Lead Scoring
Definition
Lead scoring is the systematic methodology for assigning numerical values to leads based on their likelihood to buy, using a combination of demographic fit (firmographics, title, industry) and behavioral signals (content engagement, website activity, product usage). In PE-backed B2B companies, lead scoring is the mechanism that determines which leads receive sales attention and which continue in automated nurture — making it one of the highest-leverage points of marketing-sales alignment in the entire GTM engine.
Why It Matters
Sales capacity is finite and expensive. In a PE portfolio company operating under growth mandates with margin pressure, every hour a rep spends on an unqualified lead is an hour not spent on a deal that could close. Lead scoring exists to solve this allocation problem — to route the right leads to the right reps at the right time, and to keep everything else in marketing's nurture programs until the signals say otherwise.
The problem is that most lead scoring implementations are broken. They were built three years ago by someone who no longer works at the company, they have never been validated against actual close data, and they produce scores that sales ignores. A lead scoring model that sales does not trust is worse than no model at all, because it creates the illusion of a functioning handoff process while leads fall through the cracks or get routed to the wrong stage.
For PE operating teams, lead scoring is a diagnostic tool as much as an operational one. How a portfolio company scores leads reveals how well they understand their buyer, how tightly marketing and sales are aligned, and whether the GTM engine has the instrumentation to improve over time. A company with a well-tuned, regularly validated scoring model is a company that can scale efficiently. A company with a broken or nonexistent scoring model will burn sales capacity as it grows.
What to Look For
Dual-axis scoring: fit plus engagement. Effective lead scoring separates demographic fit (right company, right title, right industry) from behavioral engagement (visited pricing page, downloaded comparison guide, attended demo webinar). A lead can be a perfect fit with zero engagement, or highly engaged but a poor fit. Both dimensions need to be scored independently to produce useful routing decisions.
Regular model validation against closed-won data. The scoring model should be validated at least quarterly by comparing scores at the time of MQL against eventual outcomes (closed-won, closed-lost, disqualified). If the company has never back-tested their scoring model, they are guessing — and the guesses degrade over time as the market and buyer behavior evolve.
Clear threshold definitions and SLA structure. There should be explicit score thresholds that trigger routing to sales, with defined SLAs for follow-up timing. If the threshold is arbitrary or if sales has no obligation to work scored leads within a defined window, the scoring model is decorative.
Negative scoring and score decay. Mature models subtract points for disqualifying signals (competitor employee, student email, no engagement in 90 days). Without negative scoring and time-based decay, stale leads accumulate at artificially high scores and pollute the pipeline.
Red Flags
- The lead scoring model was built once and has never been validated against actual sales outcomes
- Sales reps openly ignore lead scores and cherry-pick leads based on their own criteria — indicating zero trust in the model
- Scoring is based entirely on demographic data with no behavioral component, or vice versa
- There is no score decay mechanism, so leads that engaged heavily two years ago still show high scores today
- The scoring model does not differentiate between content consumption (reading a blog post) and buying signals (visiting pricing, requesting a demo) — all engagement is weighted equally