Skip to content
Curious Context Labs

case study · Operations · Custom app

Job reports done before the van leaves the driveway

How a technician's voice note and photos can become a clear report and a draft invoice the same day.

business
Home services and trades
build
Custom app
timeline
5 weeks
read
3 min

01 the business

Commercial refrigeration service company

30 technicians

Technicians write reports in the van or at home that night. Notes are brief, parts get left off and photos stay on personal phones. The office chases details for days before invoices can go out.

02 before

The problem

Technicians finish a job and write the report in the van, or at home that night. Notes are short, parts get left off and photos sit on personal phones. The office chases details for days before the invoice can go out.

  • Invoices go out days after the job.
  • Parts used are missed and never billed.
  • Customers get thin reports that do not explain the work.

03 the build

What the AI does

  1. Step 1: Takes the technician's voice note, photos and signed customer sign-off from a simple phone app.
  2. Step 2: Writes a clear job report: what was found, what was done and what is recommended next.
  3. Step 3: Lists parts and hours used and matches them to items in your price book.
  4. Step 4: Attaches the report and photos to the job in Jobber, ServiceTitan or simPRO.
  5. Step 5: Drafts the invoice in Xero or QuickBooks and sends the customer their report.

paused: waiting for a person to approve

How it was put together

  • Built a simple phone app for voice notes, photos and customer sign-off.
  • Wrote a clear report: what was found, what was done and what is recommended.
  • Matched parts and hours to the price book.
  • Attached the report to the job and drafted the invoice in the accounting system.

04 human in the loop

Where a person stays in charge

The technician checks the report on their phone before it is saved. The office approves the invoice draft before it is sent.

  • The technician checks the report on the phone before saving.
  • The office approves every invoice draft.
  • Recommendations are labelled as recommendations, not done work.

05 rollout

From kickoff to live in 5 weeks

  1. Weeks 1–2Agree the report template and price-book matching.
  2. Week 3Pilot with five technicians.
  3. Weeks 4–5Roll out to all technicians.

06 after

What changes

Reports are consistent and sent the same day, and invoices stop waiting on paperwork. Customers get a clear record of the work, with photos.

worked example · your numbers will differ

Assume 1,200 jobs a month, with 5% missing at least one billable part worth $60 on average.

1,200 × 5% = 60 jobs × $60 = $3,600 a month of unbilled parts, plus days of delay on every invoice.

What to measure

  • Days from job completion to invoice.
  • Unbilled parts found in audits.
  • Office time chasing job details.
  • Customer queries about reports.

07 evidence

What the research says

  1. In a study of 453 professionals doing writing tasks from their own jobs, access to ChatGPT cut the time taken by 40% and raised quality, as rated by evaluators, by 18%.

    Science / MIT, Experimental evidence on the productivity effects of generative AI (Noy and Zhang) (opens in a new tab) · 2023

  2. About 75% of generative AI's potential value sits in four areas: customer operations, marketing and sales, software engineering, and R&D.

    McKinsey & Company, The economic potential of generative AI: the next productivity frontier (opens in a new tab) · 2023

08 lessons

Mistakes to avoid

  • A complicated app. Technicians need three buttons, not thirty fields.
  • Skipping the technician's check. They know what they did.
  • Over-polishing language. Clear beats fancy.

09 questions

Common questions

Does it work offline?
The app can store notes and photos offline and send them when there is signal.
Which job systems does it support?
Jobber, ServiceTitan and simPRO, with invoices in Xero or QuickBooks.

keep reading

All case studies ->

Tell us how your business runs. We'll tell you where AI fits.

A free 30-minute call. You talk, we ask questions, and you leave with a few concrete ideas — or an honest answer that AI isn't the right fix yet.

Working with teams around the world.