Campaign budget optimisation
Setting one budget at campaign level and letting Meta decide how it splits across the ad sets beneath it.
Instead of ten ad sets each with their own budget, the campaign holds one and Meta reallocates continuously toward whatever is producing results most cheaply right now.
Why it matters for HVAC
Demand moves faster than a monthly optimisation call. A heat wave arrives on a Tuesday, and automated reallocation responds that afternoon rather than at the next review. For seasonal trades this is the strongest argument for handing over control.
The failure mode is specific: Meta optimises toward the cheapest result, and the cheapest result is frequently the least valuable one. Left alone with a tune-up ad set and a replacement ad set, it will fund the tune-ups, because they convert more often and cost less each.
How it is measured
Watch the spend split by ad set over time, not just the blended cost per result. If the money has quietly concentrated into one ad set, ask what that ad set actually sells.
A worked example
A campaign runs a $150 tune-up offer and a system replacement offer under one budget. Within three weeks the tune-up ad set holds 82% of spend, because it produces leads at a third of the cost. Blended cost per lead looks excellent, replacement enquiries have nearly stopped, and the revenue that pays for the year is the one that got defunded.
These figures are invented to show the arithmetic. They describe no real business.
The fix
Separate campaigns for offers with different values, or a minimum spend floor per ad set. Either keeps the expensive, valuable thing funded.
Where this is demonstrated
The Advantage+ against manual pillar scores every listed agency on what they document here, and gives you the questions that tell you which one you are being sold.