It's Tuesday morning, and rain that wasn't in yesterday's forecast just wiped out your entire route. Six jobs, gone — and every one of those customers needs a call, an apology, and a new slot squeezed into a week that was already full. While you're working the phone to rebook them, three new quote requests come in from people who'll happily call the next name on the list if nobody answers. Here's what's safe to hand to a system on a week like that, and what still needs your eye and your ladder.
Most of what floods a window cleaning company's phone isn't the cleaning itself — it's scheduling traffic: weather reschedules, new quote requests, and questions about screens, storm windows, or when a crew is coming back. All of that is safe to hand to a system, checked against your real route for the day. What's never safe to automate: deciding whether a job is too dangerous to run in current conditions, and pricing or working an unusual job — leaded glass, a hard-to-reach third story, heavy hard-water staining. That judgment stays with a person who actually knows the site.
Before anything else: when this guide says AI agent, it means a software system that looks at what's happening — a rained-out route, a new quote request coming in by text, a customer asking to move Thursday's job — figures out what's actually being asked, and handles it the way a well-trained office manager would, without pulling you or a tech off a ladder to do it. It's not the one deciding whether today's wind is too much to safely work a second story. It handles the scheduling traffic around that decision, so the person who does make it isn't also stuck on the phone all afternoon.
Most home service businesses can absorb one bad day. A general contractor pushes a start date; a landscaper mows a lawn a day late and nobody notices. A window cleaning crew doesn't get that slack, because the whole business runs on a tight weekly route: a set number of jobs, in a set order, that a two- or three-person crew can physically get to before the next round of bookings starts. Rain, high wind, or a hard freeze doesn't just cancel one job — it cancels the whole day, and every one of those jobs has to land somewhere else in a week that's already booked.
That's exactly the moment new business shows up too. A homeowner comparing quotes, or a property manager whose storefront glass is looking bad before a weekend event, doesn't wait for your rained-out day to end before calling around. They call whoever answers. If your crew is spending the morning working the phone to reschedule six existing jobs, that new call is either going to voicemail or to the next company on the list.
| What happens | Left alone | Automated first |
|---|---|---|
| Rain cancels today's six-job route | Crew calls each customer one at a time to reschedule | All six get a same-day text with the next open slot, at once |
| New quote request comes in by text or web form | Sits until someone's off the ladder and back at the phone | Answered same-day with pricing basics and a scheduling window |
| Deciding whether wind or ice makes a job unsafe today | (same either way — always needs a person) | Never automated — stays with whoever's running the crew |
| Customer asks about an unusual job — leaded glass, a hard-to-reach third story | (same either way — needs a real quote, not a form price) | Routed to a person for a photo review or site visit |
Take the free 2-minute readiness assessment
Window cleaning isn't a business with a handful of national players squeezing everyone else out. According to IBISWorld's 2024 industry analysis, there are 35,344 window washing businesses operating in the United States, and the industry is so fragmented that no single company holds more than 5% of the market.1 That's tens of thousands of small, mostly local crews all competing for the same homeowners and storefronts — which means the businesses that win aren't necessarily the cheapest or the most experienced, they're the ones a customer can actually reach when it matters.
As one small crew owner might put it, in a hypothetical but realistic scenario: "I lost count of how many jobs I've picked up just because I called someone back the same day and the other guy hadn't gotten to it yet. That's not me being better at cleaning windows. That's just me answering the phone." In a market with this many competitors and this little differentiation on price or service, response speed is one of the few edges a small crew actually controls.
Illustrative example, not a real customer: picture a two-person window cleaning business running about 14 residential jobs a week at an average $220 a visit — in line with HomeAdvisor's reported national average of roughly $220 per job, or about $10 per window — plus five recurring storefront accounts billed monthly. That's rough revenue of around $160,000 a year from residential work alone. Say a rained-out day pushes six jobs to reschedule, and working the phone to rebook all six by hand eats about 90 minutes — the same 90 minutes when three new quote requests come in and sit unanswered. If a system handles the weather reschedule automatically and answers those new requests same-day instead of after a two-day gap, and that lifts the close rate on new quotes from a baseline 20% to 30% (a realistic, illustrative improvement, not a guarantee), that's roughly one extra booked job a week — around $11,000 in added revenue over a year, plus a crew that gets its afternoon back to actually catch up on the backlog.
That's a fair instinct, and it's half right. Deciding whether today's wind is safe to work a ladder or a lift, and quoting a job with leaded glass or a genuinely hard-to-reach window, both need a person who can see the actual site — no system should touch either. But most of what fills up a window cleaning phone isn't that. It's "can you come Thursday instead of Tuesday," "how much for a two-story with screens," and "can you fit me in before my open house." Those are routine, repeatable questions with a clear right answer once a system can see your real schedule — and handling them automatically doesn't replace the judgment calls, it just clears the noise around them so your crew can spend more time on ladders and less time on hold.
This isn't a generic chatbot bolted onto your booking line — it's what Unmanually actually builds for a crew like yours: a system set up around your real weekly route, your real service area, and a firm line between routine scheduling and anything touching safety calls or unusual-job pricing, which always stays with you. It never decides whether a job is safe to run today, and it never quotes a job outside what you've told it is routine.
We also back this with a real guarantee, not a vague promise: try Unmanually for 60 days, and if it isn't saving your crew real time, whatever's left of your prepaid balance converts to account credit. That's not a cash refund on usage you've already consumed, since that reflects real infrastructure cost already spent, but it does mean you're never stuck paying for a system that isn't earning its place during your busiest weeks. For a small crew, the growth angle is direct: in a market this fragmented, the crew that answers first and reschedules fastest after a rained-out day is often the one that keeps the customer for years, not just for one job.
Not ready to commit to anything yet? That's completely fine — leave your email on our presale waitlist and we'll let you know as soon as weather-reschedule automation for window cleaning crews is live, including founding-member presale pricing before it opens to everyone else this October.
1. IBISWorld, "Window Washing in the US Industry Analysis," reporting 35,344 window washing enterprises operating in the United States as of 2024, with the industry so fragmented that no single company holds more than 5% market share.
The full pillar guide this article belongs to.
Another exterior trade where responding to a new quote request faster than the next company matters more than almost anything else.
The same crowded, fragmented-market dynamic, in a business that also runs on a tight weekly schedule.
More on the safety-call and judgment-call line — where automation should stop, referenced above.
A person still makes that call, every time. Deciding whether wind, rain, or ice makes a job unsafe or unworkable is a judgment call that needs someone who knows the site, the equipment, and the forecast. What a system can safely handle is everything after that decision: texting every affected customer at once with a same-day heads-up, offering the next open slot, and updating the day's route — instead of your team calling six people one at a time while new quote requests pile up behind them.
Not if it's scoped correctly. A system can answer common pricing questions, collect the details a quote actually needs — number of stories, screens, storm windows, when it was last cleaned — and get a customer on the schedule the same day, which is often what wins the job over a competitor who calls back two days later. It should never be the one that prices a genuinely unusual job, like leaded glass or a hard-to-reach third story; that stays with a person who can look at photos or visit the site.
Take our free 2-minute readiness assessment — it walks through this same test against your actual route and schedule, and tells you honestly what to automate first.
Take the 2-min readiness assessment