# How to prioritize LinkedIn connection requests by lead quality
> Score each LinkedIn lead from 0 to 100, then spend your personalization where it pays: handwritten notes for high-fit leads, short ones for the rest.
- HTML: https://getkinlead.com/connection-requests-for-high-fit-score-leads
- Markdown: https://getkinlead.com/connection-requests-for-high-fit-score-leads.md

**Short answer:** Score every lead against a short rubric of must-have and nice-to-have criteria, then spend your personalization where it pays. High-fit leads get a note written from something specific on their profile, mid-fit leads get a short industry note, and low-fit leads get skipped so they don't eat your daily sending room.

## Define your high-fit criteria

Fit starts with a definition you can check in ten seconds. Open a blank note and write down who actually buys from you: their title, the size of the company, the industry, and what they were doing in the week before they got in touch.

Then split that list into must-haves and nice-to-haves. A must-have is something the person has to have before your product makes any sense for them. A nice-to-have is a hint that they're paying attention to the problem right now, not a requirement.

| | Dimension | Must-have | Nice-to-have |
| --- | --- | --- | --- |
| Job title | Owns the number: VP of sales, head of growth, founder | Sits next to the number: sales manager, RevOps lead |
| Company size | 50 to 200 people with one sales team to fix | 11 to 49 people still deciding who sells |
| Hiring | Hired SDRs or posted a sales role in the last 60 days | Wrote about pipeline or quota in the last 30 days |
| Go-to-market | Already runs outbound, so the problem is real | Inbound-heavy team starting to look at outbound |
| Recent activity | Commented on a competitor's or peer's post this month | Posted anything at all in the last quarter |

## Score each profile before you send

A score is shorthand for a decision you'd otherwise make by mood. Give each must-have 20 points and each nice-to-have 5, cap the total at 100, and write the number next to the name.

The exact weights matter less than using the same weights every day. Two profiles that look the same should get the same number, otherwise you're still guessing.

1. **Write the rubric down.** Five must-haves at 20 points each, four nice-to-haves at 5 points each. Keep the arithmetic simple enough to do in your head while scrolling.
2. **Read the profile top to bottom.** Current title, company headcount, the last two posts, and anything in the About section that names the problem you solve.
3. **Add the points and record the score.** Write something like 'Acme, VP Sales, 90' in your list so the number is visible when you come back to it three days later.
4. **Set a fit line and stick to it.** Pick a number, say 70. Above it you write by hand. Below it you send a short note or skip the person entirely.

> Write one line on why a lead scored what they scored. When the score turns out to be wrong, that line is what you fix.

## Match the message to the score

Time is not evenly spread across your list. A 95 gets twenty minutes of reading and a note that proves it. A 45 gets sixty seconds and a sentence that shows you know their world.

The two failure modes are treating everyone like a 45 and calling it efficiency, or treating everyone like a 95 and burning the whole morning on nine people.

**Weak: one note for the whole list**

- Hi Mark, I'd like to connect and share how we help SaaS companies grow revenue.
- Hi Priya, I came across your profile and thought we should connect. I work with sales leaders like you.
- Hi Tom, hope you're well. Would love to add you to my network.

**Strong: note written for a 90+ lead**

- Hi Priya, your comment about reps sandbagging deals on Dana's post is an argument I lost internally last quarter. We rebuilt our comp plan around it. Worth comparing notes?
- Hi Tom, you're hiring two AEs for the mid-market team at Acme. I ran that same hiring wave at Northwind and got the ramp wrong the first time. Happy to share what I'd do differently.

## Write the note after you read, not before

The note is the whole test. If you can swap the name and the company and the note still works, you haven't personalized anything. You've just added a mail merge field.

Read the last three posts and the About section, then find the one detail you can say something true about. A comment someone left on a peer's post often beats their own post, because it shows what they care about when nobody is watching their own feed.

