Most plumbing marketing agencies are judged on the wrong thing: the monthly fee. What actually decides the outcome is how the agency is paid, because that determines who carries the risk when the ads underperform. This guide sets out the fee models and the questions to ask.
Choose a plumbing marketing agency on how it is paid, not on its monthly fee. The fee model decides who absorbs the loss when ads underperform. It matters because the job behind each lead is small: a plumbing service call averages $150 to $500, so a few wasted leads erase a week of margin.
Start from the job, because plumbing tickets are smaller than the trades a generic benchmark describes.
A standard plumbing service call runs $150 to $500 nationally, with plumbers charging $50 to $200 per hour. Emergency work carries a $50 to $100 per hour premium over the normal rate. So a single call-out is frequently a two or three figure job, not a four figure one.
Now set that against what buying the traffic costs. Published 2026 search advertising figures put home and home improvement advertisers at $90.92 per lead, from an $8.33 average click and an 8.05% conversion rate, against $66.69 per lead across all industries.
Hold the two numbers next to each other and the problem is obvious. A lead at the home improvement average is a meaningful share of a single service call before anyone has answered the phone, and that is before an agency fee sits on top of it. Plumbing only works at those prices because of what happens after the first job: the repeat call, the replacement, the maintenance plan.
That is the number to manage. Not cost per lead, and certainly not cost per click.
Four models, and the difference between them is not the price. It is who loses money when the campaign does not work.
| Model | How it is billed | Who carries the media risk | The incentive it creates |
|---|---|---|---|
| Flat monthly retainer | A fixed fee, ad spend separate | You | Predictable income regardless of result |
| Percentage of ad spend | A share of what you spend | You | Spending more raises the fee |
| Per qualified lead | A price per inquiry that meets a definition | The agency | Volume, so the definition has to be tight |
| Per booked job | A price per confirmed appointment or job | The agency | Closest to your own interest |
Read the fourth column rather than the second. A percentage of ad spend is the model to understand most carefully, because the fee rises when the budget rises whether or not the extra budget produced extra work. That is not dishonesty, it is just an incentive pointing away from you, and it should be priced in.
A retainer is not worse than a performance model by default. It buys continuity, and an agency carrying no media risk will often take on work that a performance model cannot justify. The question is whether you can tell, month to month, what you got for it.
The structural comparison between paying for traffic and paying for outcomes is set out in pay per lead against pay per click, and it applies to agency fees exactly as it applies to buying leads directly.
Eight questions. The answers are more useful than any pitch deck, and a good agency will have them ready.
Two of those decide more than the rest. Account ownership and the lead definition are where a cheap arrangement quietly becomes expensive, and they are the two most often left vague. The same discipline applied to buying leads is in the answers a lead seller should put in writing.
Because it rewards the half of the job the agency controls and ignores the half you do.
An agency can lower your cost per lead by buying broader, cheaper traffic. The number on the report improves. The quality of the inquiries falls, your crew wastes time on calls that were never jobs, and nothing in the report shows it. The trap is identical to the one written up for the neighbouring trade in why cost per lead traps roofing marketing buyers.
Cost per booked job is harder to game, because it only improves if the inquiries were real. Ask for it monthly, by source, and accept that the first few months will look worse than a cost per lead report did. That is the measurement getting more honest, not the campaign getting worse.
Three things make that number possible, and all three are your side of the arrangement:
Without those, no agency can be held to a result, and the conversation defaults to cost per lead because it is the only number both sides can see.
More than the agency choice does, which is awkward but worth knowing before you blame a campaign.
Research on online inquiry handling found that most companies respond far too slowly to web leads. In plumbing the effect is sharper than in considered purchases, because a blocked drain or a leak is bought in minutes, not weeks. The first plumber who answers often wins without competing on price at all.
That has a direct bearing on how you judge an agency. If inquiries arrive and sit unanswered for two hours, a perfectly good campaign produces a terrible cost per booked job, and switching agency will not fix it. Measure your own time to first contact before you conclude the leads are bad. The routing detail is in the speed to lead routing guide for plumbing companies.
A fair test holds your process constant. Same people, same script, same answer time, for the whole trial. Otherwise you are measuring your own enthusiasm for a new supplier.
Ask for the work, not the summary. A report showing impressions and clicks describes an outcome; it does not tell you whether anyone touched the account.
Five things should be happening, and all five are visible if you ask:
None of that requires you to understand the platform. It requires the agency to show the change and the date, and you to notice when a month passes with neither.
The reason this matters for choosing between fee models is simple. Under a retainer or a percentage of spend, this work is the only thing you are buying, so its absence is the whole problem. Under a per-lead or per-job model the supplier absorbs the cost of not doing it, which is why those arrangements tend to be self-policing.
Three situations, and recognising them saves a year of fees.
Your volume is too small to learn from. Automated bidding needs conversion data. An account producing a handful of conversions a month improves slowly whoever manages it, and the management fee is a large share of a small budget.
You cannot answer the phone reliably. Buying more inquiries into a process that drops them makes the problem more expensive, not smaller. Fix the answering first, then buy volume.
You need jobs this week. Building an account takes time. If the schedule has a hole in it now, buying qualified inquiries directly is faster than onboarding an agency, and the comparison between buying inquiry by inquiry and running the media yourself is the subject of the pay per lead guide.
Conversely an agency earns its fee when your volume is steady enough to optimise, your margin per job can absorb a slow start, and someone on your side owns the relationship. If no one internally is responsible for the account, the arrangement decays quietly whoever you hire.
If you would rather buy a measured outcome than a monthly fee, that is the model behind performance-based plumbing lead generation, where the supplier carries the media risk instead of you.
Three months, measured properly, is enough to decide.
Fix one trade, one territory and one job type, so the test is not measuring a change in demand. Hold your own answering process constant. Record every stage. Then compare cost per booked job against what the work is worth, remembering that a service call at $150 to $500 carries a different ceiling from a repipe or a heater replacement.
Agree the exit before you start. What notice is needed, what transfers, what you keep. An agency confident in the work does not need a long lock-in to prove it, and one that does has told you something about its own expectations.
There is no reliable published benchmark, and most quoted ranges come from agencies selling the service. Work it out from your own numbers instead: take the margin on a typical job, multiply by the jobs you need, and decide what share of that you can spend acquiring them. A fee that looks small against a competitor's is still too high if it does not clear that bar.
It depends on whether you can carry the media risk. A retainer gives predictable cost and continuity but leaves you exposed when the auction moves. A per-lead arrangement shifts that exposure to the supplier at a higher unit price. Smaller operators with thin margins usually do better moving the risk; larger ones with steady volume often do better keeping it and owning the account.
If someone will own it. A well-run account can beat an agency on cost, because you are not paying a management fee. The failure mode is an account nobody reviews, where search terms drift, budgets leak and the cost shows up months later. If no one has the time, paying for management is cheaper than paying for neglect.
Account ownership, and the definition of a lead. Everything else can be renegotiated. Those two determine what you keep if the relationship ends and whether you can tell what you bought while it lasts.
Expect movement in the first month and a usable read by the third, once a full cycle of inquiries has been through to booked jobs. Judging on month one is unfair, and judging on cost per lead at any point tells you less than it appears to.
Our technology is designed to measure success. With Dolead, track and measure success at the most granular level, ensuring transparency and continuous improvement.
.svg){kind=small}
.svg){kind=small}
.svg){kind=small}
.svg){kind=small}
