Expect $1,500 to $4,000 a month in management fees for a Tampa Bay home-services company, plus media budget on top, plus a one-time build in the four to five figures. The two best published surveys put average retainers at $1,557 and $2,917 a month, and a third of the market charges under $1,000. All of those numbers are true. This article explains why they disagree and how to work out your own number.
We publish pricing because refusing to is a tactic, not a policy. An agency that will not give a range before a discovery call is protecting its ability to price you by how you look rather than by what the work costs.
Tampa Bay changes the arithmetic in two ways. Auction prices run higher than in smaller Florida markets because there are more advertisers per rooftop and a steady supply of out-of-state operators buying their way in. And demand does not arrive evenly. A roofer in Clearwater can bill more in six weeks after a storm than in the previous six months. Budget on a flat monthly average and you will overspend in June and be invisible in September.
Real numbers first: what the surveys actually say
Ahrefs surveyed 439 providers in 2024. Average monthly SEO retainer: $2,917. Providers serving local-only clients: $1,557. Share of providers charging $2,000 a month or less: 68.8%.
SE Ranking surveyed 260 agencies in December 2024 and found 64% charge under $1,000 a month.
Take the local-only figure of $1,557 as your anchor if you serve one metro. Take $2,917 as the anchor if you are running multiple trades across Hillsborough, Pinellas, and Pasco with a full content and paid program. Anything meaningfully above that should come with a named team and a scope you can read in one sitting.
The survey gap that should change how you read any price list
Those two surveys are roughly threefold apart. That gap is not a contradiction, it is a description of the market.
SE Ranking's panel was 94% small local agencies. Ahrefs pulled from a broader population that included firms with national accounts and larger teams. Ask any provider one question and the gap resolves: how many clients does the person doing my work carry at once? A solo operator running eighteen accounts at $800 each has roughly two hours a month for you. That is enough to keep a Google Business Profile healthy and answer reviews. It is not enough to run paid search, publish, and maintain a site.
Price is a proxy for hours. Hours are a proxy for attention. Buy the attention, not the invoice.
The single most common way contractors get misled
Ask any quote to separate the management fee from the media budget, in writing, before you sign.
A Brandon plumbing company we talked with had been paying $2,900 a month for eight months believing that was the advertising budget. It was not. Roughly $1,400 went to Google and the rest was the fee. Nobody lied. The proposal said "$2,900 monthly investment" and the owner filled in the meaning himself.
Four line items belong on every proposal: the management fee, the media budget, one-time setup, and any software billed through the agency. If your quote is one number, it is not a quote, it is a price. And you should be able to open your own Google Ads account and see the media figure match to the dollar. If you cannot open the account yourself, that is a separate and larger problem.
One-time costs nobody mentions until the invoice
The retainer is the visible cost. The build is the one that surprises people.
A contractor website that carries real service pages, working call tracking, and forms that route to a person runs from the low four figures for a small template build to the mid five figures for a custom multi-trade site with original photography. Brand identity, vehicle wraps, and print are separate budgets, and in a market where a Riverview homeowner sees four trucks a week, the wrap does more work than most people credit it for.
Then the plumbing behind it: conversion tracking configured to fire on booked jobs rather than form views, call recording, Google Business Profile verification for each area you service, and review-request automation. Budget one to three thousand for that category. Skipping it is how a company ends up eighteen months in with no idea which channel produced which job.
Benchmarks borrowed from companies nothing like yours
You will be shown a percentage-of-revenue rule. The usual source is Gartner's 2026 CMO Spend Survey, which found companies spend an average 7.8% of revenue on marketing in 2026, up from 7.7% in 2025. Gartner surveyed 401 chief marketing officers at large enterprises. None of them dispatch trucks.
Use it as a guardrail, not a plan. A more relevant figure comes from Jobber, which surveyed 1,050 home-services owners in December 2025 and asked where the work actually originates. Referrals and repeat customers led at 59%, Facebook at 32%, Google Search at 20%, Local Service Ads at 19%. Paid channels are the minority of your revenue. Price them like an addition to a working business, not like the business itself.
Retainer, project, or performance: pick the failure you can live with
Retainer
Predictable, easy to plan around, and the default for anything ongoing. It buys effort, not outcomes, so the risk is quiet drift into a monthly report about a monthly report. Fix it by writing two or three shipped deliverables into the agreement and holding the line on them.
