Which channels produced booked electrical work?

Two channels producing calls at $60 each are not equivalent if one fills the diary with small repairs and the other with panel upgrades, EV chargers and generator installs.

Ranking on completed job revenue is the correction, and it usually reveals that the channel with the highest cost per call has the lowest cost per dollar earned.

A commercial account can be worth a hundred residential calls and arrives through completely different channels — often a referral or a bid list rather than search.

A job-type column mapped as Product / service shows residential and commercial revenue as separate lines, with how much of each traces to a channel, and referred commercial clients can go into the lead file, tagged with their referrer, so the commercial side is measurable at all.

An electrician's day contains work worth a hundred dollars and work worth eight thousand, and both arrive through the same brand and frequently the same campaign. A fault-finding call, a panel upgrade, a rewire, an EV charger installation and a generator are different products with different cycles sold by the same van.

Reported as a single job count, the cheap and fast work dominates every metric, because there is simply more of it. The channels that produce the occasional large project look expensive and unproductive on cost per lead, and they are the ones carrying the margin.

Ranking on invoiced revenue, with a job-type column mapped as Product / service, separates them, so each half's revenue is seen on its own rather than inside an average that describes neither half of the business.

A great deal of an electrician's project work is sold in somebody's house by a technician who went there to do something else. The homeowner called about a dead socket and ends up scheduling a panel upgrade, because somebody they now trust told them the panel was the real problem.

That sequence is one of the most profitable things the trade does and no marketing report credits it, because the advertisement is scored against the eighty-dollar call it produced.

Because the match is on the customer rather than the job, the later project invoice credits the channel that produced the original call. Channels that generate service volume stop looking like a low-margin commodity once the work they lead to is attached to them, which on most books reorders the list.

Scale changes what the same report is good for. With two vans and a few hundred jobs a year, one large commercial rewire can move a channel's revenue share by ten points, so the ranking is noisy and the sensible reading is directional: which sources produce paid work at all, and which have quietly produced none for two quarters. Precision is not available at that volume, and pretending otherwise invites expensive reactions to ordinary variance. The decision it supports is whether to keep paying for a listing, not how to weight two campaigns against each other.

At thirty vans the opposite applies. The volume is high enough that the ranking holds reasonably still from month to month, and the valuable cuts are the ones a small shop cannot make: revenue by service area and by job type. An import carries two breakdowns — a territory mapped as Location and a job type or technician mapped as Product / service — and a single average across a dozen service areas describes none of them, while the budget is being spent area by area rather than as one national number.

Both sizes share one trap, which is reading the attributed share as a performance score. A small contractor whose work is half word of mouth will see a large Direct and Unknown bucket and take it as failure; a large one running heavy paid media will see a small one and take it as success. Neither reading holds. The bucket describes how a business acquires customers, not how well its marketing works, and the figure worth watching is the revenue behind each named channel as it moves over several periods.