- **you:** Hi Priya, your comment under Dana's post about reps sandbagging deals is the exact argument I lost internally last quarter. We rebuilt our comp plan around it. Up for comparing notes?
- **them:** Ha, I remember that thread. We're halfway through changing our plan and it's messy. What did you change?
- **you:** Killed the accelerator and paid on collected revenue instead. Took two months before the team trusted it. I can send the one-pager we used to explain it.
- **draft:** Yes please, send it over.

```
Hi {first name}, your {thing they wrote} on {where you saw it} is the same problem we're working on at {your company}. We {what you changed} and {plain outcome}. Want to compare notes?
```

## Set a pace you can keep

LinkedIn doesn't publish how many invitations an account can send before it gets restricted, and any number changes with account age, acceptance history and how many requests are already sitting pending. Treat every published safe limit as a guess someone made about their own account.

What you can control is consistency. Send in the same block each day, keep your pending invitations low by withdrawing the ones that have gone stale, and watch acceptance rather than volume. A bloated pending pool is what makes new invitations look like spam.

Run your best leads first. If your account does hit a ceiling, you want it to hit after the 90s have gone out, not before.

> Score in the morning, send in the afternoon. Whatever room your account turns out to have, spend it on the people most likely to answer.

## Check the score against what actually happened

The score is a prediction. Acceptance and replies are the result. Keep them side by side for two weeks and you'll see which must-have is doing the work and which one is decoration.

If your 85+ leads accept at the same rate as your 50s, the rubric is measuring the wrong thing. If the 85s accept but never reply, the list is fine and the note is the problem.

| | Signal | What it tells you | What to change |
| --- | --- | --- | --- |
| High-fit leads ignore the request | The note reads like a pitch, or the profile is out of date | Rewrite the opening around their post, not your product |
| High-fit leads accept but go quiet | The note worked, the first message didn't | Open the message with the detail you noticed, not a calendar link |
| Mid-fit leads reply more than high-fit ones | You're scoring titles instead of problems | Weight recent activity higher than job title |
| Acceptance drops across the whole list | You're sending faster than the account has room for | Cut volume for a week and withdraw stale invitations |

## Doing this with Kinlead

Kinlead handles the scoring step. You describe the customer you want or paste your website, and it finds people on LinkedIn who match and scores each one from 0 to 100 against that profile. Leads below your fit line aren't kept, so the group you work from is already the top slice. [Lead groups and scoring](/features/lead-groups-and-scoring) covers how groups and campaigns fit together.

Connection requests can carry an AI-written note under 200 characters that opens with why that person became a lead. Queued messages can be read and edited on the Actions page before they send, and any campaign can be paused.

Outreach runs from your own LinkedIn account, connected through Unipile, and Kinlead keeps each account to one action every 5 minutes inside the invite and message limits you set and a send window in the lead's time zone. Replies from every campaign land in one inbox, classified by intent with interested replies first.

1. **Describe the customer.** Type the profile you want or paste your site. Scoring runs against that description, so be specific about title, size and industry.
2. **Set the fit line.** Leads under the line aren't kept. Move it up if your list feels noisy, down if you're running out of people.
3. **Pick how replies are handled.** Hand off at the first reply, let the AI reply until there's real interest, or let it book meetings on your scheduling link.
4. **Read the queue before it sends.** Open the Actions page, fix anything that sounds off, and pause the campaign if the notes drift away from your voice.

## Related

- [Lead groups and scoring](https://getkinlead.com/features/lead-groups-and-scoring.md)
- [AI LinkedIn outreach](https://getkinlead.com/features/ai-linkedin-outreach.md)
- [Book a demo](https://getkinlead.com/demo.md)

## Common questions

**What do I do with leads that score low?**

Skip them. Every request you send to a 30 is a request you can't send to a 90, and low-fit leads rarely reply in a way that leads anywhere. Send them the short version only if you have room left at the end of the day.

**Should I score before or after I look at their posts?**

After. Recent activity is usually one of your nice-to-haves, and you can't score it without reading the feed. Score last, send first thing the next morning.

**What if a high-fit lead doesn't accept?**

Leave it for two weeks, then interact with a recent post before trying again. If they still don't accept, withdraw the invitation so it isn't clogging your pending list.

**How often should I change the rubric?**

Every few weeks, and only based on what happened after you sent. Changing weights because a batch felt wrong teaches you nothing.

[Create a Kinlead account](https://getkinlead.com/signup) or [book a call](https://getkinlead.com/demo) to talk through pricing and setup.