Project
Right for a website, a rebrand, a set of trade pages. Wrong for search, which does not have a finish line. The risk is that attention ends at delivery. If you buy a site as a project, buy a separate maintenance arrangement in the same conversation or the site decays quietly.
Performance and pay-per-lead
Appealing because the risk looks shifted. The risk that actually shifts is definitional. Whoever writes the definition of a lead writes your invoice, and a duplicate, a solicitor, and a homeowner in Ocala all count unless you excluded them in writing. Pay-per-lead also rewards volume, which is the opposite of what a shop with three trucks needs. If you use it, require call recordings and a written disqualification list.
Pricing a click, a lead, and a booked job
Here is the arithmetic that actually answers "can I afford this."
LocaliQ published Google Ads benchmarks in July 2026 drawn from 3,211 US home-services campaigns using April 2024 through March 2025 data. Cost per click: HVAC $9.68, plumbing $10.49, roofing $10.70, electrical $12.18, home services overall $7.85. Cost per lead: HVAC $127.74, plumbing $129.02, roofing $228.15, electrical $93.69, home services overall $90.92.
From click to lead
Divide the lead cost by the click cost and you get how many clicks it takes to produce one lead. Roofing: $228.15 divided by $10.70 is 21.3 clicks per lead, which means about 4.7% of clicks turn into a lead. Electrical: $93.69 divided by $12.18 is 7.7 clicks per lead, or roughly 13% converting.
Read that again. Electrical clicks are the most expensive of any trade in the set and electrical leads are the cheapest. The difference is not the ad, it is what happens after the click. An electrician's page converts nearly three times better than a roofer's on the same measured data, which tells you where the cheapest improvement usually sits: not in the bid, in the page.
From lead to signed contract
Contractor Magazine, quoting Theo Prada of Queen Consultancy in August 2026, put close rates at roughly 30% to 45% in plumbing and 25% to 40% in HVAC. That is agency portfolio data rather than independent research, and it does not include roofing, so for roofing use your own historical number and nobody else's.
Say a Clearwater roofer closes 25% of qualified leads. Four leads per signed roof, at $228.15 each, is $912.60 in ad cost per contract. Angi's 2026 figure puts an average roof replacement at $9,607. So acquisition runs 9.5% of contract value. At a 20% close rate it takes five leads, or $1,140.75, which is 11.9%.
Run the same for HVAC. Angi puts an average system replacement at $7,500. At a 35% close rate you need 2.86 leads, which at $127.74 is $365.34, or 4.9% of the ticket. At 25% it is four leads, $510.96, and 6.8%. That spread between 4.9% and 6.8% is entirely your sales process, not your ad account.
The number that makes "expensive" mean something
Cost per booked job, all in, is the only figure that settles a pricing argument.
An agency billing $1,500 a month that produces six booked jobs costs you $250 per job in fees. One billing $3,500 that produces twenty-two costs $159. The larger invoice is 36% cheaper per job. Without that denominator, "expensive" is a feeling.
If your current provider cannot tell you cost per booked job by channel, the tracking was never built. That is a fixable problem and worth fixing before you change anything else, because every other decision you make is guesswork until it is.
Budgeting for a year that does not run at one speed
Most proposals assume twelve equal months. Tampa Bay does not have twelve equal months.
Take the Clearwater roofer above, targeting five signed roofs a month at a 20% close rate. That is 25 leads at $228.15, or $5,703.75 a month, and $68,445 across the year.
Spending that evenly is the wrong shape. Weight it instead: four high-demand months at 1.6 times the average, which is $9,126 a month, and eight quieter months at 0.7 times, which is $3,992.63. Four times 1.6 plus eight times 0.7 equals 12.0, so the annual total is unchanged at roughly $68,445. Same money, aimed at the weeks when a homeowner in Palm Harbor or Apollo Beach is actually looking at a ceiling stain.
Two cautions. Auction prices rise when demand rises, so your peak-month cost per lead will run above the benchmark and your quiet-month cost below it. And organic visibility does not flex like this. Local search and map visibility has to be funded through the quiet months precisely so it is standing when the demand arrives. Cut it in July and you will not have it in October.
