Which ads produced work worth doing?

Emergency search is where most plumbing ad budget goes, and it produces the most calls. It also produces the smallest average ticket, because someone with water on the floor wants the cheapest fix that works today.

The jobs that pay — repipes, sewer lines, water heater replacements, commercial contracts — often come from different searches and convert more slowly. Measured on call volume they look like poor performers, which is how budget ends up concentrated on the lowest-margin work available.

Lead resellers and directories all report the leads they sent. Matching those lists against your own invoiced jobs produces a single number per source that sums to the work you actually did — rather than several vendor reports that each claim the same customer.

Because only exact phone or email matches count automatically, a source claiming a customer it never sent has nothing to stand on.

A burst pipe at eleven at night and a bathroom repipe booked three weeks out arrive through the same brand and often the same campaign. One converts in minutes at a modest ticket; the other is quoted, compared and decided over weeks at many times the value.

Cost per lead treats them identically and therefore systematically favors emergency keywords, which are cheap to convert and cheap to serve. That is a real part of the business and it is not where the margin is.

Ranking on invoiced revenue already puts the money where it is. A job-type column mapped as Product / service makes the split explicit: emergency and planned revenue each get their own line, with how much of each traces to a channel, rather than a blended average that describes neither.

Plumbing is a repeat trade. A household that calls once and is treated well calls again, and eventually asks for the water heater, the repipe or the bathroom. That sequence is worth a multiple of the first call and is credited to nobody.

Because the match is on the customer, every later invoice credits the channel that produced them the first time. Channels that bring households rather than one-off emergencies stop looking equivalent, and on most books that reorders the list.

The unattributed share tends to be large, and it should be. Plumbing is a trade where a neighbor's recommendation, a van seen in the street and a twelve-year-old relationship genuinely produce work, and none of that can be traced to an advertisement. Reporting it as Direct and Unknown rather than distributing it across the paid channels is what makes the rest of the table believable.

The second common finding is that emergency keywords dominate the job count and underperform on revenue per job, while planned-work keywords look expensive per lead and carry the margin. Neither of those is a reason to switch a campaign off; both are reasons to bid and budget for them separately, which almost nobody does because almost nobody has ever seen the two numbers side by side.