Lead scoring: how to know who to call first

Lead scoring is a simple way to rank your leads so your team knows who to call first. Each lead gets a score based on two things: how closely they match your ideal customer, and how much buying intent they have shown. The highest scores get a call today, and everyone else gets the kind of follow up that suits where they are.

The idea matters even more in a small business, where the same two or three people answer enquiries, serve customers and keep the business running. Every call spent on a poor fit lead is a call not made to a great one.

What is lead scoring, in plain terms?

Lead scoring turns your best salesperson’s instinct into a rule everyone can follow. Instead of working through enquiries in the order they arrived, the team works through them in order of likely value.

Picture a solar installation company on a Monday morning. Over the weekend, enquiries came in from the website, a Facebook lead form and WhatsApp. One homeowner in the service area uploaded a recent electricity bill and asked about installation dates. Another person filled in a general form with no address. A third is a student researching a school project. All three count as leads. Only one needs a call before lunch.

With a handful of leads, anyone can sort this in their head. With a steady flow, several channels and more than one person on the team, gut feeling breaks down. A score keeps everyone working from the same priorities, and a low score still gets an answer: what changes is the order and the type of follow up.

What signals should a lead score include?

Most lead scoring models combine fit signals (who the lead is) and engagement signals (what the lead does), plus a few negative signals that pull the score down.

Fit: are they the kind of customer you want?

Fit is about how closely a lead matches the customers who actually buy from you. Typical fit signals include location within your service area, the type of service or product they need, and the size of the job. For B2B companies, the person’s role and the size of their company matter too. Much of this is explicit data, meaning things the lead tells you directly in a form or a chat.

Engagement: how serious are they right now?

Engagement is about behaviour, sometimes called implicit data. Requesting a call, answering your WhatsApp bot’s questions, returning to your services page or opening a proposal all show intent. Timing matters as well. Someone who needs the service this month is not the same as someone who says “maybe next year”.

Negative signals: what should lower a score?

A lead outside your service area, an invalid phone number, a job applicant who used the sales form, or a lead who has not responded to anything for a long time. A score should fade if nothing happens, otherwise last season’s hot lead stays at the top of the list forever.

A simple model might look like this:

Signal	Type	Weight
Asked for a call or a quote	Engagement	High
In your service area	Fit	High
Wants to start soon	Engagement	High
Answered the bot’s questions	Engagement	Medium
Clicked a link in an email	Engagement	Low
Outside your service area	Fit	Negative

How do you build a lead scoring model from scratch?

Start from real deals rather than theory, choose a handful of signals, and improve the model as you learn. A simple model the team understands beats a sophisticated one nobody trusts.

1. Study your won deals. Pull your recent customers and a few deals you lost. Where did the winners come from, what did they ask, and how quickly did they want to start?
2. Pick a handful of signals. Choose the ones that repeat and are easy to check automatically. If someone has to research a lead to know whether a signal applies, leave it out for now.
3. Weight each signal. High, medium, low or negative is enough to begin with.
4. Define tiers. Hot, warm and cold, or A, B and C. What matters is that the line between tiers is clear and everyone reads it the same way.
5. Attach an action to each tier. Hot leads trigger an instant alert and a same day call. Warm leads get a helpful WhatsApp message and a follow up task. Cold leads join a lead nurturing sequence until they are ready.
6. Review and adjust. At first, sit down every month with the people who talk to customers. Were the hot leads really hot? Did good leads slip through with low scores? Change the weights accordingly.

Lead scoring examples from small businesses

A dental clinic

A patient clicks an Instagram ad and starts a WhatsApp chat. The bot asks which treatment they want, whether they are a new patient and when they would like to come in. Someone who picks a treatment and asks for an appointment this week scores high, and the front desk gets an alert to call. Someone just comparing options receives clear information and an easy way to book later.

An IT services company selling to other businesses

Fit carries a lot of weight here: company size, location and the role of the person who enquired. An office manager at a company of the right size who requested a quote and visited the managed services page several times goes straight to the top of the list. A student downloading a guide stays in the nurturing flow.

How does lead scoring work inside a CRM?

In a CRM, scores are calculated automatically from everything the system already knows about a lead, and they update whenever something new happens. Nobody has to add up points or remember who looked at what.

When your website forms, WhatsApp bot, ad lead forms and phone calls all feed the same system, each lead’s record collects the details: what they filled in, how they answered the bot and when someone last spoke to them. The CRM applies your rules and updates the score. When a lead crosses into the hot tier, the right person gets a notification, and the day’s call list is sorted by score. For the bigger picture of what should happen from the moment an enquiry lands, see our guide to lead management.

This is where a CRM built around your business earns its place. The fields, stages and weights follow the way you actually sell, not a generic template: a clinic scores by treatment and timing, an IT firm by company size and role. Many off the shelf tools include some form of scoring too, so the real question is whether their model fits your sales process.

AI can add a useful layer. It can summarise a long WhatsApp conversation, spot intent such as “ready to start” or “just checking”, and suggest a score. Fully predictive scoring usually needs a lot of past data, so most small businesses start with clear, transparent rules.

Finally, a score only helps if hot leads get a fast answer, which is why lead scoring and speed to lead belong together.

What are the most common lead scoring mistakes?

Most problems come from a model that is too complicated, never tested against reality, or never maintained.

• Too many criteria. With dozens of rules, nobody can explain a score, and the team stops trusting it.
• Points for every click. If each email open adds points, everyone looks hot within weeks.
• Scores that never fade. Leads who went quiet months ago should drop down the list.
• Ignoring cold leads. A low score means “not yet”, not “never”. Change the conversation instead of deleting the contact.
• Building it without sales. The people who speak to customers know things the data does not show. Without them, the model only looks good on paper.

Quick checklist

• You know what your recent won deals had in common
• Every lead from every channel lands in the same CRM
• Your forms and bot ask the questions you need for scoring
• Each tier has a clear action and an owner
• Scores fade when a lead goes quiet
• There is a regular review with the people who sell

Start simple, then sharpen

Lead scoring does not need a data science team, just a few clear signals, sensible weights and an action attached to every score. At Kaede Digital we build scoring into a custom CRM, together with the bot questions, form fields, rules and alerts that feed it, so your team starts every morning knowing exactly who to call first.

FAQ

What is the difference between lead scoring and lead qualification?

Lead qualification decides whether a lead is relevant at all, for example whether they are in your service area and need what you sell. Lead scoring ranks the relevant leads by how likely they are to buy and how soon. In practice, qualification questions are usually part of the scoring model.

Do I need a lot of leads for lead scoring to be worth it?

Not necessarily. Even a business with a modest flow of enquiries benefits from clear priorities, especially when the same person sells and delivers the service. The more leads and the more people on the team, the more value scoring adds.

Can AI score leads for me?

AI can help a lot, for example by reading a WhatsApp conversation and spotting intent or urgency. It works best inside clear rules your team understands, rather than as a black box. That way everyone knows why a lead got its score.

How often should I update my lead scoring model?

Review it regularly: monthly at the start, then every quarter once it is stable. A new service, a new area or a campaign that brings in a different audience is also a good reason to check it again.