Every business with more enquiries than sales time faces the same problem: which prospects should the sales team call first? Answering by gut feeling means the loudest enquiries get attention, not the most promising ones. Answering by arrival order means a casual browser who filled in a form five minutes ago outranks a serious buyer who enquired yesterday. Lead scoring solves this by giving each prospect a number based on evidence: who they are, what they have done, and how closely they resemble past customers who actually bought. The number is not magic, and it is not a prediction carved in stone. It is a disciplined way of ranking, so that limited sales time goes to the prospects most likely to be worth it.
The Lead Scoring Calculator helps you build and test that ranking. It lets you assign points to criteria and see how different prospects score. The calculator cannot tell you which criteria matter for your business; that knowledge comes from your own history of who bought and who did not. The guide below explains how to build the system so that the numbers mean something, and how to keep it honest over time.
Two kinds of signal: fit and behaviour
Lead scoring traditionally combines two kinds of signal. Fit describes who the prospect is: their company size, industry, location, role, and budget range. A prospect who matches your ideal customer profile scores high on fit, because similar customers have bought before. Behaviour describes what the prospect has done: visited your pricing page, downloaded a guide, attended a webinar, replied to an email, requested a demo. Behaviour signals interest and timing. A prospect who requested a demo yesterday is signalling readiness in a way that a prospect who downloaded a guide six months ago is not.
Both kinds matter, and neither is enough alone. A perfect-fit prospect who has shown no interest may not be ready to buy, and may resent a hard sales call. A highly active prospect who is a poor fit, a student researching a project, a competitor browsing your site, may consume sales time without ever becoming a customer. A good scoring system combines the two, so that the top of the list holds prospects who both fit and have shown intent. Many teams set a threshold: above a certain score, the lead goes to sales immediately; below it, the lead stays in marketing for further nurturing until its behaviour score rises.
Negative scoring is just as important and is often forgotten. A prospect who unsubscribed, who works for a company far outside your market, who entered a clearly fake email address, or who visited only your careers page should lose points. Without negative scoring, junk leads accumulate high scores through harmless activity, and the sales team learns to distrust the system. Every point a lead can gain should have a corresponding way to lose points for signals that argue against a sale.
A worked example
Suppose you sell accounting software to small businesses. You decide on a simple system out of 100 points. Fit is worth up to 50: the prospect's company has 5 to 50 employees (20 points), operates in a country you serve (15 points), and the contact is a finance decision-maker rather than a junior employee (15 points). Behaviour is worth up to 50: visiting the pricing page (10 points), downloading a guide (5 points), attending a webinar (10 points), and requesting a demo (25 points). Negative signals subtract: unsubscribing removes 30 points, and a company with over 500 employees, outside your target market, removes 20.
Now consider three leads. Lead A is a finance director at a 20-person firm in your market who requested a demo: 50 fit points plus 25 behaviour points, 75 in total, and the demo request alone justifies an immediate call. Lead B is a junior employee at a 10-person firm who downloaded a guide and attended a webinar: 35 fit points plus 15 behaviour points, 50 in total, worth nurturing with more content but not yet a sales call. Lead C is a student who visited the pricing page five times and downloaded everything: low fit, moderate behaviour, perhaps 25 points, and clearly not a sales priority despite the activity. The ranking matches what an experienced salesperson would conclude, but it reaches the conclusion consistently, for every lead, without anyone having to read each enquiry in full. That consistency is the system's real value.
Set the sales threshold by looking at your history, not by guessing. If past customers typically scored above 65 at the point they bought, a threshold around 65 is a reasonable starting point. Then watch what happens: if sales complains that leads above the threshold are poor, the weights are wrong, not the threshold. If good leads sit below the threshold for weeks, the weights are wrong in the other direction. A scoring system is a hypothesis about your customers, and like any hypothesis, it should be revised when the evidence disagrees with it.
Common mistakes
The first mistake is scoring everything. If you assign points to twenty criteria, the score becomes noise, because small differences on irrelevant criteria swamp the important signals. Start with five to eight criteria that you genuinely believe distinguish buyers from browsers, and add more only when the data shows they help. Simplicity is not a compromise here. It is what makes the system understandable to the sales team, and a system the sales team does not understand is a system they will ignore.
The second mistake is never updating the weights. Markets change, products change, and the signals that predicted buying last year may not predict it this year. Review the system quarterly: compare the scores of leads that became customers with those that did not, and adjust the weights toward the criteria that actually separated them. If a criterion no longer separates buyers from non-buyers, remove it, even if it felt important when you chose it.
The third mistake is treating the score as a verdict rather than a ranking. A lead scoring 80 is not guaranteed to buy, and a lead scoring 40 is not guaranteed to ignore you. Scores rank priorities; they do not predict individuals. Salespeople should still use judgement on individual leads, and the system should support that judgement, not replace it. When a salesperson overrides the score with a good reason, record the reason. Those overrides are valuable data for the next revision of the weights.
The fourth mistake is building the system in isolation from sales. If marketing designs the scoring alone, sales will distrust it, and rightly so, because they are the ones who live with the consequences. Build it together: ask sales which signals they have found meaningful, agree the threshold jointly, and review the results together. Shared ownership is what turns a calculator output into a tool the team actually uses.
Honest limitations
Lead scoring works best when you have enough history to learn from. A brand-new business with ten past customers cannot build a reliable system from data, and should start with a simple, judgement-based version, then refine it as the customer base grows. Scoring also inherits the biases in your history: if your past customers skew toward one industry because that is where you happened to market, the system will keep favouring that industry, potentially blinding you to new markets. Review the criteria for that kind of self-reinforcing loop, and deliberately test leads from outside the usual profile from time to time.
Privacy matters too. Score only the data prospects have given you or that you legitimately hold, be transparent about how you use it, and respect the data protection rules that apply in your market. A scoring system built on data you should not have is a liability, not an asset.
Closing
Combine fit and behaviour, include negative signals, start simple, set the threshold from your history, and revise quarterly with sales at the table. To compare the cost of acquiring leads against their value, use the ROI Calculator, and to understand the speed at which your team follows up, try the Email Response Time Calculator. For the advertising metrics behind lead costs, read CPM CPC and CPA Calculator Guide.