Research an account before anyone works it
Runner owns the research recipes, the evidence record and the fit scorecard your team fills in. Sources and timestamps land beside every claim.
Built into Runner
- Runner reads a company's own website and keeps what it says, with the source
- Every claim lands with its source and the time it was read
- A fit scorecard your team fills in, with the evidence behind each score
Worth knowing
Runner does the reading on a research account you connect yourself, then files the evidence for your team to review. What comes back is evidence for a decision — never permission to contact anyone.
Runner leaves a record of the reading, not a verdict
A research run ends one of two ways: with evidence on the account record, or with the reason there isn't any. A run that fails, breaks a source rule or is refused for its budget keeps nothing, so a half-read company never sits on the record looking researched.
What does land is labelled a fact, an inference, an unknown or a risk, and carries its source and the time it was read. The account's research page lists the risks first, and a claim that's no longer fresh says so.
Nobody edits a claim in place. A teammate marks it verified, disputed, corrected or dismissed with a note, or files a correction that threads beneath it, and the original stays visible, marked as superseded.
Your team turns the evidence into a fit assessment, one product at a time
A company's facts are shared across every product you run. Fit is scored separately for each, so a strong match for one product says nothing about another.
On the fit card, a person gives each of five criteria up to two points, ticks the evidence each score rests on and adds a line on why, and the card flags a score with nothing ticked behind it. The queue the account lands in comes from rules we wrote rather than a model: a criterion resting only on unknowns keeps it out of the top queue, evidence marked as contradicting itself holds it at manual review, and a recorded disqualifier outranks everything.
Each save is kept alongside the earlier ones, with the version of the scorecard that produced it.
The short version
People give the points, rules pick the queue, and the decision still waits for someone to write down why.
A person qualifies the account, with the evidence attached
Moving an account to qualified is a separate decision, and it's refused without a written reason, without evidence from that account's own record, while a recorded risk goes unnamed in the decision, or without a next step that's ready to go — one with an owner, a due date and a measure of success.
None of it decides whether anyone may be contacted. That's a separate check that reads consent and opt-outs, and research isn't one of its inputs.
Questions buyers ask
Does account research find email addresses or contact details?
No. A research run reads what a company publishes about itself and files it as evidence about that company. It creates no contact, and Runner has no source for finding people or their email addresses.
What does it actually read on a company's website?
On your connected research account, the company's own public website: which company a domain belongs to, and a summary of the site and what the company sells. Each recipe's card names the kinds of source it may read and its spending ceiling before you run it, and a recipe with nothing set up to read for it refuses rather than improvising.
Whose account does the web reading run on?
Yours. Runner does the reading on a research account you connect yourself, on your own key, with no shared research account behind it. Before each run the launcher says which reader is about to run: a live read of the public web, or a preview on sample output that visits no outside websites.
If an account lands in the top queue, can we start emailing it?
Not on the strength of the queue. The queue is a suggestion about where to spend your time. Qualifying the account is a person's decision with a written reason, and whether anyone there may be contacted is settled by the separate consent and opt-out check.
Can we research a whole list of companies at once?
Yes, as a batch: pick the accounts and one recipe. The runs go one after another, and a company researched with the same recipe shortly before is handed back rather than read again.
What happens to old fit assessments when the scorecard changes?
They stay as they were. Every saved assessment records the scorecard version that produced it, and the history flags the change, so an old total isn't compared with a new one as if the two measured the same thing.
Use what already works
Connect your own research account to supply the public-page reads. Runner turns those reads into cited evidence and a review queue; your team keeps the final decision.
How teams often buy this job
Usually bought as Clay.
Clay
Trademarks of their owners. No affiliation or endorsement.