Sheets
A sheet that keeps running
Sheets are a remote workspace for your lists, with schedules, signals, and webhooks built in. Ask in plain language and the agent sets it up; every column stays yours to adjust.
Agents in SheetsThe Swiss Army knife
for agentic GTM.
Clay-like tables, GTM MCPs, and data APIs used to be three products. We forged them into one. It makes sense the moment you try it.
Used by GTM and product teams at
MercorWe work backwards from one question: how should GTM work with agents? In that world, research takes minutes, every idea gets tested, and anyone can build production systems without being an engineer. That world needs a new kind of tool.
Be water, my friend.
Empty your mind of tables and tools. Ask in a chat, open it as a sheet, call it from code. Same list, any shape.
Table tools
Built for your team. Every table is clicked together by hand.
✓ Team– Agents
Data tools for agents
Built for your agent. The data comes back in the chat, and the rest is up to you.
– Team✓ Agents
pipe0
Built for both. Your agent builds it, your team can see it, and nothing is rebuilt in between.
✓ Team✓ Agents
ToolTeamAgents
Table tools
Built for your team. Every table is clicked together by hand.
Data tools for agents
Built for your agent. The data comes back in the chat, and the rest is up to you.
pipe0
Built for both. Your agent builds it, your team can see it, and nothing is rebuilt in between.
From a tangled stack to one platform.
Enrichment, lists, routing, sync, and automations, orchestrated in one place instead of spread across tools like these.
Clay
Zapier
n8n
Chili Piper
Hightouch
Polytomic
Lusha
BetterContact
More of your list, found.
Same 150 people, two waterfalls. The difference is which providers ask first.
Share of 150 profiles with a mobile number found
pipe0 waterfall
94%
Traditional waterfall
72%
Two cheap providers
36%
Mobile numbers found from 150 LinkedIn profiles, same rows for every route, August 2026. The traditional waterfall asks the cheapest provider first. pipe0 paid about 12¢ per number found. A traditional waterfall costs around 52¢ per result. Method and data
Work anywhere.
The app, your coding agent, Slack, or your own code. Every surface runs on the same data model.
MCP
Claude, Codex, and Cursor
Connect the MCP server and your agent gets every search and pipe as a tool, with data back in the same call. When a list gets big, it writes it to a sheet your team can open.
MCP serverSlack
Ask from Slack
Mention @pipe0 in a channel. It researches the account, finds contact data, and answers in the thread, with a link to continue in pipe0.
Slack agentAPI
Build it into your product
Typed REST endpoints and a TypeScript SDK for every search and pipe. Ship enrichment behind your own UI, billed per result found.
Enrichment APISheets
A sheet that keeps running
Sheets are a remote workspace for your lists, with schedules, signals, and webhooks built in. Ask in plain language and the agent sets it up; every column stays yours to adjust.
MCP
Claude, Codex, and Cursor
Connect the MCP server and your agent gets every search and pipe as a tool, with data back in the same call. When a list gets big, it writes it to a sheet your team can open.
Claude Code
Codex
Cursor
Slack
Ask from Slack
Mention @pipe0 in a channel. It researches the account, finds contact data, and answers in the thread, with a link to continue in pipe0.
Florian
@pipe0 who in #customers works in engineering?
API
Build it into your product
Typed REST endpoints and a TypeScript SDK for every search and pipe. Ship enrichment behind your own UI, billed per result found.
6–12x more cost‑efficient than Clay.
Premium enrichment data, pay as you go. No subscription required, and you pay only for data a provider found.
Mobile number
average cost per result
- pipe0
- 10¢
- Typical
- 25–100¢
Work email
average cost per result
- pipe0
- 3¢
- Typical
- 8–30¢
Mobile number
average cost per result
- pipe0
- 14¢
- Typical
- 25–100¢
Work email
average cost per result
- pipe0
- 4.5¢
- Typical
- 8–30¢
Actions
CRM writes, Slack messages, and sequence enrollments on your own connections. No second meter that grows with every step.
$3*per 100k actions
* On plans above $300/mo
Florian Martens
Founder · Berlin and San Francisco
Built data pipelines at Cloudflare, then joined Cursor as its first growth engineering hire and built its GTM data platform.
“Don't trust any vendor's find rate, ours included. Run a real list through pipe0 next to your current tool. If the difference isn't obvious, we did something wrong.”
Questions teams ask before switching.
Do I have to rebuild my Clay tables?
No. Export the table as a CSV and import it into a sheet, or describe the workflow and let the agent add the columns. The rows you already enriched stay as they are.
How is pipe0 different from Clay?
Clay made spreadsheet enrichment mainstream and has the larger template library. pipe0 gives you the same table workflow with tables up to 2M records, plus a full API and MCP server on the same engine, so agents can build and run the work too. Full comparison
How do you keep the data accurate?
A waterfall stops at the first answer, so the first provider decides accuracy. We benchmark providers against premium data and only add one when it finds data the others miss. Before you send, run email verification as its own step. How we benchmark
Can pipe0 do CRM enrichment?
Yes, without a separate product. Put a sheet on a schedule: pull the accounts created yesterday from HubSpot, Salesforce, or Attio, enrich them, and write the results back to the CRM. Schedules
Do I need a subscription?
No. Buy credits as you go and they never expire. Monthly plans lower the price per credit when your volume grows. Pricing
Can I take my data with me?
Always. Export any sheet to CSV at any size, write rows back to HubSpot, Salesforce, or Attio, or read them over the API.
Can I use my existing provider contracts?
Yes. Store your own API key for a provider and pipe0 uses it instead of the managed connection, for a small platform fee per call. Connections
Do I need to be technical?
No. Most work starts as a sentence to the agent, and every column can be adjusted by hand. Engineers get the same engine as an API, a TypeScript SDK, and an MCP server.
Give your GTM team its AI moment.
Run a few rows of a list you already enriched somewhere else. Compare the hit rate and the cost per contact.