A definition you can quote, the calculation with weights derived from outcomes, an illustrative worked example, and why fit and intent belong on separate axes.
Free Forever • No Credit Card Required
Quick answer
A fit score measures how closely a lead or account resembles the customers you serve best, using stable attributes rather than anything the lead has done.
It is deliberately blind to behaviour. The question is not whether this account is interested; it is whether they would be a good customer if they were. Keeping those two questions apart is the entire reason a separate fit score exists.
Fit score = Σ (attribute weight × match value), normalised to a readable scale such as nought to one hundred.
Firmographic: industry, employee count, revenue band, location. Technographic: systems already in use that make you relevant or irrelevant. Role: the seniority and function of the individual contact where the score is applied to a lead rather than an account.
Binary, meaning the attribute matches or it does not, or graded, so a company just outside your target size band earns partial credit rather than nothing. Graded matching is usually more informative and slightly more work to maintain.
Derived from your own closed-won, closed-lost, and retention history. Attributes that separate winners from losers earn more weight. Attributes everybody assumes matter frequently separate nothing, and finding that out is one of the more valuable outputs of the exercise.
Hard constraints handled outside the weighting: an unsupported country, a prohibited industry, a company below a viable size. These should exclude an account from the scored population rather than merely reduce its total.
These weights are invented to demonstrate the arithmetic. They are not recommended values.
Attributes and weights: target industry 30, employee count within band 25, decision-making role 20, country served 15, relevant existing system 10. Total available weight is 100, which makes the normalisation trivial and the score directly readable as a percentage of ideal fit.
An account matches target industry, giving 30. It matches employee count, giving 25. The contact is a departmental manager rather than the budget holder, which the graded scale treats as half credit, giving 10. The country is served, giving 15. No relevant existing system is recorded, giving 0.
Fit score = 30 + 25 + 10 + 15 + 0 = 80 out of 100. The denominator here is total available weight, so an 80 means this account matches four fifths of what your best customers look like. Note the missing attribute: it scored zero, but zero and unknown are different things, and a system that cannot distinguish them will understate accounts whose data is simply incomplete.
Fit score drives allocation. It decides which accounts deserve outbound effort, which belong on a target list, which get senior attention, and which are served perfectly well by self-serve signup and documentation.
It also disciplines marketing. A campaign producing volume with a low average fit score is producing work rather than pipeline, and the fit distribution of leads by source is one of the few marketing measures that sales will trust instinctively.
The most consequential error. A combined score cannot distinguish a perfect account that is not looking from a poor account that is very active, and those two require opposite responses.
Workshop-derived weights encode what the team believes about its market. Some of that belief will be right and some will be a decade out of date, and only outcome data separates them.
Scoring only on conversion rewards attributes that predict a fast yes. Some of those same attributes predict a fast cancellation, and a fit model tuned purely to closing will systematically recruit churn.
A score computed from a minority of known attributes carries the same visual authority as a complete one. Reporting completeness alongside it is the cheapest available correction.
Products move upmarket or downmarket, new segments appear, and a fit model that has not been revisited quietly keeps recruiting last year's customer.
A useful fit model has weights traceable to outcome data, includes retention in its derivation, treats hard constraints as disqualifiers, reports completeness, and shows a visible difference in win rate and retention between high-fit and low-fit accounts. Sales can see the contributing attributes and correct them when they are wrong.
A weak one is a set of round numbers agreed in a meeting, applied to sparse data, combined with behaviour into a single total, and never tested against what actually happened to the accounts it rated.
| Situation | What it means | The right response |
|---|---|---|
| High fit, high intent | A good account that is actively looking | Contact immediately with a named owner |
| High fit, low intent | A good account that is not in market yet | Nurture, target list, and outbound over time |
| Low fit, high intent | An active lead you cannot serve well | Self-serve, or decline politely and early |
| Low fit, low intent | Neither suitable nor interested | No allocated selling attention at all |
| Unknown fit, any intent | Too little data to judge suitability | Enrich or ask, then score before allocating |
Store the fit score on the account record with the contributing attributes visible, and let representatives correct attribute data directly. Fit is stable enough that it can be calculated at creation and refreshed occasionally, which makes it far cheaper to maintain than behavioural scoring and, for allocation decisions, frequently more useful.
Continue with lead management software, AI CRM capabilities, what a CRM is, sales automation, CRM for small business, and product features.
Every team in this space loses revenue to the same recurring gaps. Here is what they cost you and how HelloGrowthCRM closes each one.
Fit and intent were combined into one score, and nobody can tell a good account from an active one.
Keep them as two axes and act on the combination. A high-fit, high-intent account needs contacting today. A high-fit, low-intent account belongs in nurture. A low-fit, high-intent lead is usually a distraction. One combined number makes those three indistinguishable and routes them identically.Two axes, not one number
The scoring weights were set in a workshop and reflect what the team believes rather than what happens.
Derive weights from closed-won and closed-lost history, and include retention alongside conversion. Attributes that everyone assumes matter frequently show no separation at all, while an unglamorous attribute such as team size or an existing system often does most of the predictive work.Weights from outcome history
Most accounts have half their attributes missing and the score is computed anyway.
Report data completeness alongside the score, or withhold a score below a minimum number of known attributes. A confident-looking figure derived from two fields out of eight will be trusted exactly as much as a well-supported one, which is precisely the problem.Completeness reported with score
An account in an unsupported country scored well because it matched every other attribute.
Treat hard constraints as disqualifiers rather than as negative points. Unsupported geography, a prohibited industry, or a company below a viable size should exclude an account from the scored population entirely, because no amount of matching elsewhere makes them serviceable.Disqualifiers separate from weights
AI-powered CRM with the features you need to close more deals.
Common questions about using HelloGrowthCRM in your industry.
Join small businesses that close more deals with HelloGrowthCRM.
Free Forever • No Credit Card Required
Free Forever • No Credit Card Required
Prefer email? Write to sales@hellogrowthcrm.com