If you’re tired of sifting through “hot leads” who never reply, or you’re just convinced your CRM is guessing wrong, custom lead scoring is worth your time. This guide is for anyone who wants to actually use Leadmagic’s lead scoring to find real prospects—not just fill dashboards with pretty numbers.
Whether you’re in sales ops, running your own outreach, or just tasked with “making the leads better,” you’ll get a practical, step-by-step walkthrough. No buzzwords, no magic formulas—just the actual process, with honest notes on what’s worth your time (and what’s not).
Why bother with custom lead scoring?
Let’s be real: default lead scoring rarely fits your business. The built-in stuff in most CRMs is generic, sometimes even counterproductive. The real value in Leadmagic is building a system that fits your signals—things that actually predict who’ll close, not just who opens emails.
Custom scoring helps you:
- Waste less time on “leads” who’ll never buy
- Prioritize the outreach that actually gets replies
- Give sales reps a list they trust (and will actually use)
- Avoid the endless debates about what a “good lead” looks like
It takes a little setup, but it saves a lot of headaches.
Before you start: What you need
Don’t dive into the tool yet. You’ll need:
- Access to your Leadmagic admin account
- A clear idea of what makes a lead “good” (honestly, write this down)
- Any extra data sources you want to bring in (CSV lists, CRM fields, etc.)
- Some patience—scoring is part science, part guesswork, and part “we’ll see”
If you don’t know yet what a “good” lead looks like, talk to your best sales rep (or look at your last 10 closed deals). You need at least a rough sketch.
Step 1: Map out your scoring criteria
Before clicking anything, decide what you’ll actually score on. Don’t make this complicated.
Common ways to score leads: - Demographics: Company size, industry, job title, location - Behavior: Website visits, downloads, email opens/clicks - Fit: Do they match your ideal customer profile? - Engagement: Recent replies, meeting booked, etc.
Pro tip: Don’t try to use everything. Pick 3-5 things that actually matter. If you score on 15 signals, you’ll drown in noise and tweak forever.
What to skip: Scoring based on “clicked the newsletter once in 2022” or “followed us on Twitter.” Unless you know it matters, ignore it for now.
Write down your list. Seriously—open a doc, jot it down. You’ll thank yourself later.
Step 2: Log into Leadmagic and find the scoring settings
Now, head into Leadmagic.
- Log in with your admin credentials.
- From the main dashboard, find “Settings” or the gear icon (top right).
- Click “Lead Scoring” in the settings menu.
If you don’t see it, you might not have the right permissions. Ask your admin or check your plan—some features may be locked behind higher tiers.
Step 3: Set up your custom scoring rules
This is where the magic happens (pun only slightly intended).
3.1. Create or edit a scoring model
- Click “Create Scoring Model” (or “Edit” if you’re modifying the default one).
- Give your model a clear name: “2024 Inbound Scoring” beats “Test 2.”
3.2. Add your criteria
For each signal you wrote down, add a rule:
- Pick the field: e.g., “Job Title,” “Company Size,” “Visited Pricing Page”
- Set the condition: e.g., “contains ‘Director’,” “over 100 employees,” “visited in last 7 days”
- Assign a score: e.g., +10, +5, -10
Tips: - Stick to round numbers for now. You’ll tweak these later. - Use negative scores for red flags (like “student” job titles if you’re B2B). - Don’t be afraid to leave some things out. Too many signals = messy results.
3.3. Combine multiple signals
You can (and should) stack rules. For example:
- “VP Title” and “Company Size > 500” = +20
- “Clicked Demo” = +15
- “Generic Email Domain” (like gmail.com) = -10
Most systems let you use AND/OR logic or even “score decay” (lower scores if signals are old). Don’t overthink the logic at first—keep it simple.
3.4. Save and activate
Once your rules are set, hit “Save” and make sure your model is active.
Step 4: Test your scoring model with real data
Don’t trust your model yet. Run it against real leads.
- Pick a batch of leads—recent ones, closed-won, closed-lost.
- See what scores they get.
- Ask yourself: Do the “best” leads actually score high? Are there any surprises?
What usually happens: Some junk leads score way too high, or your best customers barely register. That’s normal on your first pass.
Step 5: Adjust and tune your scoring
No model is perfect out of the gate. Use these tips:
- Look for “false positives”: Are there junk leads with high scores? Figure out why.
- Watch for “false negatives”: Are good leads getting low scores? What signal are you missing?
- Tweak the points: Don’t be afraid to bump scores up or down by 5-10 points at a time.
- Remove useless criteria: If you realize “Downloaded Whitepaper” never actually closes, drop it.
Pro tip: Revisit your model every month for the first quarter. After that, once a quarter is fine.
Step 6: Set up alerts and automations (optional, but handy)
Lead scoring isn’t much good if nobody sees the results.
- Set up alerts for hot leads (e.g., score > 70)
- Create workflows: auto-assign top leads to reps, trigger emails, etc.
- Build simple dashboards: Let sales see their best leads at a glance
But don’t go nuts. Automate just enough to help your team, not so much that leads get spammed or ignored.
Step 7: Train your team (and get feedback)
The best scoring model in the world is useless if nobody trusts it.
- Walk your sales team through how it works (keep it simple)
- Ask for honest feedback: “Are the best leads actually at the top of your list now?”
- Adjust if needed—don’t get defensive, just fix what’s broken
Skip: Don’t make a 20-slide deck. Just show a couple of real leads and scores.
What actually works (and what doesn’t)
What works:
- Using real sales data, not “gut feel,” to pick your criteria
- Keeping your model simple at first, then adding complexity only if needed
- Regularly reviewing and updating your scores
What doesn’t:
- Blindly copying someone else’s scoring model
- Overcomplicating things with too many signals or fancy logic
- Setting it and forgetting it—lead scoring isn’t “one and done”
If you’re not sure if something matters, leave it out. You can always add it later.
Summary: Keep it simple, tweak as you go
Custom lead scoring in Leadmagic isn’t rocket science, but it does take some up-front work and a bit of trial and error. Start simple, focus on what actually predicts a good lead for your business, and don’t be afraid to adjust as you go. The best models are the ones you keep tuning—not the ones you set and forget.
And remember: Any lead scoring is better than letting your CRM guess. Keep it practical, and your sales team will thank you.