Your Lead Score Is One Number Doing Two Jobs

Most lead scoring models hand the sales team a single number between 0 and 100, and that number is quietly answering two completely different questions at once. One is whether this lead is the right kind of buyer. The other is whether they're ready to buy right now. When you collapse both into one score, a research analyst poking around your pricing page can outrank a VP at your perfect-fit account — and your best SDR spends Tuesday calling the analyst.

Scoring decides who your reps don't call

A B2B SaaS team generating 1,000 marketing leads a month doesn't have a lead problem — it has a capacity problem. Five SDRs running maybe 15 real conversations each per week can work somewhere around 300 of those 1,000 leads with any seriousness. The score's only honest job is to pick which 300. Everything below the line gets an email and silence. So the model isn't ranking interest so much as rationing the single most expensive resource you have, which is a human being's attention — and most teams have never looked at it that way.

Why one number breaks

Behavioral points are easy to rack up, and firmographic points are capped. A lead earns +20 for a pricing-page visit, +15 for watching a demo video, +10 for three sessions in a week — intent signals compound, because a motivated browser generates a lot of them. Fit signals don't: company size, industry, and job title are a fixed handful of attributes worth maybe 30 points total. So the top of a single-score list skews hard toward whoever clicked the most, not whoever looks like your actual customers. The competitor running a teardown and the student writing a paper both look electric in your analytics. The CFO at a 2,000-person account who downloaded one whitepaper and went quiet looks cold.

Split the score into two axes

The fix isn't a smarter formula. It's refusing to add the two questions together in the first place.

Fit answers: does this account look like the ones that actually close and stay? Build it from your CRM, not your gut — pull two years of closed-won and weight the firmographics that correlate with retention, not just with signing. An industry that closes fast and churns in six months should score lower, not higher.

Intent answers: is this person showing buying behavior right now? And "now" is load-bearing — intent is perishable. A pricing-page visit from 60 days ago isn't intent; it's history. Score intent on a decay curve so a signal loses most of its weight within two or three weeks.

Then plot every lead on a 2x2 — fit on one axis, intent on the other — and four quadrants fall out, each with its own play:

• High fit, high intent — Priority. Roughly 90 leads a month in our 1,000-lead example. Lead-to-opportunity around 22%, opportunity-to-close around 25% — call it 5% lead-to-close. These get an SDR call within the hour; speed-to-lead is worth more here than anywhere else.

• High fit, low intent — Nurture. The biggest quadrant you actually care about, maybe 260 a month. They'll buy, just not this week. An SDR cold-calling them now converts at maybe 3%. Put them in a fit-based nurture track and let an intent trigger, not a calendar, tell sales when to step in.

• Low fit, high intent — Triage. Maybe 220 a month, and this is the trap. They generate beautiful engagement and close at around 1%. Route them to self-serve and your PLG flow; don't spend a human on them.

• Low fit, low intent — Hold. The remaining 430 or so. Newsletter, nothing more.

The reallocation, in arithmetic

Run the single-score model and the top 300 are mostly whoever generated the most clicks: about 90 Priority leads and about 210 from the low-fit, high-intent quadrant. Your reps spend roughly 70% of the week on a pool that closes at 1%. The yield is about 90 × 5% + 210 × 1% ≈ 4.5 + 2.1 = 6.6 deals — and 210 SDR conversations burned to produce two of them.

But route by quadrant instead, and the week looks different. Reps work all 90 Priority leads hard and fast, plus the high-fit leads whose intent has genuinely crossed the line — say 40 a month graduate out of Nurture when their decayed intent score spikes. That's 130 leads worked, not 300. The deal count barely moves, because the Priority quadrant was always where the deals lived. What changes is that you just freed more than half your SDR capacity. That capacity doesn't vanish — it goes into faster follow-up on Priority, into working the nurture list the moment triggers fire, and into the open opportunities your reps were too buried to advance.

The single score never changed who was going to buy. It only changed who your team spent the day chasing.

What to do this week

Pull last quarter's closed-won and closed-lost, and tag each deal with where the lead sat at hand-off: fit high or low, intent high or low. You'll almost certainly find most revenue came from one quadrant and most SDR hours went to another. That gap is the entire argument — you won't need a slide to make it.

Then make two changes to the model. Cap intent's contribution so it can't outvote fit: a workable starting rule is to gate, not sum — a lead needs to clear a real fit floor before its intent points count toward routing at all. And add decay to every behavioral signal, so the score reflects this week's buying behavior rather than last quarter's.

Lead scoring will never tell you who will buy. Done right, it tells your five reps which 130 conversations are worth having before Friday — and that was the only thing you were ever really asking it for.