Services
Paid Media Management
Advertising is the easiest half of this to buy and the hardest half to make worth buying. The difference is what happens after the click.
8+ years building growth systems for service businesses. Leads strategy, media buying, offer design, sales conversion, and CRM architecture at Agency Logics.
Published
Optimising for leads is how accounts quietly fail
Most ad accounts we inherit are optimised for lead volume, because leads are the last event the platform can see. That is a measurement artefact, not a strategy, and it produces a predictable pathology: cost per lead falls, the owner is told things are improving, and the number of booked jobs stays flat or drops.
It happens because cheaper leads are usually worse leads. The platform, told to find people who submit forms, finds people who submit forms, including the price shoppers, the wrong service areas, and the not-yet-ready. Optimising toward the booked job instead means feeding the real outcome back into the platform, which requires the CRM to be wired to report it. That dependency is the whole reason we do not sell media management on its own.
| Measured on | What the platform learns | What the owner sees |
|---|---|---|
| Leads | Find people who fill in forms | Falling cost per lead, flat revenue, rising frustration |
| Booked jobs | Find people who become customers | Higher cost per lead, higher revenue, fewer wasted calls |
What the engagement covers
- Account build or takeover, in your accounts, with campaign structure set around service lines and margin rather than around what the previous agency inherited.
- Creative production and testing, because in a saturated local market the creative carries more of the result than the targeting does.
- Offer design. The most common fix in a struggling account is not the bidding, it is that the offer gives a stranger no reason to act now.
- Conversion tracking wired to the CRM, so the platform optimises toward booked jobs. See revenue operations for how attribution is built.
- Budget pacing and seasonality, which for most trades is the single largest lever nobody pulls. Spending evenly across a year in a business with a seasonal demand curve wastes money in both directions.
What we do not do
We do not take a media-only engagement where the response layer is broken. It is the most reliable way to produce a disappointed client and a bad case study, and after eight years the pattern is clear enough that it is not worth the revenue.
The documented programs
| Program | Period | Ad spend | Tracked sales | ROAS | Jobs |
|---|---|---|---|---|---|
| DFW specialty contractor | Jan to Dec 2025 | $234,073 | $3,922,673 | 16.76x | 93 |
| Multi-state service business | Jan to Dec 2024 | $101,361 | $1,339,016 | 13.21x | 82 |
| Specialty contractor | 2025 reporting year | $10,007 | $369,967 | 36.97x | 27 |
Read the third row carefully, because it is the most instructive one. $10,007 of spend produced $369,967 in tracked sales at 36.97x. That multiple is not the product of better bidding than the first program. It reflects a much smaller spend into a high-ticket service where the conversion machinery was already working. The lesson generalises: the leverage sits in ticket size and close rate, and the ad account is downstream of both.
Common questions
- What ROAS should a service business expect?
- There is no universal number, because ROAS depends on ticket size, close rate, and margin far more than on ad platform skill. Across three documented Agency Logics programs the figures were 16.76x, 13.21x, and 36.97x, blended at 16.3x. Those are historical results on specific accounts with specific economics, not a forecast for yours. A business with a $600 average ticket and a business with a $40,000 average ticket cannot be held to the same multiple.
- Why do you insist on fixing follow-up before increasing spend?
- Because spend multiplies your existing conversion rate rather than improving it. If leads currently wait six hours for a first response, doubling the budget buys twice as many leads that convert at the same poor rate, and you pay twice as much to learn that. Fixing response first raises the multiplier that all future spend runs through.
- How long before paid media works?
- Usually 60 to 90 days to get past the learning phase and into stable economics. That is materially slower than database reactivation, which produces appointments in about two weeks. This is why we typically start with the systems that pay back fast, then scale spend once the machine behind it is proven.
- Do you require a minimum ad budget?
- Not a fixed one, but the arithmetic sets a practical floor. Below a few thousand a month, platforms cannot gather enough conversion data to optimise, and management overhead eats the return. If your budget is under that, reactivation and response are a far better use of the same money, and we will say so rather than take the account.
- Who owns the ad accounts?
- You do. Accounts, pixels, audiences, and historical data stay in your name. An agency holding the ad account hostage is a common and indefensible practice, and it makes the eventual transition worse for everyone.
Have an account that is not working?
Nine times out of ten the diagnosis is one of three things: the account optimises for leads instead of jobs, the offer gives no reason to act now, or the response after the click is too slow to convert what it catches. All three are visible quickly.
Apply For a Growth Partnership