6 min read
This is an illustrative story. Rosa is a composite based on common situations in small cleaning businesses, not a real person, and the numbers are realistic examples rather than reported results. The steps and tools are real and you can copy them.
It is 9:40 on a Tuesday night, and Rosa Delgado is sitting in her car outside her last job of the day, scrolling through six new quote requests. One says “3 bed 2 bath, how much?” Another has four blurry photos of a kitchen and no address. A third wants a move-out clean “this Friday.” She knows from experience that whoever replies first usually wins the job, and she also knows she will not get to any of them until tomorrow evening.
Rosa runs a one-woman residential cleaning business. She is good at cleaning and good with clients. What was slowly sinking her was quoting: the back and forth to get the basics, the guesswork on price, and the two-day gap before a customer got a number.
Where the time was really going
When Rosa finally sat down and counted, the problem looked like this:
- About 15 quote requests a week, arriving by text, Facebook messages and her Google Business Profile.
- Most requests were missing at least two key details (square footage, pets, how long since the last deep clean, parking).
- Each request took three or four messages to fill in, often spread over a day or two.
- Her pricing lived in her head, so quotes for similar homes varied by $40 or more depending on how tired she was.
- Average time from request to quote: roughly two days. She estimated she lost a third of leads simply because someone else replied faster.
She did not want a chatbot pretending to be her, and she did not want to pay for complicated software. She wanted a faster, fairer way to get from “how much?” to a real number.
Step 1: Writing down the price list that lived in her head
The first breakthrough had nothing to do with AI. Rosa spent one evening writing her pricing rules down in plain sentences: a base price by bedrooms and bathrooms, an add-on for homes over 2,000 square feet, extra for pets, a deep-clean multiplier if the home had not been professionally cleaned in three months, flat add-ons for inside the oven and inside the fridge, and a move-out rate.
Then she opened a project in her AI assistant, pasted the rules in as instructions, and asked: “Using only these rules, what would you charge for a 3 bed, 2 bath, 1,800 square foot home with one dog, last cleaned six months ago? Show the math.” She checked the answer against what she would have charged. It matched within a few dollars. She ran ten past jobs through it and adjusted two rules that turned out to be vaguer than she thought.
That alone made her quotes consistent for the first time.
Step 2: A one-message intake that gets everything up front
Next, Rosa asked the assistant to help her write a friendly reply that she could send to every new inquiry within minutes, even from a parking lot. The final version thanked the customer, asked six short numbered questions (bedrooms and bathrooms, rough square footage, pets, when it was last deep cleaned, any add-ons, preferred dates) and asked for three photos: kitchen, main bathroom and the messiest room.
She saved it as a text shortcut on her phone and as a saved reply in her Facebook and Google Business Profile inboxes. Our comparison of spreadsheets, free CRMs and AI assistants for tracking leads helped her settle on a simple spreadsheet to log each request.
Customers answered the numbered questions in one go far more often than they answered her old open-ended “can you tell me more about the home?”
Step 3: Photos plus rules equals a draft quote
When the answers and photos came in, Rosa pasted them into the same AI project with a short prompt: “Draft a quote using my pricing rules. Point out anything in the photos that suggests heavy buildup or clutter I should ask about. Keep the message warm and under 120 words.”
The assistant does not decide the price. It calculates from her rules and flags things she might miss, like a stovetop that clearly needs extra time. Rosa reads every draft, adjusts if needed and sends it herself. Nothing goes to a customer without her looking at it first.
She also made one firm rule about privacy: she never pastes in a full address, door codes or alarm codes. The quote only needs the neighborhood. Details like access codes go in her own notes on the day of the job, not into any AI tool.
Step 4: Following up without feeling pushy
About half of quotes used to get no reply at all. Rosa had the assistant write two short follow-up messages: one for two days later (“Just checking this works for you, happy to adjust the date”) and one for a week later offering her next open slot. She set calendar reminders to send them. It takes her about ten minutes a week in total.
The results after two months (illustrative)
- Time from request to quote dropped from about two days to about two hours on most requests.
- Time spent on quoting fell from roughly six hours a week to under two.
- Quotes for similar homes became consistent, which cut down on awkward “my neighbor paid less” conversations.
- She booked noticeably more of the jobs she quoted, mostly because she was now often the first reply.
The total cost was one AI subscription she already had for writing social posts. No new software, no website chatbot, no automation platform.
What Rosa would tell another solo service owner
- Write your pricing rules down before touching AI. The assistant is only as consistent as the rules you give it.
- Fix the intake first. Most of the delay was missing information, not slow math.
- Keep yourself as the sender. Customers hire a person. The AI drafts; you decide.
- Keep sensitive details out. Addresses, codes and payment details do not belong in a chat window.
- Test on old jobs. Running ten past quotes through the rules shows you where they are fuzzy.
This same pattern (clear rules, a better intake, AI drafts, human sends) works for any trade that quotes jobs: lawn care, painting, pet sitting, handyman work. It is the same idea that helped the solo mechanic in our story about not losing calls while under the hood. Once quotes go out quickly, the next win is reviews, and our guide to answering every Google review in 15 minutes a week picks up from there. If your business gets inquiries through Google, it is also worth reading Google’s own Business Profile help center to make sure your hours, service area and messaging settings are accurate, so the leads arriving are the ones you actually want.
What is the slowest step between “how much?” and a booked job in your business? Tell us in the comments, and we may turn it into a future walkthrough.
Related reading
- Spreadsheet vs Free CRM vs AI Assistant: The Best Way for a Solo Business to Track Leads
- How a Solo Mobile Mechanic Stopped Losing Calls While Under the Hood (An Illustrative Story)
- How to Answer Every Google Review With AI in 15 Minutes a Week (Without Sounding Like a Template)
- Google Will Let a Free AI Agent Text Your Customers. Fix Your Business Profile Before You Say Yes
