Lead scoring for small businesses: sort hot, warm and cold leads

Lead scoring gives every inquiry a number based on what it tells you, so you know which one to answer first. For a small service business, a rule-based score works well: start each lead at a base number, add points for signs of a real buyer (a budget that fits, a date you can do, full contact details), take points away for warning signs, then sort leads into hot, warm and cold. You can build it in an afternoon and improve it from your own results.

Why a small business needs a score at all

When ten inquiries arrive on a Monday, you can't answer all of them in the first hour. A score decides the order, and the order matters, because speed wins work.

The classic study here is a 2011 Harvard Business Review audit of 2,241 US companies by Oldroyd, McElheran and Elkington, which found that firms replying within an hour were far more likely to qualify a lead than those who waited. Response times haven't improved much since. In a 2026 mystery-shopper test, Workato sent demo requests to 114 B2B companies: only one sent a personal email within five minutes, and the average email reply took 11 hours and 54 minutes.

You don't need to answer everything in five minutes. You need to answer the right ones first. We covered the reply side in why the first reply wins the job. Scoring is the part that tells you which reply to send first.

Rule-based scoring vs AI scoring

Large sales teams use predictive scoring, where software learns from thousands of past deals. A business that books 5 to 50 jobs a month doesn't have that much history, and a model trained on too little data guesses.

Rules work better at this size for three reasons:

• You can read them. "Budget fits: +15" is something you can check and argue with.
• They work from day one. No training period.
• They match how you already think. You already know a wedding inquiry for next Saturday is less likely than one for next June.

Salesforce's own guide to lead scoring recommends this kind of method for teams starting out: find your baseline rate of leads that become customers, then give points to the attributes that close above that baseline. It also recommends negative scores for spam, generic email addresses and places you don't serve.

The signals that predict a booking

Most service inquiries carry the same handful of clues. Group them so no single clue swamps the rest.

Can you reach them? A real phone number and an email address that works are good signs. A lead with no way to reply is close to worthless.

Did they tell you enough? A complete form with a detailed message beats a one-line "how much?".

Does the budget fit? If you ask for a budget range, a range that covers your price is one of the strongest signals there is. A range well below your minimum is one of the strongest negatives.

Does the date work? A date inside your booking window is good. A date in the past, or one so close you can't deliver, is not.

How did they find you? Repeat clients and referrals close at higher rates than strangers from a search ad.

Buying words. "Book", "deposit", "contract", "start date" in a message usually mean someone is past browsing.

These match the old BANT checklist (budget, authority, need, timeline) that most sales teams learn first. HubSpot's write-up of BANT and its newer BANTER version adds engagement and review, which in a small business mostly means "did they answer my questions?".

A simple points table to start with

Start every lead at 40 and adjust. These numbers are a starting point, not a rule. Change them once you see your own results.

Then set bands. A practical split: 70 or more is hot, 45 to 69 is warm and under 45 is cold. Hot leads get a reply the same hour. Warm leads get one the same day. Cold leads get an automatic reply with your prices and a booking link, and a person only if they answer.

Spam is its own band

Spam isn't a cold lead. It's noise, and it should never trigger your follow-up emails. Mark these down hard: messages pitching you SEO or development services, people looking for a job, messages stuffed with links and names made of random letters. A form with a bot check that fails is a strong sign too.

Keep spam visible in a separate filter for a few weeks so you can rescue the odd real lead the rules caught. Then trust it.

Check your scores against what actually closed

A score is a guess until you test it. Every month, look at the leads you won and lost:

1. Did hot leads close more often than warm, and warm more than cold? If not, your weights are off.
2. Which signal shows up in most wins? Give it more points.
3. Which signal shows up in wins and losses alike? It isn't telling you anything. Cut it to a few points.
4. What did you lose that scored hot? Write down why. That reason may be a signal you're missing (we cover how in our post on inquiry form questions that filter out bad-fit leads).

Clean data makes this work. In Salesforce's State of Sales 2026, 79% of high-performing sales professionals said they prioritise data hygiene, against 54% of underperformers. For a small business that just means recording the outcome on every lead, every time.

How Delivvo scores every inquiry the moment it arrives

Delivvo scores each new inquiry by rules you can read, so you open your lead board and answer the hot ones first. There's no AI in it and nothing to train.

Here is what happens when someone fills in your inquiry form:

1. The lead lands on your board with its source (form, booking page, import) and an owner if you work in a team.
2. Delivvo works out a score from 0 to 100. It starts at 40 and adds or takes points for contact details, how complete the form is, whether the budget fits, whether the date sits in your window, buying words, repeat clients and referrals.
3. The lead gets a label: Hot, Warm, Cold or Spam. Hot starts at 70 and Warm at 45. Spam signals (sales pitches, job requests, link stuffing, random names, a failed bot check) push a lead into Spam.
4. The reasons are on the lead page. You see each signal and its points, so you know why a lead is hot, not only that it is.
5. Automations can act on it. Rules can send an auto reply with your price list and booking link, or assign the lead, and rules never run on Spam.

On Business and up you can set your own weights, so the score follows what closes for you. The person who owns each lead is always clear, and a lead that waits too long can be flagged to a manager on Team and up. See the board on the lead board page.

FAQ