Early access Montr BI opens to teams in India, the US and the GCC in waves. Get your spot →

Sales

Lead scoring without data science: a 20-minute method

Build a simple lead score from fit and interest using a spreadsheet or your CRM. No models, no data team. Includes a scoring table you can copy.

25 Sep 2026 · 5 min read · Montr team

Lead scoring without data science: a 20-minute method

Lead scoring sounds like something for big companies with data teams. It is not. At its simplest, it is a short list of rules that tell your team who to call first.

You can build a working score in 20 minutes, with a spreadsheet and your own judgment.

Why bother

You have 40 new leads this week and time to call 10. Which 10?

Without a score, people call whoever is easiest or whoever came in last. With a score, they call the leads most likely to buy. The score does not need to be right every time. It needs to beat guessing.

Two questions, two scores

Every good score answers two things.

  1. Fit. Is this the kind of customer you serve well?
  2. Interest. Are they acting like someone who wants to buy soon?

Score each on its own. A perfect fit who never replies is a slow lead. An eager lead who cannot afford you is a waste of time.

Step 1: write down your best customers

Open your last 20 customers. For each, note:

  • What did they buy?
  • How big were they?
  • Where did they come from?
  • How long did they take to decide?

Look for repeats. If 15 of 20 came from a referral, that is a signal. If most are in two cities, that is a signal. Use your own data, not a guess.

Step 2: score fit

Pick three to five fit signals and give each a number of points. Here is an example for a business that sells to local companies.

Signal Points
In your service area +20
Company size in your sweet spot +15
Right job title (owner, manager) +10
Wrong industry -20
Free email address and no company name -5

Keep numbers small and easy to add. Round to 5s and 10s.

Step 3: score interest

Interest is behaviour. Pick actions that mean something.

Action Points
Asked for a price or a quote +25
Replied to your message +15
Visited your pricing page +10
Opened your email +2
No reply after 3 messages -15

Be careful with opens. Email apps preload images, so opens are noisy. Treat them as a weak signal at most.

Step 4: set the line

Add fit and interest. Then split into three groups:

  • Hot (say 60 or more): a person contacts them today.
  • Warm (30 to 59): automatic follow-up, then a person.
  • Cold (under 30): nurture, check again in a month.

Pick your lines by looking at last month’s leads. Would the score have put your actual buyers in “hot”? If not, move the numbers.

Step 5: test it on old leads

Take 50 old leads where you know the outcome. Score them. Check:

  • Did most won deals land in hot or warm?
  • Did most lost or silent leads land in cold?

If yes, you are close. If no, adjust. Do this twice and stop. Perfection is not the goal.

Step 6: put it where people will see it

A score in a spreadsheet nobody opens does not help. Put it on the lead record, next to the name. Sort the list by score. Make “hot” show up first on a Monday morning.

A worked example

Priya runs a small interior design studio. Her last 20 clients were mostly homeowners within 30 km, with a clear budget, who asked for a site visit in the first message.

Her rules:

Signal Points
Within 30 km +20
Mentions a budget range +15
Asks for a site visit +25
Replied within a day +10
Looking for “ideas only” -15

A lead who is nearby, shares a budget and asks for a visit scores 60. That is hot, so Priya calls today. A lead from another city who wants ideas scores 10 or less. That one gets a helpful email and a check-in next month.

After a month, Priya notices that “asks for a site visit” predicts almost every project she wins. She raises its points. That is how a simple score improves.

How often to revisit

Check your score every quarter, and after any big change in your offer, pricing or audience. Look at the last 30 won deals and ask whether the score ranked them well. If it did not, adjust two or three numbers, not the whole model.

Mistakes to avoid

  • Too many rules. Past about eight signals, nobody can explain a score, and nobody trusts it.
  • Scoring on things you cannot check. If you cannot see a signal reliably, leave it out.
  • Mixing fit and interest into one number too early. Keep them separate while you test. Combine them later.
  • Never calling the cold ones. A cold lead who is a great fit might need a different message, not silence.
  • Treating the score as the truth. It is a sorting tool. A person still decides.

Keep it honest

  • Review quarterly. Your market changes. So should your rules.
  • Do not score on things you should not use. Avoid anything that could treat people unfairly, such as guesses about age or background from a name.
  • Do not let the score override a conversation. If a “cold” lead calls and says “I need this by Friday”, treat them as hot.

Doing it in Montr

Montr BI does not have a built-in scoring model, and we would rather say that than invent one. What it does have is the pieces to build your own:

  • Custom fields on contacts, such as a “score” number and a “temperature” label.
  • Saved views with filters, so you can see hot leads at the top.
  • Automation triggers when a field changes, a tag is added or a deal stage moves. You can add points or tags when someone replies, and alert a salesperson when a lead crosses your line.

Read the lead scoring definition, start with the CRM, then build the rule in workflow automation. The first workflow guide shows the steps.

The short version

Score fit and interest separately. Use small numbers. Test on old leads. Put the score where people look. Review it every quarter.

Put your growth on autopilot. Keep your hand on the wheel.

Join the waitlist. We open Montr BI to teams in India, the US and the GCC in small groups, and we set up your first workflows with you.