How it works
A scoring model adds or subtracts points based on two kinds of signals:
- Fit: facts about the lead, such as location, company size, service requested or project type.
- Behavior: what they’ve done, such as replying quickly, asking for a quote or opening several messages.
Rules might look like: +20 for being inside the service area, +15 for asking about a specific date, −30 for asking about a service you don’t offer. Leads above a threshold go to the top of the queue or to a senior rep.
Some tools use statistical or machine-learning models instead of hand-written rules. Either way, the output is a number, and it is only as good as the data and assumptions behind it.
Example
A roofing company gets dozens of inquiries a week across SMS and WhatsApp. It sets simple rules:
- Inside the three-county service area: +20
- Mentions a leak or storm damage: +25
- Homeowner, not a renter: +15
- Asking only for a price with no address: −10
A message reading “Storm took shingles off, water coming into the attic, I own the house in Elm Grove” scores high and is answered first. A vague “how much for a roof?” still gets a reply, just after the urgent ones.
Lead scoring vs. lead qualification
- Lead scoring ranks many leads with a number, often automatically and before a real conversation.
- Lead qualification is a person’s decision about one lead, usually after asking questions.
Scoring helps with prioritizing. Qualification helps with deciding. A small team handling a modest number of inquiries may get more value from consistent qualification questions than from a scoring model.
Why it matters for messaging teams
In chat, the strongest signals are in the words the customer wrote. That makes it tempting to let software score leads from message text. Treat any automated score as a suggestion that a person checks, because tone, sarcasm and missing context can mislead a model.
Common mistakes
- Scoring before you have enough data to know which signals matter.
- Never revisiting the rules after your services or market change.
- Using a low score as a reason not to reply at all.
- Trusting an opaque score nobody on the team can explain.
inChat does not offer autonomous lead scoring. Its AI lead qualification is an in-development concept designed to suggest qualification that a person confirms.