Describe your buyer. Get the list.
Say who you sell to and Discovery goes and finds them — companies or people — checks each one against what you asked for, and writes them into a table with the evidence attached. Then the rest of the platform emails them.
- Companies or people
- Evidence on every row
- Priced per row
- Lands straight in a table
- No seat, no contract
A list you can defend, not a scrape.
The hard part of a prospect list is not finding names, it is knowing which ones actually match. Every row here carries what it was matched on.
Start from a brief
Describe the customer in plain words — industry, size, location, what they do — instead of assembling a filter query.
Or start from a domain
Point it at a customer you already have and it builds the ideal-customer profile for you, then finds more like them.
Companies or people
One axis, not six confusing categories. Ask for organisations or for the individuals inside them.
Signals, not just firmographics
Name a signal — hiring for a role, recently funded — and it pulls the sources that carry it, on top of the general web search.
Checked before it lands
A verification pass tests each candidate against your criteria and drops the ones that do not hold up, rather than handing you everything it saw.
Location that is actually right
City is read out of the listing's own address rather than stamped from your search text, so a metro-area search does not mislabel the suburb.
Straight into a table
Results arrive as rows in TinyTables with a progress banner while the run works — no CSV, no import step.
Enrich in the same breath
Found a company but need the head of marketing and their address? The enrichment columns pick up where the finder stopped.
Pay for rows, not seats
Credits are spent per record. There is no per-seat licence and no annual minimum.
From a sentence to a working list.
- 01
Say who you want
Write the brief, or hand it a domain and let it infer the profile from a customer you already like.
- 02
It searches, then checks
It picks its sources from the signals in your brief, pools candidates, and runs a verification pass over them before anything is kept.
- 03
The rows show up in a table
Each one lands in TinyTables with what it matched on. Enrich them, form-gate them, or point a campaign at the table.
It uses the rest of your TinyCommand.
Same login, same data, same bill. No export step between the thing that finds the answer and the thing that acts on it.
What Discovery is not.
Every tool’s marketing page tells you what it does. This is the part that usually isn’t on one, so you can find out now instead of after you’ve built on it.
- This is not a licensed B2B database. There is no Apollo, ZoomInfo or Clearbit underneath — everything is found on the live web and structured by a model. That is why it costs a fraction of those, and why coverage varies by industry and country.
- It is not a replacement for a list you have already validated. Treat the output as strong candidates with evidence, not as verified fact.
- Runs are started by hand. Scheduled and recurring prospecting exists in the service but is not exposed in the product, so do not plan around it.
- People search works from public web results and pages. It does not read LinkedIn, so someone whose only presence is a LinkedIn profile will come back thin or not at all.
- Email addresses are a separate enrichment step with their own per-row cost. A found company does not automatically arrive with a contact.
- Results are cached for about a week, so a company that changed something yesterday may answer with last week's picture.
- Row counts are not guaranteed. A narrow brief in a small market returns what genuinely matches rather than padding to a round number.
Questions people actually ask.
How is this different from enrichment?
Enrichment fills in rows you already have. Discovery finds rows you do not have yet. Most people use them together: find the companies, then enrich for the person and the address.
Is this a Clay alternative?
It covers the same job — build a targeted list, verify it, enrich it, act on it — inside the platform that already holds your tables, workflows and email, and priced per row rather than per seat. It is not a like-for-like feature clone.
Where does the data come from?
Live web search and page fetching across general and specialised sources, with a model structuring what comes back and a verification pass filtering it. No purchased contact database.
What does a run cost?
Credits per record — 20 for companies, 10 for people, with email verification priced separately at 5. You are charged for rows that are written, not for the attempt.
Can I start from a customer I already have?
Yes. Give it a domain and it builds an ideal-customer profile from that company, then finds others that match it.
The list is the hard part. It shouldn't cost you a seat.
Describe the buyer, get verified rows in a table, and let the workflows and campaigns you already have take it from there.