Agency, freelancer, or do it yourself: choosing marketing help
What each option really costs a window cleaning company, the red flags in a contract, what a fair price is on a small ticket, and why a trial is reasonable.
Window cleaning companies get pitched harder than almost any trade, because the businesses are small, the owners answer their own phones, and the lists are cheap to buy. If you have a Google listing, you already know this. The pitch is always the same shape: page one, a number of leads per month, and a contract before anything has happened.
The question is not whether marketing help is worth it. It is which kind, at what price, given that your average job is around $260 and there is no room for a $2,500 monthly retainer to be wrong for six months.
What the three options actually cost
Doing it yourself costs three to five hours a week plus whatever you spend on ads. That is real money if those hours would otherwise be billable. A technician hour is worth $80 to $120 on site, so five hours a week of your own time is $400 to $600 a week of opportunity cost, which is more than most retainers.
A freelancer runs $300 to $500 a month for one channel done competently, usually Google Ads management or Google Business Profile and reviews. You still pay the ad spend on top. Freelancers are good value and they are one person, which means one person's vacation, one person's illness, and one person's opinion.
An agency runs $1,000 to $3,000 a month for a bundle, again plus ad spend. At the bottom of that range you are buying one channel and a report. At the top you should be getting ads, tracking, a site that converts, review generation and follow-up, with someone accountable for the number of booked jobs rather than the number of clicks.
Add it up before you sign. At $1,500 a month in fees plus $1,500 in ad spend, you need roughly 12 extra jobs a month at $260 just to cover the cost, before your own labor and materials. That is not an argument against paying it. It is the number you should be able to say out loud in the first meeting, and so should they.
What an owner can honestly do alone
Three hours a week, spent well, beats most cheap retainers. Here is what fits in it.
Thirty minutes on reviews: check the tally, text the customers who did not review, reply to every new one. This is the highest-return hour in the trade and it cannot be outsourced well, because the ask happens at the door. Across 14,348 window cleaning listings, the median company has 7 reviews and only 10.1% have passed 100, so consistency alone puts distance between you and the field.
Thirty minutes on the Google Business Profile: add photos from the week, answer questions, post once. Sixty minutes on the ad account: read the search terms report, add negatives, listen to two recorded calls. Sixty minutes on follow-up: call every quote from the last two weeks that has not booked.
What you cannot realistically do yourself is build a fast website, set up conversion tracking correctly, or run a Meta campaign that does not quietly waste half its budget. Those are the jobs worth buying.
Red flags in the pitch
- A 12-month contract with no exit. Six months is defensible while a site and a campaign mature. Twelve with no out is a retention strategy, not a service.
- They own the ad account. Your Google Ads and Meta accounts must be in your name, with you as admin, and the same goes for your Google Business Profile and your domain. When the relationship ends, you should keep the history. Ask this question first, because the answer sorts the field fast.
- Guaranteed rankings or a guaranteed number of leads. Nobody controls Google's results, and a guaranteed lead count usually means the count includes wrong numbers and spam form fills.
- No call recording and no call tracking. In a phone-driven trade, an agency that cannot play you the calls it generated is asking you to take its word for the only thing that matters.
- Reports built on impressions, reach and "engagement". Those are inputs. Ask for calls, form fills, booked jobs and cost per booked job. If the report does not have those four lines, it is decoration.
- One price for every trade. Window cleaning is not roofing. A $9,000 roof supports a spend per lead that a $260 window job does not, and anyone quoting the same plan to both has not done the math on yours.
- No mention of your website. 32% of window cleaning companies on Google have no site at all, and sending paid traffic to a profile with no site is the most common way money disappears in this trade. The website guide lists what the site has to do before ads make sense.
What fair looks like on a small ticket
Fair is a fee that is a sensible fraction of the revenue it produces, with the math visible.
For a one-truck company doing $150,000 a year, $500 to $900 a month in fees is proportionate, usually aimed at one channel done properly plus reviews and follow-up. For a three-truck company past $400,000, $1,500 to $2,500 buys a real stack and should return several times that in booked work or be cut.
Ad spend should be quoted separately from management fees, paid on your own card, and visible to you in your own account. A management fee expressed as a percentage of ad spend creates an incentive to spend more, which is worth knowing even when the partner is honest.
Ask for the first 90 days in writing: what gets built, in what week, and which number is being moved. Then ask what happens in month four if the number has not moved. A partner who has answered that question before will answer it in a sentence.
Why a trial before a contract is a reasonable ask
Every other purchase in your business gets tested. You demo a water-fed pole system before you buy one. You work a technician for two weeks before you put him on the schedule alone. Marketing is the one category where the standard practice is to sign first and see later, and there is no good reason for that.
A short trial answers the questions a sales call cannot. Do they answer on Tuesday afternoon. Do they set up conversion tracking correctly. Does anything actually get built in the first two weeks. Is the reporting readable by a person who cleans windows for a living.
Ask for it plainly: "Run one thing for two weeks before I sign anything." Some will say no, and that is information. If you want to compare what you are being pitched against the wider field, the benchmark report shows where window cleaning companies actually stand on reviews, ratings and websites, and the FAQ covers the questions owners ask most often before they buy.
Frequently asked
Is it better to hire in-house? Not at this size. A competent marketing employee costs $50,000 or more, which is well past what a one-truck or three-truck company can justify. In-house makes sense somewhere past a million in revenue with several locations.
Should I pay per lead instead? Pay-per-lead sounds safe and usually is not, because the same lead is sold to three companies and you are bidding against them on speed. It can work as a supplement, never as your only source. Owning your own reviews, profile and site is what makes leads cheaper over time.
How long before I can judge results? Paid ads show a signal in two to three weeks and a real read at 60 days. Reviews compound over a quarter. Local search rankings and AI answers take three to six months. Anyone promising a map pack position in 30 days is describing something they cannot deliver.
Every service on this site is available as a free 14-day trial, no card, so you can test the answer to all of this before you sign anything: start here or text or call (385) 832-6175.