The question we hear most from service businesses is also the easiest to get wrong: “What should our Meta budget be?” Spend too little and the algorithm never gathers enough signal to optimize. Spend too much, too fast, and you scale losses before you’ve found a winner.
Start with your numbers, not a benchmark
Your budget should be driven by your economics — your average job value, your close rate, and the most you can profitably pay for a booked job. If a new customer is worth $2,000 and you close one in four leads, you can afford a healthy cost per lead, and your budget should reflect that math.
Respect the signal floor
Meta’s optimization needs roughly 30–50 conversions per ad set per week to stabilize. Work backwards: if your target cost per lead is $40 and you need about 40 leads a week to learn, you’re looking at a starting budget near $1,600/week for that campaign.
- Far below the signal floor, the algorithm stays stuck in learning and costs stay high.
- At or above it, performance settles and scaling becomes predictable.
Scale on proof, not hope
Once an offer is profitable, we scale roughly 20–30% every few days — fast enough to grow, slow enough to keep the algorithm stable. The businesses that win treat budget as a dial tied to ROAS, not a fixed line item.
The short answer
Most service businesses we work with start around $3,000/month in ad spend, gather signal quickly, and scale from there. The right number is whatever lets you buy booked jobs below your max — and then buy more of them.
Rated 4.9 by the businesses we grow.
Want this done for you?
Tell us about your business and goals — we’ll come back with a straight assessment of where the growth is.