Fit and intent lead scoring in 2026: the dual model that decides who gets worked

A single lead score hides the one thing reps need to know. "Hot" leads that cannot buy are not hot. They are distractions with a calendar invite. Fit and intent lead scoring fixes that by keeping two numbers apart: fit (can they buy) and intent (are they moving now). This guide is the full model, plus a note at the end on where Chronic runs the same logic for you, so the scoring is not another spreadsheet you maintain by hand.

TL;DR

• Build fit and intent lead scoring as two separate scores. Never mash them into one number.
• Fit = should they buy? Firmographics, technographics, hiring, stack, geo, constraints.
• Intent = are they moving now? Site behavior, engagement, triggers, reply sentiment.
• Gate it: high intent plus bad fit never reaches the front of the queue.
• Add decay, missing-data rules, and tiers like A1, A2, B1 that map to a clear action.
• Run it daily as a prioritized queue, signal-based personalization, and meeting routing.

---

Why dual scoring exists (and why the single number stopped working)

Buyers do their homework without you. Then they arrive with a shortlist and a preference already formed.

• Gartner found 67% of B2B buyers prefer a rep-free buying experience for most of the journey (survey of B2B buyers, reported March 2026), up from 61% the year before. That means fewer obvious hand-raisers and more invisible evaluation. Source: Gartner press release, March 9, 2026.
• 6sense's 2025 research says the pre-contact favorite wins about 80% of the time, and in 95% of deals the winning vendor is already on the Day One shortlist. Source: 6sense 2025 Buyer Experience Report.
• Gartner has also long reported that buyers spend only about 17% of their total buying time meeting with suppliers, split across every vendor in the running. Source: Gartner, the B2B buying journey.

So if your model still rewards "downloaded an ebook," "opened 3 emails," or "visited the careers page once," you are ranking noise. By the time someone looks like a hand-raiser, they have usually already picked a favorite. Scoring's job is to find the right accounts early and act on real evaluation signals.

Dual scoring fixes the core failure: it separates "can buy" from "wants to buy now," so neither one hides behind the other.

---

Define dual scoring (so it stays simple)

Dual scoring = two independent scores with guardrails between them.

Fit score (0-100)

How well does this account match the ICP, the constraints, and the buying reality?

Inputs you can get without a data science team:

• Firmographics: industry, company size, revenue, region
• Technographics: stack, tools installed, integrations
• Hiring: relevant job openings, headcount growth
• Stack signals: competitor tools, complementary tools
• Geo: where you sell, compliance boundaries, language
• Basic constraints: funding stage, security needs, sales-motion fit

Intent score (0-100)

How likely is this account to act soon?

Inputs you can collect or infer:

• Site behavior: pricing page, comparison pages, docs
• Engagement: email replies, clicks, LinkedIn activity
• Trigger events: funding, leadership change, new product launch
• Reply sentiment: "curious," "timing," "budget," "not now"
• Frequency and recency: a behavior counts for more when it is recent

Rule: you never collapse fit and intent into a single score. Sales needs clarity about which lever is high, not a blended number that explains nothing.

---

Build fit and intent lead scoring in a week

No data science. No six-month RevOps overhaul. Just a model you can ship and tighten.

Step 1: lock the ICP, then stop pretending it is "everyone"

You cannot score fit without an ICP.

Minimum ICP fields (write these down):

1. Industries you win in (2-5 max)
2. Size band (employees or revenue)
3. Tech environment (must-have, must-not-have)
4. Geo rules (sell, do not sell)
5. Deal motion (SMB, mid-market, enterprise)
6. Compliance and security assumptions (SOC 2, HIPAA, and the like)

Keep it honest. If your ICP does not describe your last 20 wins, it is fiction, and every score built on it will be too.

---

Step 2: choose fit inputs that correlate with "they can buy"

Fit scoring should not be "nice to know." It should block bad pipeline before it reaches a rep.

