Which leads should you call first?

The uncomfortable finding underneath this question is that the ordering rule most businesses use — call the ones that look best — is usually worse than calling them in the order they arrived. Not because lead quality is a myth, but because the information available at arrival is thin, and the cost of acting on it is paid in the only currency that reliably matters at that stage, which is minutes.

The decay is steep enough to swamp most sorting decisions. Between the five-minute mark and the thirty-minute mark the odds of qualifying an enquiry fall by more than an order of magnitude, and against a next-day reply the gap is larger again. A scoring step that delays every enquiry by twenty minutes in order to call a slightly better one first has, in most realistic distributions, given up more than it gained — and it has done so invisibly, because the enquiries it deprioritised never appear as losses.

This is not an argument against qualification. It is an argument about where qualification belongs. Deciding who gets a proposal, who gets a site visit, who gets the senior person and who gets a polite decline are all decisions worth making carefully, and all of them are better made after a conversation than before one. What does not survive scrutiny is putting a judgement in front of the first contact, where it costs the most and is based on the least.

There are real exceptions and they are structural rather than score-based. Enquiries with a hard external deadline attached, enquiries from an existing customer, and enquiries arriving through a channel with a contractual response commitment all justify jumping the queue — because in each case the priority comes from a fact about the enquiry rather than from a prediction about the person. A useful test: if the rule requires guessing how likely someone is to buy, it belongs after the call, not before it.

The version of this that actually gets implemented is a routing rule rather than a scoring model. Newest first, with a named owner per channel, a deadline, and an escalation when the deadline passes. It is less interesting than a scoring model and it survives contact with a busy Tuesday, which the scoring model does not.