Account scoring for B2B SaaS: firmographic, technographic, and growth signals

Account scoring is the dimension most B2B teams underweight, and it’s also the one with the highest predictive power for SaaS sales.

The reason: B2B SaaS purchases happen at the company level, not the contact level. A perfect-fit person at the wrong-size company doesn’t become a customer. A whole buying committee at the right-size company often does, even if individual contacts are imperfect. Scoring that ignores the company in favor of the person systematically underrates the leads that actually close.

This post is the third in a series on the four dimensions of lead scoring. The first two covered engagement and profile fit, the person you sell to. This one covers account fit: the company. We’ll walk through what each signal category contains, how to weight them, and how to keep account scoring from becoming a stale third-party-data layer.

What account scoring measures

Account scoring asks “is this the right company to sell to?” The signals fall into four categories.

Firmographic. The basic facts about the company.

• Company size (headcount or revenue)
• Industry / vertical
• Geographic location
• Company type (private, public, PE-backed, agency, etc.)

Technographic. What’s in their tech stack.

• Uses platforms your product integrates with (HubSpot, Salesforce, Snowflake, Slack, etc.)
• Uses platforms your product replaces or competes with
• Has a stack signal that suggests technical sophistication (Kubernetes, dbt, modern data stack indicators)

Growth signals. Indicators of momentum or change.

• Recent funding round (typically Seed → Series C is the sweet spot for most B2B SaaS)
• Hiring spike, especially in functions that buy your product
• Recent product launch
• Leadership change (new CRO, new VP of Marketing, etc.)
• M&A activity

Engagement at account level. Multiple contacts from the same company engaging.

• Multiple contacts at the account in your CRM
• Multiple contacts engaged in the last 30 days
• Contacts spanning different functions (the “buying committee” forming)

How to weight by category

For most B2B SaaS:

Firmographic should anchor the score. Company size, industry, and geography are the strongest non-engagement signals of fit. They typically deserve the heaviest weights in account scoring.

Technographic is often the highest-leverage signal. If your product integrates with HubSpot, the fact that the lead’s company runs HubSpot is more predictive than most other fit signals. Technographic data can meaningfully improve conversion in tight ICP segments. Don’t underweight it.

Growth signals are predictive but noisy. A recent funding round is real signal but the timing rarely aligns with your buying window. Score modestly. Hiring spikes in target functions are stronger than generic growth signals.

Account-level engagement deserves its own treatment. Multiple contacts engaged from the same account is a near-binary signal that something is happening, and it should crank account fit score meaningfully. The buying committee is forming.

The negative signals worth scoring

Account fit is the dimension where negative scoring has the highest payoff, and where most teams skip it entirely.

Worth scoring negative:

• Company size below your minimum ICP band (−15 to −25)
• Company size above your maximum ICP band, if you’re not enterprise-ready (−10)
• Geographic location you can’t serve (−15 to −20)
• Industry you’ve explicitly disqualified (e.g., government, healthcare for non-HIPAA tools) (−15)
• Competitor companies (−25)
• Existing customer accounts (−999, effectively block from the funnel)

Without negative account scoring, leads from disqualified companies still rack up engagement points and end up “Hot” despite having zero chance of closing. This is the same reason engagement scoring needs negative points and decay: without a way to move down, every score eventually climbs into the red zone and stops meaning anything.

Where account scoring goes wrong

A few common failure modes:

Static firmographics. Company size doesn’t update on its own in HubSpot. A startup that grew from 20 to 400 employees over 18 months is still “small” in your CRM unless you refresh enrichment. Schedule re-enrichment quarterly, or use a continuous enrichment provider.

Industry mapping mismatch. HubSpot’s default Industry property is messy. Lots of companies appear under “Marketing & Advertising” who actually sell SaaS, security, etc. Build a cleaner Industry property with controlled values, or trust enrichment over self-reported industry.

Single-source enrichment. If you enrich exclusively from one provider (say, Clearbit), you inherit all of their data quality issues. Cross-reference with at least one other source for high-stakes scoring decisions, especially for company size.

Treating tech stack as a static field. Tech stack changes. The most predictive technographic signals come from continuously updated sources (HG Insights, BuiltWith, Wappalyzer). Annual snapshots are quickly stale.

Ignoring buying committee signals. A score that doesn’t reward “multiple contacts from this account engaged this month” misses the strongest leading indicator of an active buying process.

How to validate

The validation loop for account scoring:

1. Pull every closed-won deal from the last 90 days.
2. Look at the company-level signals at the time of MQL conversion (size, industry, tech stack, growth).
3. Compute the average account-fit score for closed-won accounts.
4. Compare to the average account-fit score across your overall MQL pool.
5. Identify which specific signals over- or under-correlated with close.

The most common finding from this exercise is that one or two specific technographic signals are doing 80% of the predictive work, and several other signals you’ve been weighting are doing nothing. Tune accordingly.

Account scoring in HubSpot

HubSpot’s lead scoring tool supports scoring on the Company object as well as the Contact object. You can build a Company-level Engagement score, Fit score, or Combined score (Marketing Hub Enterprise, since combining engagement and fit into one score isn’t available on Professional). The score lives on the Company record, not the Contact.

The constraint: the Company-level score doesn’t automatically flow to Contact-level scoring without configuration. Most teams either replicate fit signals at both levels (introducing redundancy) or live with the gap (losing the signal).

The cleaner pattern is to score company-level signals on the Company object and contact-level signals on the Contact, then combine them at the dashboard or alert level rather than baking the combination into a single property.

Where this fits

Account fit is one of four dimensions in a complete lead scoring framework. The others: profile fit (the person), engagement (recent activity), and deal scoring (the opportunity itself). Account fit alone is a partial picture; a perfect-fit account with a wrong-fit contact is still not a buyer. Composing the four into a play your sales team can route on is covered in the series wrap-up.

A note on tooling

kenbun scores accounts and contacts as separate dimensions, supports negative rules natively, and exposes the account-fit score per signal so you can see which firmographic, technographic, or growth indicators are contributing. See it on your HubSpot data.