How to score LinkedIn leads by profile fit
Learn how to qualify LinkedIn leads automatically by defining a target customer profile and using a fit score to filter out bad leads before outreach.
Get startedShort answer
Qualifying LinkedIn leads automatically requires defining a strict Ideal Customer Profile (ICP) and using a scoring system to filter out anyone who doesn't match. You set a minimum fit score based on job title, industry, and company size, then only move leads above that threshold into your outreach sequence.
Define your non-negotiables
Most people start outreach with a broad job title search, like Founder or CEO. That's a mistake, because you end up with a list of people who can't buy what you sell. Before you score anyone, write down the attributes that make a lead a zero no matter what their title says.
Firmographics are the hard boundaries. If your product only works for companies with a certain employee count, or you can't support customers in a region, anyone outside those bounds is a wasted connection request. Write the boundary as a number or a named list. Mid-market is not a boundary, 50 to 200 employees is.
Personas decide who can say yes. A marketing manager at a 10 person agency has different problems and a different budget than a VP of marketing at a 500 person firm. Pick the department and the seniority level, and be honest about whether a director can sign or only a founder can.
Keep the list to three or five things you can check from a LinkedIn profile in ten seconds. If you can't see it on the profile, you can't score it, and you'll be guessing at the rest.
Company size
- Category
- 50 to 200 employees
- Example target
- Under 10 or over 1,000
Industry
- Category
- SaaS, fintech
- Example target
- Retail, manufacturing
Geography
- Category
- North America
- Example target
- EMEA or APAC
Seniority
- Category
- Director or VP
- Example target
- Junior or intern
Build a scoring rubric
A rubric turns a profile into a number, which lets you rank a thousand people and decide where to draw the line. It also makes the decision repeatable, so two people on your team sort the same list the same way.
Start with your most important requirement and give it the most points. If the job title is the only thing that matters, it gets 50. If industry is nice to have, it gets 20. The weights should reflect what you'd actually bet money on, not what sounds thorough.
Add negative points for red flags. A current employee of a direct competitor, someone whose headline says they're a consultant looking for work, or a region you don't serve should fall below the line on their own. Negative weights usually do more work than positive ones, because they catch the leads you'd otherwise spend a week on.
- 1
Assign positive weights
Give 50 points for the exact target title, 30 for the right industry, 20 for company size. That totals 100.
- 2
Apply negative weights
Subtract 100 for competitors, agencies selling to your buyer, or regions you don't serve.
- 3
Set the fit line
Pick a minimum, such as 70. Anyone at 69 or below is dropped before a message gets written.
Paste this job
Score this LinkedIn profile from 0 to 100 against my ICP. +50 exact title: Head of Marketing, VP Marketing, CMO +30 industry: B2B SaaS or fintech +20 company size: 50 to 200 employees -100 current employer is a competitor: Acme, Northwind -100 headline mentions open to work or consultant Return the total, the lines that matched, and the one line that decided the score.
Filter leads before the first message
Plenty of teams message everyone in a broad search and qualify on who replies. That spends your reputation on people who will never buy and fills your inbox with conversations that go nowhere. It also makes the list impossible to learn from, because you can't tell which attribute brought the reply.
Filtering first means connection requests land on people who fit. When the person on the other end actually has the problem you solve, the request reads as relevant instead of random, and the first line you write has something real to say.
The point is precision, not volume. Twenty messages to people who match beat two hundred to a mixed list, and the smaller list is the one you can follow up on properly.
The manual way
- Export 500 people from a search
- Open every profile in a browser tab
- Delete the ones that don't fit by hand
- Spend an afternoon qualifying one list
The scored way
- Run the list through a scoring pass against your ICP
- Drop everyone below the fit line before anything sends
- Only qualified leads enter the campaign
- Re-score the same list next month when your ICP changes
Where scoring goes wrong
A rubric is a model of your customer, and models drift. These are the failures that show up most often.
- Scoring on headline keywords. People write their own headlines, so Founder | Growth | AI tells you almost nothing about the company they run or whether they buy software.
- Weighting six weak signals. Averaged together they produce a score nobody can explain or defend. Three strong attributes are easier to check and easier to change.
- Never reading the bottom of the list. Look at the leads that landed just under the line every couple of weeks. That's usually where the rubric is wrong, not the leads.
- Scoring once and never again. A lead who scored 40 last quarter is still a 40 unless something changed, and you won't see the change unless you re-run the list.
- Letting the number replace reading the profile. The score sorts the list. You still open the top twenty and read them before you write anything.
- The fix for most of these is the same: fewer attributes, checked more often, with a person reading the top of the list before a campaign goes out.
Refine the fit line as replies come in
Your first fit line is a guess. Adjust it from the quality of conversations, not from how tidy the list looks.
If nobody replies, the line is probably too high, or the message is wrong. Work out which one before you lower the score, or you'll fix the wrong problem.
If you get replies but no meetings, the line is too low. You're reaching people who like the topic but can't approve a purchase.
Pull the leads that landed five or ten points under the line and read ten of them. If they look like customers, lower the threshold. If they look like the people you already rejected, leave it where it is.
Do this every few weeks rather than every day. A fit line moved after two replies is just noise, and you'll never learn what the last change did.
Never lower the fit line past your non-negotiables. A great title doesn't fix a company that can't buy.
Write the first message from the score
The score tells you which attribute made someone a lead. Use that in the first line. If the lead scored on company size and industry, say what you noticed about the company.
A connection note under 200 characters that opens with why that person became a lead reads like a person wrote it. Generic praise doesn't, and the reader has seen it forty times this month.
Skip the compliment. Love your content is the same message everyone sends, and it tells the reader nothing about why they're hearing from you today.
Example
- You
Hi Priya, saw Northwind is hiring two SDRs, which usually means outbound is about to get busier. Curious how you're handling the LinkedIn side of it.
- Prospect
We are, though the LinkedIn piece is the part nobody owns yet. What do you do?
- You
Mostly the first touch and the follow-up, so the SDRs start with replies instead of a cold list. Worth 15 minutes next week?
Doing this with Kinlead
Kinlead finds people on LinkedIn who match the customer you describe, or your website, and scores each lead from 0 to 100 against that profile. Leads below the fit line are not kept.
Kept leads go into lead groups and scoring. Any campaign can point at any group, and you can export a group to CSV or import a CSV of LinkedIn profile URLs if you already have a list.
Outreach runs from your own LinkedIn account through Unipile, at one action every five minutes inside the daily limits and send window you set. Queued messages can be read and edited on the Actions page before they send.
- 1
Describe the customer
Enter your customer description or website URL so Kinlead knows what a good lead looks like.
- 2
Score and filter
Profiles are scored 0 to 100 against that description, and anything under the fit line is dropped.
- 3
Point a campaign at the group
Send connection requests and follow-ups from your own account, inside the limits you set.
Frequently asked questions
Related
Get started with Kinlead
Prefer markdown? How to score LinkedIn leads by profile fit.md