Ask three contractors how much to budget for Local Services Ads and you will get three round numbers — a thousand, five thousand, "whatever the agency recommends." None of them can tell you why that figure and not one half or double it. That is a problem, because in a pay-per-lead channel the budget is not a marketing preference. It is a direct function of how much work you can do and what a booked job is worth to you. The right number is one you can derive on the back of an envelope and defend line by line.
This piece is about arriving at that number — the "what number" question, not the mechanics of how Google paces it through the week. We will work backwards from the only ceiling that actually matters: your capacity to fulfill jobs.
Start with capacity, not a spend target
Most budget conversations start at the wrong end. They ask "how much should I spend?" when the disciplined question is "how many jobs can I fulfill next month, and what is one worth?" Local Services Ads charge per lead, not per click — the ad sits above the map pack and organic results, and you pay when a searcher contacts you. So every dollar buys leads, and leads only matter to the extent you can convert and service them.
If your crew can complete 30 net-new jobs a month, buying enough leads to book 60 is not ambition — it is waste. Beyond your fulfillment line, extra leads either go unanswered (which quietly hurts your ranking, since responsiveness is a performance factor) or turn into overbooked, poorly-served customers who leave the reviews that sink you. Capacity is the anchor. Everything downstream is arithmetic.
How much to budget for Local Services Ads: a bottom-up derivation
Here is the full chain for one illustrative trade — an HVAC company in a mid-size metro that wants to fill 30 booked jobs a month from LSA. Every input below is a figure you should replace with your own; they are illustrative, not benchmarks to copy.
The logic runs in four moves. Convert booked jobs into contactable leads using your close rate. Gross up contactable leads into raw leads, because a large share of raw leads — third-party estimates put it around 45% — are unbookable (wrong geo, out of scope, spam, or unreachable). Multiply raw leads by your cost per lead. Then subtract the slice of spend you can expect back through Google's auto-credit system.
| Step | Input (illustrative) | Result |
|---|---|---|
| Monthly fulfillment capacity you want to fill from LSA | 30 booked jobs | 30 jobs |
| Close rate on contactable leads | 40% | 75 contactable leads needed (30 ÷ 0.40) |
| Unbookable share of raw leads | ~45% (so 55% are contactable) | ≈ 136 raw leads needed (75 ÷ 0.55) |
| Average cost per lead for the trade/metro | $65 | ≈ $8,840 gross spend (136 × $65) |
| Credit recovery buffer (Google auto-credit) | ~6.5% recoverable | ≈ $8,265 net monthly spend |
| Implied true cost per booked job | $8,265 ÷ 30 | ≈ $275 per booked job |
Now the number defends itself. Roughly $8,300 a month yields about 30 booked jobs at a true cost of about $275 each. Whether that is a good budget is not a matter of opinion — you check the last row against your target. If a booked HVAC job carries an average ticket and margin that comfortably clears $275 in acquisition cost, the spend is justified and you could even push it if you had capacity. If your target cost-per-booked-job is $200, the current close rate or lead mix does not support this spend yet, and the fix is operational, not budgetary. That is the whole point of deriving the figure: it tells you not just how much to spend but whether to.
Cost per lead varies enormously by trade and market — commonly cited around $53 on average, but ranging roughly $12 to $180 — so the single most important thing you can do is replace $65 with your own trailing number from your lead report. The structure of the derivation holds; only the inputs change.
Weekly budget cap vs. actual spend
Google does not take a monthly number. You set a weekly budget cap, and this is where owners routinely misread their own accounts. The cap is a ceiling, not a target. Google paces charges beneath it and, in most weeks, spends less than the maximum because demand and the auction rarely fill every dollar. Over a full month, actual spend can climb toward roughly the weekly cap times about 4.3 — but treating the cap as your expected spend will leave you consistently short of the number your derivation demands.
Translate the monthly figure the other way. An ≈$8,265 monthly plan is about $1,920 a week of expected spend, so you would set the weekly cap somewhat above that — enough headroom that a busy week does not exhaust the budget by Thursday and stop showing your ad through the weekend, when high-intent emergency searches often convert best. The cap protects you from a runaway week; it does not, by itself, guarantee you spend enough to hit capacity.
Why both over- and under-budgeting cost money
Under-budgeting is the obvious loss: your cap trips mid-week, your ad goes dark, and jobs you had the capacity to book go to whoever is still showing. In a channel where being present at the moment of search is most of the game, an ad that is switched off for two days a week is not saving money — it is refusing revenue.
Over-budgeting is the quieter, more insidious loss. Past your fulfillment line, incremental leads have a falling return. You pay full price for leads you cannot service well, and the marginal booked job — the one you squeezed in past capacity — often gets rushed work and a mediocre review. Because reviews and responsiveness feed your ranking, over-spending can degrade the very signals that keep your cost per lead low. The budget has a right size, and drifting above it does damage that does not show up on the invoice.
Two levers move the whole model without touching the spend line. Recovering more of the unbookable ~45% through fast, disciplined lead handling raises your effective close rate, and diligently disputing genuinely bad leads pushes credit recovery toward the top of that 6–7% band. Both lower true cost per booked job — which is the number your budget ultimately has to justify.
Frequently asked questions
What is a good starting budget for Local Services Ads?
There is no universal number. Size it from the jobs you can actually fulfill, your average cost per lead, your close rate on contactable leads, and the roughly 45% of raw leads that are unbookable. A single-truck trade in a mid-size metro often lands in the low four figures per month, but a defensible figure comes from the derivation — not a round number.
Is the LSA weekly budget a spending target or a cap?
It is a cap, not a target. Google paces charges under the weekly maximum and typically spends less than the ceiling. Over a month, actual spend can approach roughly the weekly cap times about 4.3, but demand and the auction usually keep it below that. Set the cap above your expected weekly spend so you do not run out mid-week.
Should I raise my LSA budget or improve my close rate first?
Usually close rate first. Because a large share of raw leads are unbookable and part of your spend is recoverable through Google auto-credit, improving speed-to-lead and lead triage lowers your true cost per booked job without spending an extra dollar. Only raise the budget once you are booking near the limit of what you can fulfill.