The figure we refuse to publish
We will not give you a cost per lead for Google's Local Service Ads. Not for roofing in St. Petersburg, not for anything.
Google publishes no such data. Every number you have seen is one agency's internal portfolio presented without methodology, and those estimates disagree by two to threefold. A number that swings that far is not a benchmark, it is an anecdote with a dollar sign.
What is measurable is placement. A WebFX study of 500 SERP runs in May 2026 found Local Service Ads appeared on 78 of 100 home-services keywords, always in position one, with the local pack on 98 of 100 and regular Google Ads on only 26 of 100. The placement is clearly worth having. Its price in your ZIP codes is something you learn by running a capped test for thirty days, and any agency claiming to know it in advance is guessing on your dime.
The Performance Max migration and your next budget
Google is moving Local Service Ads into Performance Max pay-per-lead. Google Ads Help states the US home-services rollout began in August 2026, that manual bidding and industry Target CPA are being deprecated, and that historical reporting does not carry over.
Practical effects on budgeting. Export your existing cost and lead history now while it is still retrievable. Assume less direct bid control, which shifts the levers toward review volume, response time, and the quality of the signals you feed the system. And treat any 2027 projection built on pre-migration numbers as provisional, including one from us.
Where the money quietly leaks
Assets in somebody else's name. If the domain, ad account, Google Business Profile, tracking numbers, or site are owned by the agency, your exit costs you every bit of history you paid to create. Everything in your name, agency granted access. Confirm it before signing.
Reporting that costs more than it reveals. A monthly deck of impression graphs is billable hours spent describing motion. One page with leads by source, booked jobs, and cost per booked job takes less time to make and is worth more.
Lock-ins without a performance exit. A twelve-month initial term is defensible for search work that genuinely compounds. Twelve months with no out and no defined milestones is not. Ask for month-to-month after the initial term and asset release on request either way.
Duplicate spend across trades. Multi-trade shops routinely pay for overlapping campaigns bidding against themselves in the same geography. If you run HVAC and roofing from one company across Wesley Chapel and Land O' Lakes, someone should be checking that.
Scope you never asked for. Social posting, extra articles, an unused chatbot. If you cannot connect a line item to a booked job in two steps, cut it and put the money into media.
Frequently asked questions
Why do Tampa marketing quotes vary so much for the same work?
Because the surveys behind them describe different businesses. Ahrefs found an average retainer of $2,917 across 439 providers in 2024, while SE Ranking found 64% of 260 agencies charge under $1,000, with 94% of that panel being small local shops. Price mostly tracks hours of attention. Ask how many accounts your assigned person carries, and the range stops looking mysterious.
How much should a five-truck company spend each month?
Work backward from jobs, not from a percentage. If you want five signed roofs a month, close 20% of leads, and roofing leads cost $228.15 per LocaliQ's 2026 benchmarks, that is 25 leads and about $5,700 in media, plus a management fee. Most five-truck Tampa Bay shops running a real paid program land between $4,000 and $8,000 a month all in.
Do I keep my Google Ads account if I fire the agency?
Only if it was created under your own account with the agency granted manager access. If the agency built it inside their manager account and owns it, you lose the conversion history, audience data, and campaign structure when you leave, and your next provider starts from zero. Verify ownership of the ad account, domain, and Google Business Profile before you sign anything.
How should I budget around hurricane season?
Set an annual media number, then weight it by month rather than dividing by twelve. Push over half your paid budget into the high-demand months and hold the rest back for the quiet ones. Keep organic and map visibility funded year-round at a steady level, because that is the channel that has to already be in place when demand spikes and cannot be switched on in a week.
What is a fair contract length?
Three to six months for paid advertising, since it produces measurable data quickly. Six to twelve months for search and content work, which genuinely compounds slowly and cannot be judged at week eight. In every case, insist on month-to-month terms after the initial period and written asset release on request. A provider unwilling to release your assets is telling you something useful.
Keep reading
- Land O' Lakes Plumbers: Google Local Service Ads
- Riverview HVAC: The Referral System That Books Jobs
- Marketing Strategy for Safety Harbor Small Businesses
Bring us your last twelve months of jobs by source and we will build this calculation for your trade, your close rate, and your average ticket, on one page you keep. No obligation and no proposal unless you ask for one. Start with our approach to contractor marketing, or just book a call.
