Fit, before the first message

ICP Scoring: Rank Every Lead Before You Contact One

You get about 100 invitations a week. Order decides the result. LinkedGrow reads your website, builds the profile of who actually buys from you, and scores every person it finds against it. The strongest fits rise to the top of the queue with the post that surfaced them attached, and weak matches never make it into the list at all.

Profile read from your siteScored before contactEvidence on every score

7-day trial · Cancel before day 7 · Everything included

179+ founders already run their LinkedIn with LinkedGrow
Quick answer

What is ICP scoring and why does it matter?

It ranks each prospect against your ideal customer profile before anyone is contacted, so the strongest fits get your limited invitations first. LinkedGrow builds the profile from your website, shows the post behind every score, and keeps weak matches out of the queue entirely.

You cannot contact everyone, so order is everything.

The invitation budget is fixed. What changes your results is who gets the ones you have. Most prospecting treats the list as a queue to work through top to bottom, in whatever order the search returned it. That order is arbitrary, and it means the best prospect of the month might sit at position 400 until the list goes stale.

01
400

A LinkedIn search sorts by its own relevance, which knows nothing about who buys from you. Working it top to bottom spends your best week of invitations on whoever happened to rank first.

02
0

Eyeballing profiles works for 10 people and collapses at 100. Without a consistent measure, the decision comes down to whoever had the most interesting headline, which is not a buying signal.

03
100

The ceiling is not going up. Spending it on people who were never going to buy is the most expensive mistake in outreach, and it is invisible until a quarter later when the pipeline is thin.

04
2

Poor targeting looks identical to good targeting for a fortnight, because the invitations go out either way. The gap only appears when the replies do not.

We contacted 300 people and the two that closed were both in the last twenty we got to...

How it works

From your address to a ranked queue

Scoring is only as good as the profile behind it, so the profile is the part you get to correct first.

1

It proposes the profile

Paste your website address and the agent works out the buyer, the roles, the industries, the sizes and the topics. It takes about a minute to read and fix, which is far less than writing it from nothing.

2

You correct it

Add the job titles it missed, remove the industry you never win in, tighten the headcount band. Every change moves every future score, so the correction is worth more than any single lead decision.

3

It ranks what it finds

Candidates are scored against the corrected profile, deduplicated across your agents, and ordered so the best fit of the week is the first person you see. You approve the queue or let it run.

How the score is built

A score you can argue with

An opaque number is useless, because the only sensible response to a bad ranking is to fix the profile behind it. Everything here is visible and editable.

Measured against your own words

The baseline

The profile comes from your website, not from a category you picked off a list. Roles that decide, industries you win in, headcount that can afford you, and the topics those people follow. You edit any of it in one screen.

  • Read from your site
  • Editable in one place

The evidence stays attached

Each scored lead links to the post or comment that surfaced them. You can see what the score was reacting to, which is what makes it possible to tell a wrong score from a wrong profile.

  • Source post per lead
  • Matched fields shown

Weak fits never queue

Sourcing finds more people every week than the account can safely contact, so the floor matters as much as the ranking. Anything below it is dropped rather than sitting in the list waiting to waste an invitation.

  • Hard floor
  • Queue stays short

One profile per agent

Selling to agencies and to recruiters at once means two profiles and two scores, so run an agent for each. They share the account's daily budget and deduplicate against each other, so nobody hears from both.

  • Separate agents
  • Shared deduplication
Before you ask

Questions about ICP scoring

Short answers to what people ask most often.

Still have a question?

Write one line about what you sell and we will answer honestly.

Email us directly

Ranking each prospect against your ideal customer profile so the strongest fits get contacted first. With about 100 invitations a week available, the order you work the list in decides most of the result.

The agent reads your website once and proposes the roles, industries, company sizes and topics that describe your buyer. You correct anything it got wrong in a single screen, and the corrected version is what every score is measured against.

Yes. Every lead links back to the post or comment that surfaced it, alongside the profile fields it matched. You can disagree with a score, and disagreeing is useful because it usually means the profile needs one line changed.

Weak matches never enter the queue, which is the point of scoring at all. Sourcing finds far more people than you can safely contact, so the job is choosing rather than reaching everyone it found.

CRM scoring usually runs on people already in your pipeline, using behaviour you recorded. This runs before contact, on strangers, using public signals and the fit criteria you set, to decide who is worth an invitation at all.

Spend the 100 invitations well

Contact the best twenty, not the first twenty.

The agent reads your site, scores everyone it finds against that profile, and puts the strongest fits at the top with the evidence attached. Give it your website address and see the first ranking.

7-day trial · Cancel before day 7 · Everything included