Fit input categories (use 3-6 from this list)

1) Firmographics (baseline)

• Industry match
• Employee band match
• Revenue band match
• Ownership or funding stage (if relevant)

2) Technographics (real buying constraints)

Examples:

• "Must use Salesforce" (or HubSpot)
• "Uses Snowflake, Segment, dbt"
• "Runs Shopify Plus"
• "Already uses competitor X"

Technographics work because they predict integration fit, implementation time, and switching friction.

3) Hiring and org signals (timing plus capacity)

Hiring is both fit and intent. Treat it as fit when it reflects capability:

• Hiring SDRs: building outbound muscle
• Hiring RevOps: tooling changes underway
• Hiring security: readiness for enterprise vendors

4) Stack compatibility and deal-motion fit

Examples:

• If you sell to PLG SaaS, "has self-serve signup" matters.
• If you sell to agencies, "lists service offerings" matters.

5) Geo and compliance hard stops

If you cannot sell there, that is a fit kill switch. No debate.

---

Step 3: build intent inputs that correlate with "they are moving"

Intent scoring fails when it is built from vanity engagement. Use actions that imply evaluation.

High-signal intent events (ranked)

1. Pricing page views (especially repeated)
2. Competitor comparison page views
3. Product docs, API, or security page views
4. Demo page views or scheduling-tool opens
5. "Reply with timing" or "send info" responses
6. Trigger events: funding, new VP Sales, tooling migration

If you want a fuller signal list, see the companion piece: The trigger engine: 25 real-time outbound triggers that beat static lists in 2026.

Intent is about recency

A pricing visit two hours ago beats ten ebook downloads last quarter. So every intent model needs recency weighting and decay built in from day one.

---

Step 4: ship a model you can explain in 30 seconds

Stop trying to be perfect. Ship a model sales can predict, then calibrate it.

A practical dual scoring model

Fit score (0-100): a weighted checklist.
Intent score (0-100): weighted events with decay.

Fit scoring example (template)

Fit factor	Points
Industry in ICP	+25
Employee band match	+15
Core tech match (must-have)	+20
Competitor installed (if you replace it)	+10
Hiring for relevant roles	+10
Geo allowed	+20
Hard disqualifier present	set Fit = 0

Hard disqualifiers (examples):

• Geo blocked
• Industry forbidden
• Uses a platform you cannot integrate with
• Too small to afford you

Intent scoring example (template)

Intent event	Base points	Notes
Pricing page view	+25	cap at 2/day
Competitor comparison view	+30	strongest evaluation signal
Security or compliance page view	+20	enterprise motion
Demo page view	+35	near hand-raise
Email reply, positive	+30	sentiment-based
Trigger: new VP Sales	+15	add if within 30 days

Decay rule (simple): for each day since the last intent event, multiply intent by 0.92. That is roughly a 50% drop in 8 to 9 days. Intent rots fast, and the math should reflect it.

If you want a cleaner version, use buckets instead:

• 0-2 days: 100% of points
• 3-7 days: 70%
• 8-14 days: 40%
• 15+ days: 20%

---

Step 5: create tiers sales can run without thinking

Scores are useless if the next action is unclear.

Recommended tiers (A1, A2, B1)

Define tiers by two thresholds, not one.

Example thresholds

• Fit high: 70+
• Fit medium: 50-69
• Intent high: 70+
• Intent medium: 40-69

Tier map

• A1 = high fit + high intent. Action: work it now. Call plus email, fast follow-up.
• A2 = high fit + medium intent. Action: personalized sequence plus trigger watch.
• B1 = medium fit + high intent. Action: qualify quickly with tight messaging, do not burn time.
• C = everything else. Action: nurture, low-touch, or ignore.

This is where fit and intent lead scoring becomes operational instead of theoretical.

---

Guardrails: stop "high intent, bad fit" from stealing rep time

This is the main reason sales ignores scoring. The model keeps lying, so reps stop trusting it.

