FIT · 0–5
Is there a plausible buyer, problem, and offer?
A 3 means the role, company, and problem align. A 5 adds an exact buyer, exact problem, and current reason to act. Funding or hiring can prompt research; neither proves need.
Private, in-browser worksheet
Score the evidence, not the person. Fit and relationship remain separate so neither dimension can hide weakness in the other.
Privacy: all worksheet logic runs locally in this page. The tool does not ask for a person’s identity, persist inputs, set tool-specific cookies, or transmit scorecard data. The site’s general analytics and privacy terms still apply to page visits; see the Privacy Policy.
This worksheet adapts the two-axis framework in Every’s guide to finding clients in your existing network. It does not use a hidden model or universal conversion benchmark.
FIT · 0–5
A 3 means the role, company, and problem align. A 5 adds an exact buyer, exact problem, and current reason to act. Funding or hiring can prompt research; neither proves need.
RELATIONSHIP · 0–5
A 3 means repeated interaction, a mutual collaborator, or a credible introducer. A social connection alone can still be cold.
TIMING
A role change, known budget cycle, public launch, direct request, or offered introduction may be relevant. Timing is a note to verify, not a magic multiplier.
EVIDENCE
Limited evidence routes to verification first. Mixed evidence calls out remaining inferences. Clear evidence is current, specific, and appropriate to use.
The order matters. Weak evidence is checked before any score-based recommendation, and no lane authorizes outreach.
If the record is mostly inferred, verify role, relationship path, and timing before treating either score as actionable.
Fit below 3 stays out of a sales lane. Relationship below 3 is not called warm, even when fit is high.
When both axes qualify but no current signal exists, maintain the relationship or set a later review rather than manufacturing urgency.
A plausible trigger still needs a current source before it supports a contextual first touch.
Strong inputs create a human-review priority. They never become a probability, forecast, or automatic outreach decision.
A reviewer still chooses whether to reconnect, request an introduction, begin a direct conversation, or do nothing.
Read how Every handles warm-lead discovery, including authorized relationship sources, pending review, and controlled external sends. If you are comparing the whole operating system, model both stacks with the service-business stack cost calculator.
Full methodology
Find clients in your existing network without cold outreach
Source mapping, scoring rubrics, timing, warm paths, failure modes, and outreach-law references.
It applies a transparent set of review rules to four inputs: business fit, relationship strength, timing, and evidence quality. The output is a suggested review lane and next step, not a conversion probability or sales forecast.
A close relationship can be a poor business fit, while a well-matched buyer may have no credible relationship path. Keeping the axes separate prevents a high score on one dimension from hiding weak evidence on the other.
No. The scorecard does not search for people, import people, identify email addresses, send messages, or make outreach decisions. It reviews numbers you enter for one anonymous opportunity.
No. A review result is not consent, legal advice, or a compliance decision. You remain responsible for the relationship context, channel expectations, applicable law, recipient, and whether any contact is appropriate.
No. This worksheet runs in the page, sends no scorecard inputs over the network, and does not store them in cookies or browser storage. Resetting or closing the page clears the worksheet.