More contractor campaigns are killed prematurely than are left running too long. A fortnight of disappointing numbers feels like evidence, and usually is not.
Wait for enough conversions, not enough days
The question is never “has it been long enough” but “have enough people converted for this to mean anything”.
A campaign producing three bookings a week tells you very little after two weeks and quite a lot after two months. One producing thirty a week is readable far sooner. Judge by the count of outcomes, not the calendar, and be aware that below roughly a couple of dozen conversions you are mostly reading noise.
Match the window to the sales cycle
Contractor sales cycles differ enormously by service, and a single lookback window will misread at least one of them.
- Emergency work converts the same day. A week of data is meaningful.
- Replacements take a fortnight or more: quotes, financing, a decision. Judging those on a seven-day window shows you spend without the bookings it produced.
- Large projects can run a month or longer.
If your replacement campaign looks unprofitable, check whether your reporting window is shorter than your sales cycle before touching it.
Do not judge a seasonal campaign in the wrong season
An HVAC campaign assessed in a shoulder month is telling you about the month. So is a roofing campaign assessed in a quiet spell for storms.
Compare like with like, this spring against last spring, rather than against the peak you have just come out of.
Before concluding the campaign is the problem
Four things fail more often than the campaign does, and all four look identical in the reporting:
- Nobody answered. Check the answer rate first. This is the most common cause of a campaign that “does not work”.
- Calls are not being counted. If conversions only track forms, the campaign may be performing while appearing not to.
- The landing page is wrong for the search. Emergency traffic arriving on a general services page converts poorly no matter how good the targeting is.
- The job mix is wrong. The campaign may be booking work you do not want, which reads as poor performance and is actually poor targeting.
Work through those before switching anything off. Killing a campaign that was being let down by the phone simply moves the problem to the next channel.
Scale in steps, not leaps
When something is working, the temptation is to double the budget. That usually degrades performance: large jumps push the system into less qualified inventory and reset whatever it had learned.
Raise by a fifth or so, let it settle for a week or two, then check cost per booked job before raising again. If efficiency holds, continue. If it slips, you have found your ceiling, which is useful information rather than a failure.
And scale against capacity, not just profitability. Bookings you cannot service become cancellations and complaints, which is a worse outcome than the spend you saved.
Pause rather than delete
Pausing keeps the history: the conversion data, the learning, the search term record. Deleting throws it away, and restarting later means beginning from nothing.
Pause when you are booked out, when a season has turned, or while you fix something the campaign depends on. Reserve stopping entirely for campaigns that have had a fair test, on enough data, with the four checks above ruled out, and still do not pay.
Frequently asked questions
How long before I can judge a PPC campaign?
Count conversions rather than days. Below roughly a couple of dozen you are mostly reading noise, so a campaign producing three bookings a week needs months, while one producing thirty needs weeks.
My campaign looks unprofitable. What should I check first?
Your call answer rate, whether calls are counted as conversions at all, whether the landing page matches the search, and whether the campaign is booking work you actually want. All four look like campaign failure in the reporting.
How fast should I increase budget?
By around a fifth at a time, with a week or two to settle between raises. Doubling pushes the system into less qualified inventory and discards what it had learned.
Should I pause or delete an underperforming campaign?
Pause. It keeps the conversion history, the learning and the search term record. Deleting means starting from nothing if you return to it.
Where to start
Before your next decision to switch something off, count the conversions it has actually produced. If the answer is under a couple of dozen, you do not yet have a result; you have a hunch with a spend figure attached.
Real Time Marketing has worked with home service contractors since 2016: plumbing, HVAC, roofing, electrical and other trades, nationwide, from our office in Bradenton, Florida. Agreements are month-to-month. See our pay-per-click management or book a strategy call.
