Residential routes · Commercial contracts
You know roughly who is due this month, because you have been doing it long enough to know. It lives in a notebook, a phone, and the bit of your memory that gets used for nothing else. It works right up until you add a second van.
The two halves of this business have opposite problems. Residential leaks revenue quietly, through customers who were due in March and nobody called. Commercial leaks it loudly, through bids that took an evening to price and never got followed up.
01 / Two businesses in one
Most companies do both, and the mistake is running them as one thing. A recurring residential customer and a commercial bid are different businesses that happen to share a ladder.
Small tickets, high frequency, and density is everything. The revenue is not in winning new customers, it is in nobody forgetting that the Hendersons are due again in April. Miss that and you have quietly lost a customer without a single complaint.
Bigger tickets, longer cycles, and a bid process. Storeys, access, frequency, insurance certificates, net-30 terms and a decision-maker who is not the person you met on site. The leak here is the bid you priced and never chased.
02 / The road
Between somebody deciding their windows are filthy and you being paid for cleaning them, there are six places the job can fall on the floor.
Page three for "window cleaning" and your town, and a Google listing with two photos from 2019.
A site built around the towns you actually serve, and a listing that agrees with it.
WebsiteA site that looks like a hobby, on a phone, in four seconds, against a competitor who looks insured.
Your real work, your real van, your insurance said out loud. This is most of the sale.
WebsiteA three-field form, then six texts to establish storeys, access, frame cleaning and how many panes.
A request form that asks the questions you would have asked, so the first reply can be a number.
WebsitePer-pane on Tuesday, per-hour on Thursday, and a commercial bid that takes an evening.
Your pricing rules, applied the same way every time, by anyone.
AirtableThursday exists on a whiteboard. Recurring customers get remembered, or they do not.
Routes, frequency and next-due dates that surface themselves instead of waiting to be remembered.
AirtableInvoices chased by text, and no idea which commercial account is quietly at 60 days.
Invoices raised from the job, and one screen showing who owes what and for how long.
AirtableThe single biggest hole in a residential window cleaning business is not lead generation. It is that a customer who was due in March never got called, never complained, and simply stopped being a customer. Nobody notices that leak because nothing about it looks like a problem.
03 / What I'd build
Which one depends entirely on where your road is leaking. Nobody needs them all at once, and starting with everything is how a project stops halfway.
Website
Found in the towns you serve, credible in four seconds on a phone, and a request form that asks what you would have asked. Five business days, and you write none of the copy.
See how it works →Airtable
Quoting, jobs, scheduling and reporting in one place, built around how you already work. Start with the single piece costing you the most. Your account, your data.
See how it works →04 / The obvious question
Both are genuinely good at this and if you are a one-van operation running standard residential work, buy one and stop reading. Where they stop fitting is when the commercial side gets real — multi-visit contracts, per-building pricing, certificates and renewal dates — or when your pricing has rules the form cannot express. Then you end up with the software and a spreadsheet next to it, which is the point where a build pays for itself.
The full comparison →05 / Fit
Before you commit to anything
The assessment reads the whole path a job takes — how it reaches you, how it gets captured, and where it lands once it does. Those last two are exactly what a base or a custom tool fixes, so the answers tell us whether a build is worth doing and which piece is worth building first.
Three findings on screen — the joins where information is getting re-typed from one tool into another.
My read on which of it is a systems problem, which is a front-door problem, and which is neither. Written by me, not generated.
A fair number of these end with "your existing tool already does that." Cheaper to find out here than after a quote.
Q3 2026 · finish it and you are entered to win a website built free · closes September 30
The prize is a website build, which may well not be what you came here for — take the assessment for the read on your operation and ignore the giveaway if it is not relevant to you.
The newsletter
Short, practical notes for people running a shop: what to automate, what to leave alone, and what your software is quietly costing you in time. Written from inside a working print shop, not from a product roadmap.
Free · one click to leave
Join solo operators and small teams who would rather fix the system than work more hours.
Free · no pitch · reply within one business day
Tell me how work reaches you and I will tell you which of the six is costing you most. Free, and if the answer is "buy Jobber", that is what I will say.
Short on time? Get just the free site audit — two minutes, your URL and your town.