Why it matters
Two things quietly drain Meta budgets: changing things too often, and changing them for the wrong reasons. Big edits to an ad set, like a sharp budget jump or a new audience, can send it back into the learning phase. That’s the period where Meta’s system is still figuring out who responds, and results tend to be less stable while it lasts.
Local service budgets are usually modest, and daily results bounce around. You might get three leads Monday and none Tuesday. If you react to every swing, nothing runs long enough for you to know whether it’s actually good.
Pacing matters too. You want spending spread evenly across the month, so you aren’t out of budget by the 20th or buried in leads during a week your crew is already booked solid.
How we do it
- Agree on what success means. Before we test anything, we decide what counts: cost per lead, lead quality, and, where your CRM data allows, which leads became paying jobs. Clicks and likes aren’t the scoreboard.
- Test one variable at a time. We test meaningful differences, like a before and after photo against a crew video, or an instant form against a landing page, instead of tiny tweaks. Changing one thing at a time keeps results readable, and Meta’s built-in A/B testing tool can split audiences cleanly when a test calls for it.
- Give tests enough time and data. We let ads gather a fair number of results before calling a winner. With smaller local budgets that can take a while, and we’ll tell you plainly when a result is still too early to read.
- Pick the right budget setup. We choose between daily and lifetime budgets, set either at the campaign level (Advantage+ campaign budget) or for each ad set, depending on how much control a campaign needs. Meta may spend a bit more on some days and less on others while keeping the overall total in line, so we judge pacing across the week, not day by day.
- Adjust budgets gradually. When something works, we raise the budget in steady steps instead of big jumps that can reset learning. When it doesn’t, we look for the cause before cutting it off. We also adjust for your season and your schedule, easing off when you’re full and leaning in before a slow stretch.
- Review weekly, report monthly. Every week we check spend, frequency, cost per lead and lead quality, and make the changes the data supports. Your plain-English monthly report covers calls, forms, cost per lead, what we tested, and what’s changing next.
What you get
- A testing plan that spells out what we’re testing, why, and how we’ll judge it
- Budget setup and pacing matched to your monthly ad spend and your busy and slow seasons
- Weekly optimization based on leads and jobs, not vanity metrics
- A running log of tests and results, so nothing gets retested by accident
- A monthly report with calls, forms, cost per lead and next steps
- Direct access to the person running your ads when your schedule or capacity changes
Common mistakes we fix
- Editing every day. Constant tweaks keep ad sets stuck in the learning phase and make results impossible to read.
- Calling tests too early. A couple of leads in the first few days isn’t enough to pick a winner.
- Testing too many things at once. A new audience, new creative and new form all launched together leave you guessing about what made the difference.
- Big budget jumps. Doubling spend overnight can shake up delivery and send costs swinging while the system adjusts.
- Judging by cost per lead alone. Cheap leads that never answer the phone aren’t a bargain, so we look at which leads actually turn into jobs whenever we can.