Guardrail 1: a fit gate on the priority queue

If fit is under 50, the account never appears at the front of the queue, even if intent is 100. Send it instead to marketing nurture, a "disqualified but watching" list, or a junior qualifier if you insist.

Guardrail 2: hard-stop rules

Examples:

• Student, consultant, or competitor email domain: disqualify
• Countries you do not sell into: disqualify
• Size under your minimum: disqualify

A hard stop beats "but they visited pricing" every time.

Guardrail 3: intent caps to prevent spammy inflation

Cap repeated behavior:

• Max pricing views counted per day
• Max email clicks counted per sequence step
• Filter bot traffic (user agent, known IPs, or at least "time on page under 3 seconds")

Guardrail 4: decay, always

No decay turns your model into a museum of old behavior.

---

What to do when data is missing (because it will be)

Data gaps are normal. Your model must degrade gracefully instead of guessing.

Missing fit data

If you cannot confirm a fit input, do not guess. Score it neutral.

• Unknown industry: +0, not -25
• Unknown employee count: +0, not disqualify
• Unknown tech stack: +0, not "bad fit"

Then add a data-completeness label so reps see the tier and the confidence behind it:

• Fit completeness: 0-100%
• Intent completeness: 0-100%

The fix for thin data is enrichment, not optimism. Fill the missing firmographic and technographic fields from a real source before the fit score leans on them.

Missing intent data

With no dedicated intent feed, you can still score intent from email engagement and replies, LinkedIn activity, and trigger events (funding, hiring, job changes). It is weaker, but it still beats treating every MQL as sales-ready.

---

Examples: fit plus intent scoring for three business types

Same framework. Different inputs.

Example 1: B2B SaaS (mid-market)

Fit inputs

• Industry: SaaS, fintech, B2B services
• Employees: 50-500 is the sweet spot
• Tech: must use HubSpot or Salesforce
• Tech: uses Segment or a data warehouse (integration maturity)
• Hiring: RevOps, demand gen, SDR manager
• Geo: US, Canada, UK, AU

Intent inputs

• Pricing plus security page within 7 days
• Viewed integration docs
• Replied with "send details" or "timing is Q2"
• Trigger: a "migrating CRM" job post

Tier behavior

• A1: book a meeting this week, route to an AE
• A2: keep the sequence running until a trigger flips intent to high

If you anchor your actions to real targets, see the numbers-focused companion: Outbound benchmarks in 2026.

---

Example 2: a lead-gen agency selling outbound services

Fit inputs

• Client type: B2B service providers, SaaS, agencies (agencies buy from agencies)
• Offer maturity: a clear niche (the site shows a vertical or case studies)
• Team size: 5-50
• Stack: already uses Instantly, Apollo, HubSpot, or Close
• Geo: English-speaking markets if delivery requires it

Intent inputs

• Viewed case studies plus pricing
• Opened 3+ emails and clicked the booking link
• Trigger: hiring SDRs, posting for a "cold email specialist"
• Reply sentiment: "we need more meetings"

The guardrail that matters Agencies get plenty of "high intent" from underfunded founders. Gate it by minimum budget, minimum team maturity, and proof they actually sell B2B.

---

Example 3: a services firm (consulting, dev shop, security services)

Fit inputs

• Industry alignment (regulated industries if that is your wedge)
• Budget proxy: company size, funding, or recent initiatives
• Tech: uses platforms you specialize in
• Geo and compliance: where your team can deliver

Intent inputs

• Trigger: breach news, a new compliance deadline, a new IT leader
• Viewed security or process pages
• Downloaded a specific service brief (not a newsletter signup)

Routing

• A1 goes to a senior closer fast. Services deals are won or lost on discovery quality.

---

Make dual scoring run the day, not sit in a dashboard

Scoring that lives in a dashboard is decoration. The point is the daily loop it drives.

