the first question

Who to pitch

A list of companies is easy to get and nearly worthless. What you need is a short list you could defend out loud: each one with the case for it, the case against it, a real person to write to, and the shortest warm path in. Customers, distribution partners, resellers, companies worth investing in: whichever of those you asked for. That is what a run gives you before it writes a word.

Scan your site for free Then: what to send →
01

It works out who the company actually is, before anything else runs.

A target is identified by its website, not by its name. The studio opens the site and follows the pages that name an operator (footer, terms, privacy, about), reads what they sell and to whom, and only then starts searching. Everything downstream is handed that as settled fact.

It reports back the pages it opened and the same-named businesses it ruled out. That second list is the point: "Northwind" is four companies, and a pitch researched against the wrong one is worse than no pitch at all.

You see the pages read and the companies ruled out, on every run.

02

It scores fit against your brief, and argues both sides.

Your brief says what you are hunting for, in your own words: new customers, agencies who could resell you, companies worth investing in, whatever it actually is. Every company found is scored against that, and the score comes with a written argument and the signals underneath it: the ones for and the ones against, side by side, each rated by how recent it is and how often it recurs.

A tool that only ever says yes is a tool you stop reading. The signals against a company are what make a shortlist worth the name.

Fit thesis · signals for and against · recency and recurrence on each.

03

Decision-makers, with sources. No source, no name.

A name appears only where a credible public source supports it, and the pitch links what was read. Every name is rated: confirmed when the company or the person states the role, likely when a secondary source does, unverified when it is inference. Where nothing backs a name, the role comes back blank for enrichment rather than filled with a guess.

A LinkedIn profile is linked only where one was actually read. Where it was not, you get a search instead, drawn dashed, so it can never pass for a profile the studio found.

The confidence rating is about the name, not the person.

04

Lookalikes, so one good target becomes a quarter's worth.

Every run ends with companies like the one you just researched. Tick the ones worth pursuing and the button counts them ("Run 3 pitches"), so a shortlist goes out in one action. A bracketed name is a gap in the research rather than a target, so it never gets a checkbox.

Already have the list? Paste it. Up to a hundred companies at a time, researched and drafted while the tab is closed.

A run is a background job. Start it and leave. It finishes without you.

05

The warm path in, from the network you already have.

Upload your connections once and every target is checked against them. Where somebody you already know can open the door, you get the chain: you, the connector, the company, and who they are to you.

Email addresses in a LinkedIn export are deleted at import, and your network never travels on a shared pitch.

And the intro request comes drafted. That half is on the next page.

the honest bit

This half is free, and it is the easier half.

Point the scan at your own site and it will tell you who it would hunt for you, without an account and without a card. We give it away because a list of names is not the thing that is hard to get, and it is not the thing you are behind on.

What takes an afternoon per company is the other half: the deck, the emails, the proposal and the call sheet, in your voice, with nothing invented. That is the half Prospektor is actually for.

What to send →