Most lead scoring models in use today are built around firmographic fit: company size, industry vertical, job title, annual revenue. If a company checks enough boxes against your ICP, the lead scores high. If it doesn't, it scores low. That logic is sound as far as it goes — but it stops at the wrong question. Fit tells you whether a company could buy from you. It tells you nothing about whether they're ready to buy right now.
This is the gap that breaks most scoring implementations within the first year. RevOps teams invest weeks configuring the model, reps trust it for a quarter, and then someone starts noticing: the highest-scored leads aren't converting any better than the mid-tier ones. The model isn't broken — it's just answering the wrong question.
What Fit Scoring Actually Measures
A firmographic fit score is a static snapshot. A company with 150 employees in B2B SaaS with a VP of Sales who matches your buyer persona — that company will score 85 regardless of whether they visited your pricing page this morning or haven't touched your site in eight months. The data inputs to a fit score change slowly: company headcount, industry classification, technographic stack. They're useful for filtering out clearly wrong-fit prospects, but they create no meaningful signal about urgency.
The problem compounds when you have a large inbound volume. Consider a scenario like this: a growing logistics software company runs a content campaign that generates 200 form fills in a week. Every single one of those leads gets a fit score based on demographics. Reps work the list roughly in order of that score. But three of those 200 leads had spent 18 minutes reading your pricing page and compared your tool against two competitors in the same browser session — and they're buried at position 47, 112, and 163 because their company happened to be 40 employees instead of 75. Someone who matches your ICP perfectly but has zero urgency gets called first.
That's not a calibration problem. That's a structural gap in what the model is designed to measure.
Intent Signals: What They Are and What They're Not
Behavioral intent signals capture what a prospect is actively doing right now. The most reliable ones within your own data are: pages visited (especially high-intent pages like pricing, integrations, or case studies), time-on-page, return visit frequency, email engagement patterns (opens, link clicks, which links), and demo or trial requests. Third-party intent data — where a vendor aggregates content consumption signals across their network — adds another layer, though it comes with variable quality and significant latency.
We're not saying intent signals are infallible or that fit scoring is useless. The real argument is that neither works well alone. A company with perfect fit but no behavioral signals is a prospect worth nurturing — not worth prioritizing in today's call queue. A company showing high behavioral urgency but weak fit is worth a quick qualifying call, not an extended sequence. The combination is what creates a meaningful priority signal.
The industry term for combining these two axes is sometimes called a fit-intent matrix: segment your leads into quadrants based on where they fall on each dimension. High fit + high intent = call immediately. High fit + low intent = nurture with value content. Low fit + high intent = quick disqualification call. Low fit + low intent = don't call.
The 30-Minute Window Problem
There's a well-established pattern in B2B inbound sales where contact rates drop sharply after the first 30 minutes following a form submission. The exact numbers vary by industry and deal size, but the directional reality is consistent: a prospect who just submitted a demo request is at peak engagement in that window. They haven't yet moved on to the next task. They're likely still in the mental context of the problem they were trying to solve when they found you.
A fit-only scoring model running on a weekly batch job — which is how many CRM scoring implementations are configured — cannot serve this window. By the time scores update and route to reps, the urgency window has already closed for the highest-intent leads. Even a model running overnight batch updates is too slow for same-day inbound.
Real-time intent scoring changes the operational model. When a prospect hits your pricing page for the second time in three days, that event should trigger a score update immediately — not be aggregated into next Monday's batch run. For RevOps teams, this means the architecture of the scoring system matters as much as the logic inside it. A perfectly calibrated model running on stale data will still miss the window.
Where Fit Scoring Earns Its Place
None of this means firms should abandon firmographic scoring. It still serves two critical functions. First, it's the right filter for top-of-funnel qualification — before any behavioral signals exist, fit scoring helps route leads to the right sequences or disqualify them from sales attention entirely. A lead from a company with 12 employees applying to a tool priced for 50+ seat teams probably shouldn't enter the sales queue at all, regardless of intent.
Second, fit scoring is the right calibration layer for outbound prospecting, where by definition no inbound behavioral signals exist. When reps are building targeted lists for cold outreach, they need ICP alignment as the primary prioritization axis because intent data simply isn't available. This distinction — fit is primary for outbound, intent becomes decisive for inbound — is something many organizations miss when they build a single unified scoring model for all lead types.
Practical Steps for Adding Intent to an Existing Model
For most RevOps teams, the path forward isn't ripping out the fit model — it's building an intent layer on top of it. The sequence that tends to work:
Start by auditing which behavioral events you already capture. Most CRMs have some activity data: email opens, link clicks, form fills. Map those events to an intent score, even a simple one. Pricing page visit = +15. Email click = +5. Returned within 72 hours = +10. The weights don't need to be perfect on day one — they need to exist and be applied in real time.
Then combine the two scores into a composite. The simplest approach is a weighted blend — for example, 50% fit / 50% intent for inbound leads, adjusted over time as you observe which combinations actually correlate with conversion. Over a quarter, you'll have enough closed/lost data to recalibrate.
The single most important operational change: stop accepting batch-update architectures for inbound scoring. If your system only re-scores leads once a day, the intent component is largely useless for time-sensitive inbound. The scoring update needs to happen within minutes of a triggering event — not hours, and certainly not days.
What a Combined Score Actually Looks Like in Practice
Picture a lead that came in from a content download three weeks ago. At that point, they scored 72 on fit (right company size, right industry, relevant title) and near zero on intent. They got routed into a nurture sequence. Two days ago they came back, spent 11 minutes on your pricing page, clicked through to your integrations page, and then submitted a contact form. Their composite score is now 91 — fit 72 unchanged, intent now 89 based on the session activity plus the contact form submission. They move to the top of the call queue with a reason code that tells the rep exactly what happened: "Returned after 21 days, read pricing + integrations, submitted contact form."
That rep doesn't need to guess what angle to take on the call. The context is already there. That's the difference between a fit-only score that routes a lead into a queue and a fit-intent composite that tells a rep what's happening and when to act on it.
The shift from static fit scoring to real-time composite scoring is less about model sophistication than it is about asking the right question. Fit answers: could they buy? Intent answers: are they trying to? Both questions are useful. Only one of them tells you who to call first today.