Step 1: build a daily prioritized queue

Every morning, sales should see:

1. A1 accounts sorted by intent recency
2. A2 accounts sorted by fit, then by recent triggers
3. B1 accounts with a strict "qualify fast" playbook

Step 2: tie personalization to the signal that fired

Personalization is not "Hi {FirstName}." It is opening on the reason the account scored high:

• Competitor page viewed: open with "switching from X"
• Hiring signal: open with "scaling the outbound team"
• Security page viewed: open with "security review pack"

If you want reply-handling rules to pair with this, see: The follow-up engine: 12 reply-handling rules.

Step 3: route meetings so speed wins

Routing by tier:

• A1: route to an AE immediately, SLA under 5 minutes
• A2: route to the SDR who owns the sequence
• B1: route to a qualifier or a fast-disqualify motion

Also route by context:

• Enterprise intent plus security signals: route to an enterprise AE
• Agency buyer: route to an agency specialist
• Services buyer with an urgent trigger: route to a senior consultant

---

Where Chronic fits: it runs this loop for you

The model above is sound. The problem is keeping it alive. Weights drift, intent feeds break, and the daily queue becomes a chore nobody owns. Most teams stitch this across a data tool, a sending tool, and a CRM, then maintain the seams by hand.

Chronic is an autonomous revenue operator, not a dashboard you tend. You give it the ICP, the constraints, and the offer. It finds the accounts, enriches the fit fields so the score runs on real data, watches the intent signals, scores both, applies the fit gate and decay, then works the resulting tiers: writing and sending outreach from managed, warmed mailboxes, handling replies, and booking the meeting. It asks for approval when a decision needs you and stays quiet when it does not. The scoring model is not a spreadsheet you babysit every Friday. It is the operator's standing instruction for who to work and why.

If you are comparing how this differs from a data tool or a CRM, start here:

• Chronic vs Apollo
• Chronic vs HubSpot
• Chronic vs Salesforce

---

FAQ

What is fit and intent lead scoring?

Fit and intent lead scoring is a dual model where fit measures how closely an account matches your ICP, and intent measures how likely it is to buy soon based on behavior and triggers. Keeping them separate stops "bad fit but noisy" accounts from hijacking sales time.

What is the biggest mistake teams make with dual scoring?

They collapse it into one number. A single score hides the problem: high intent can mask bad fit. Sales learns the model lies, then ignores it.

How do I set weights without historical win-loss data?

Start with operator logic. Weight hard constraints highest (geo, size, must-have tech) and evaluation behaviors next (pricing, competitor comparisons, security pages). Then run a two-week calibration: pull 30 A1 accounts, ask "would we actually work these?", and adjust weights until the answer is yes.

How should intent decay work in practice?

Use a simple decay that forces freshness. Either multiply intent by 0.92 per day since the last intent event, or use time buckets (0-2 days = 100%, 3-7 = 70%, 8-14 = 40%, 15+ = 20%). Decay keeps your queue from filling with accounts that were interested last month.

What if I only have contact-level intent, not account-level intent?

Roll it up. Account intent = max(contact intent) plus a bonus when multiple contacts show signals. Buying is a group decision: one person clicking is good, three people clicking is real.

How do I stop reps from cherry-picking leads outside the model?

Make the queue the system of record. Require dispositions on A1 and A2 daily, and route meetings and credit through the scored queue. When pipeline is tied to the model, behavior follows it.

---

Build it this week, then tighten it every Friday

Ship the first version in seven days:

1. Write the ICP constraints.
2. Implement the fit checklist and hard stops.
3. Implement intent events with decay.
4. Publish the tiers (A1, A2, B1) with actions.
5. Launch the daily queue and routing.

Then run a weekly calibration: review 20 scored accounts, adjust one weight, add one guardrail, delete one vanity signal. That is how fit and intent lead scoring becomes a working system instead of a dashboard ornament, whether you run it by hand or let an operator run it for you.