Cost per lead has not moved but cost per booked job is up. Where do I look?
Cost per booked job equals cost per lead divided by booking rate. If cost per lead is flat and cost per booked job rose, booking rate fell. That narrows the search to four candidates: source mix shifted toward a weaker source, call handling degraded, demand type changed, or scheduling capacity tightened. Check them in that order, because the first two are the cheapest to verify.
Start with the identity, not the dashboard
Cost per booked job = cost per lead / booking rate. That is not an approximation, it is an identity. Any movement in the output has to come from one of the two inputs. When one input is flat, you already know where the change lives, and you can stop reading channel reports.
Four candidates, ordered by how fast you can rule them out
- Mix shift. Each source held its own booking rate, but a source that books poorly grew as a share of total leads. Recompute the blend with last quarter's source weights. If the problem disappears, it is mix, not performance.
- Handling. Answer rate, hold time, after-hours coverage or rep behavior changed. Look at unanswered and abandoned calls before anything else; they are the most common cause and the easiest to fix.
- Demand type. Same source, same volume, different callers. More price shopping, more out-of-area, more calls about work you do not do. This shows up as a rising share of calls that were never bookable.
- Capacity. Callers who wanted to book could not be given an acceptable appointment window. Booking rate falls for reasons marketing cannot touch.
The measurement that separates handling from demand
You need a bookable-call classification: of the calls that arrived, which ones were a real prospect for work you actually perform, in your area, at a time you could serve. Booking rate on total calls mixes demand quality and rep performance into one number and settles no arguments.
Booking rate on bookable calls only is a performance metric. The share of calls that were bookable is a media quality metric. Splitting the two turns a monthly blame conversation into two separate, fixable problems. AI call analysis produces the classification at full volume rather than from a hand-sampled dozen calls.
What to do with the answer
If it is mix, the fix is budget allocation and you should be comparing sources on cost per booked job rather than cost per lead in the first place. If it is handling, it is a staffing and coaching problem, and media changes will only mask it.
If it is demand type, tighten targeting, negatives and geography. If it is capacity, adding budget makes the number worse, not better, and the honest move is to slow spend until schedule availability recovers.
Topics: booking rate · diagnostics · cost per booked job · call handling
Have a version of this question about your own business?
The useful answer usually depends on which systems you run and how they're connected. That's a conversation, not a blog post.