and plenty of labs start there. But files end up as attachments or pasted links, and Notion pages don’t leave Notion cleanly. Tabulous keeps files and notebook entries in your drive as ordinary documents, and gives lab data real types, from DNA and protein sequences to plate-reader exports that become tables you can query.

Where is our data stored, and is it used to train AI?: Your files stay in your own Google Drive, SharePoint, or Box, and everything else Tabulous stores is on AWS. The built-in assistant runs on the enterprise AI services from AWS and Google Cloud, so your data is never used to train models, and requests are processed in the US. Tabulous is SOC 2 Type 2 compliant.
Can we use it for regulated work?: Tabulous keeps a full history of who changed what and when, and nothing an AI drafts is saved until a person approves it. For validated GxP work, we also offer sign-and-lock controls and a separate test workspace for validating changes before they go live. Talk to us about your requirements.
How much does it cost?: $50 per editor per month, with unlimited free viewers. The built-in assistant is available to editors and viewers alike, billed at cost plus 10%. Most people spend $5–$20 a month on it, and heavy users around $30–$60. Admins can set org-wide and per-user spend limits. AI tools you already pay for, like Claude or ChatGPT, connect at no extra charge.
Do you offer academic pricing?: Yes. Academic labs get discounted pricing. Contact us and we’ll set it up.

Set up in an afternoon