Lead scoring: how to score leads and prioritise sales
When enquiries pile up, someone has to decide who gets called first. Lead scoring answers that with points: every contact gains or loses based on who they are and what they do, and anyone above a threshold gets attention straight away.
Key takeaways
- Lead scoring works on profile (who the contact is) and behaviour (how much interest they show).
- Start simple: six to ten criteria, one threshold, a monthly review.
- Include negative points for signals of low likelihood.
- Scoring people is profiling under GDPR; design for that from the start.
What lead scoring is
A method for ranking contacts by how likely they are to buy. It combines two dimensions. Profile measures how closely the contact resembles your ideal customer. Behaviour measures how strong their interest is. A director in your target sector who never opens an email, and a student who downloads everything, each score high on one dimension. Neither is a priority.
What it is for
- Prioritising attention: strong interest gets a fast reply.
- Aligning marketing and sales: both agree on what a sales-ready contact is.
- Improving campaigns: you see which channels bring contacts that actually progress.
- Automating: nurture sequences for people who are not ready yet.
It works best once you have reasonable volume and a defined sales funnel. With ten enquiries a month, just call them all.
The criteria
| Type | Criterion | Example points |
|---|---|---|
| Profile | Decision-making role (B2B) | +15 |
| Profile | Right company size or sector | +10 |
| Profile | Inside the area you serve | +10 |
| Behaviour | Requested a quote or meeting | +30 |
| Behaviour | Visited the pricing or services page | +10 |
| Behaviour | Opened and clicked emails | +5 |
| Behaviour | Replied on WhatsApp | +10 |
| Negative | Outside your area or profile | −15 |
| Negative | No activity in 60 days | −10 |
Swipe the table sideways to see all columns.
These points are illustrative. Your model has to reflect your business.
Building the model
- Define what a sale is and when a lead is ready for sales.
- Review your recent customers: what did they have in common, and what did they do before buying?
- Pick six to ten criteria across profile and behaviour.
- Assign points, weighting actions close to purchase more heavily.
- Add negative points and let old activity decay.
- Set the threshold at which the contact passes to sales.
- Agree the response: who calls, within what time, and what happens if there is no answer.
- Run it for a month and adjust.
A spreadsheet is enough to start. As volume grows, a CRM calculates the score and triggers the actions.
Worked example: a language school (illustrative)
Someone downloads a placement test (+10), lives in the school's city (+10), visits the pricing page (+10) and asks a question on WhatsApp (+10): 40 points. With a threshold of 35, sales take it the same day. Another person downloads the test from abroad (+10, −15) and never returns: −5. They get the newsletter only, if they consented to it.
Checking the model works
- Close rate by score band: higher-scoring contacts should buy more often. If they do not, the criteria are wrong.
- Response time for contacts above the threshold.
- Share of contacts above the threshold: if it is nearly all or nearly none, move the threshold.
- Sales feedback: if the "hot" leads are not hot, revise.
GDPR and scoring people
Lead scoring processes personal data to evaluate aspects of a person, which is profiling. The European guidelines on automated individual decision-making and profiling, endorsed by the EDPB, set out that this processing must satisfy the GDPR's general principles. Article 22 also gives people the right not to be subject to a decision based solely on automated processing that produces legal effects or similarly significantly affects them.
Ordering sales calls rarely reaches that bar, but you should still:
- say in your privacy notice that you profile for commercial purposes;
- have a clear lawful basis for the processing;
- use only the data you need and avoid special categories;
- honour access, objection and erasure requests;
- keep a human in any decision that affects the customer, such as terms or pricing.
This is general information, not legal advice. Check what applies in your jurisdiction.
Common mistakes
- Models with dozens of criteria nobody maintains.
- Scoring behaviour only and ignoring profile.
- Never subtracting points.
- Setting the threshold once and never revisiting it.
- Not agreeing with sales what happens above the threshold.
Frequently asked questions
Do I need a CRM for lead scoring?
Not to start. A spreadsheet with criteria and points works. A CRM automates the calculation and the follow-up actions.
What is the right threshold?
There is no universal number. Start with an estimate and adjust by comparing the scores of people who actually bought.
How many criteria should I use?
Six to ten is usually enough. Long models are hard to maintain and hard to explain.
Does it suit small businesses?
Yes, if there is a follow-up routine. Even a simple model helps decide who to call first.
Can I score people using personal data?
Yes, within GDPR: clear purpose, data minimisation, transparency, security and respect for data subject rights.
Sources
About this article. Produced by the MetaEX team with AI assistance, from the official documentation listed under Sources. Examples marked as illustrative are not client data. Last updated on 11 November 2026.
Want a read on your digital operation?
Talk to us: we first understand your situation, then tell you what makes sense.
Book a meeting