Leads are up but jobs are not. Is that marketing's fault or the phone's?

The chain, and why order matters

Leads become jobs through a sequence, and a failure anywhere upstream makes every downstream number unreadable. If a fifth of calls never get answered, your booking rate is computed on a biased sample and tells you nothing reliable about rep performance.

So the sequence is fixed: answered, bookable, booked, completed. Diagnose in that order, always, and stop at the first step that is clearly broken.

Step one: did the call get answered

Answer rate, abandon rate and average hold time before abandon. Then the same numbers broken out by hour and day of week. Almost every service business has a specific window where coverage fails, usually the first hour of the morning, the lunch hour, or after close.

This is measurable directly from phone system data and it is the most commonly overlooked cause of a lead-to-job gap. A marketing increase that lands on unchanged staffing produces exactly this pattern: more leads, more missed calls, flat jobs.

Step two: was the call ever bookable

Of the calls that were answered, what share were a real prospect for work you perform, in an area you serve. Everything else is spam, vendors, existing customers with a billing question, wrong numbers, and requests for services you do not offer.

A rising share of non-bookable calls following a media change is a media problem, full stop. Broader match types, wider geography or a new audience will do it. Classifying calls this way at full volume rather than by sampling is what AI call analysis is for, and it is the only way to make the distinction without an argument.

Step three: bookable and not booked

Now, and only now, you are looking at handling. Booking rate on bookable calls, by rep and by call reason. Then the reasons: price objection, schedule availability, the caller wanted a quote over the phone, the call ended without a next step.

Schedule availability is a capacity finding, not a rep finding, and it belongs back with operations. Price objections concentrated in one rep are a coaching finding. Calls ending without any attempt to book are the largest and most fixable category in most companies, and they respond directly to rep coaching built on evidence from real calls rather than on manager impressions.

Publish the split, not the blame

Once the chain is instrumented, the monthly conversation changes shape. Instead of marketing and operations trading assertions, each step has a number and an owner. The report should show where the leads went, not who was at fault.

That is the practical value of connecting media data to phone data to job data: the argument becomes a diagnosis. Everything else in marketing intelligence builds on having that chain instrumented.