Trades and home services: quotes out faster, and followed up every time
An illustrative electrical, solar and HVAC business with 25 technicians takes enquiries through four channels, takes days to quote and re-keys every job into Xero. One intake, AI-drafted quotes approved by an estimator, automatic follow-up and a proper Xero connection release office time and stop paid-for quotes going cold.
This is a worked example: a realistic scenario we've built to show how we'd approach the problem. The organisation is fictional, and every number is a stated assumption or a calculation from those assumptions. It is not a client case study and makes no claim about results.
The situation
Work arrives through four doors: the phone, the website form, a shared inbox and texts to individual technicians. The office logs what it can into the job-management platform, but some enquiries live in a notebook or on a technician's phone until someone remembers them.
Install quotes need a site visit. Technicians send back photos and voice memos, and the estimator writes each quote from scratch, usually in the evening. Quotes take days to go out, and nobody owns follow-up on the ones that haven't been accepted.
When a job is finished, the accounts officer re-keys materials, invoice lines and customer details into Xero, because the two systems were never fully connected. Review requests go out only when someone remembers on a Friday afternoon.
Constraints we'd work within
- The job-management platform works well for scheduling and dispatch, and technicians know it. Replacing it would cost more in retraining and lost bookings than it would return.
- Technicians are on the tools all day. Anything that adds steps on site, or needs a laptop, won't be used.
- Quotes carry safety, warranty and pricing commitments, so the estimator must approve every scope and price before anything reaches a customer.
- Customer names, addresses and photos of homes are personal information. With turnover above the Privacy Act's $3 million small-business threshold, the Australian Privacy Principles apply, including to any AI tool that reads site photos.
- Follow-up and marketing messages must meet Spam Act requirements, and changes have to land in stages around the summer air-conditioning peak.
What the Diagnostic would find
Findings by lens, with the commercial effect of each. See how the Diagnostic works.
| Lens | Observation | Commercial effect |
|---|---|---|
| Operations & process | Enquiries arrive through four channels and are logged by whoever picks them up. There is no single list of open enquiries, so response time depends on who happens to be in the office. | Slow first responses on high-value install enquiries, at the exact moment the customer is collecting other quotes. |
| Workforce capability | The estimator writes install quotes from photos, voice memos and memory, formatting each one from scratch. Scope wording varies from quote to quote. | Quote turnaround measured in days, regular estimator overtime, and loose scope descriptions that turn into variations and disputes later. |
| Systems & technology | The job-management platform and Xero exchange invoice totals only. Materials, customer changes and payments are re-keyed, and the two systems hold different versions of the same customer. | Office hours lost to duplicate entry, plus mismatched records that slow debtor follow-up and month-end. |
| Growth & marketing maturity | No defined follow-up on unaccepted quotes and no routine review request after completed jobs. | Quotes the business has already paid to produce go cold, and the Google Business Profile shows few recent reviews for the volume of work done. |
| Data & reporting | The owner sees revenue in Xero and jobs in the job-management platform, but not enquiry source, time to quote or quote win rate in one place. | Marketing spend and hiring decisions are made on feel, and nobody can see which job types or suburbs win most often. |
| AI readiness & opportunity | Site photos, notes and several years of past quotes already sit in the job-management platform, and most install work falls into a small number of standard assemblies. | A realistic first AI use with a clear human approval step: drafting quotes against a price book, rather than a speculative project. |
The moves
Every recommendation resolves to one of nine moves: build, buy, improve, integrate, automate, remove, change, train or leave alone. Building is only one of them.
- Leave alone
The job-management platform's scheduling, dispatch and technician app. They work, the team knows them, and the gains sit in the gaps around the platform, not inside it.
- Buy
A business phone and SMS service that logs every call against the customer and texts back missed calls, rather than building any telephony.
- Integrate
Web form, shared inbox and call log feed one enquiry record in the job-management platform, and completed jobs flow to Xero with materials, payments and customer details, so nothing is typed twice.
- Improve
The price book: standard install assemblies, labour times and exclusions written down once, so quotes are consistent and an AI draft has something reliable to work from.
- Build
A quote-drafting assistant that reads the technician's checklist, photos and voice note, matches them to the price book and produces a draft for the estimator to edit and approve. Nothing is sent without approval.
- Automate
Quote follow-up until each quote is won or closed, booking confirmations and day-before reminders, and one review request after every completed and paid job.
- Change
One named owner for the open-enquiry and open-quote lists each morning, with agreed targets for time to first response and time to quote.
- Train
Short sessions for technicians on capturing site photos and voice notes to a checklist, and for the office on checking AI drafts and handling exceptions.
The system, end to end
Every enquiry becomes one record in the job-management platform, the site visit adds photos and notes, and AI drafts the quote for the estimator to approve. Accepted quotes become scheduled jobs, completed jobs flow to Xero, and every step feeds follow-up messages, review requests and a daily dashboard.
- Phone, web form and email
- One enquiry record
- Site visit capture
- AI quote draft
- Estimator approval
- Follow-up and reminders
- Job management to Xero
- Reviews and dashboard
Enquiry intake
Web form, shared inbox and phone-system call log mapped into the job-management platform as one record with source, suburb and job type. Customers get an immediate acknowledgement with a realistic next step.
Quote-drafting assistant
AI reads the site checklist, photos and voice note, matches them to price-book assemblies and drafts the scope, exclusions and options. The estimator edits and approves, and every edit is logged so the price book improves over time.
Follow-up and reminders
SMS and email sequences for unaccepted quotes, booking confirmations and day-before reminders. Consent is recorded at enquiry, every message identifies the business, and commercial messages carry an unsubscribe honoured within five working days.
Job management and Xero integration
Customers, invoices, payments and materials kept in step between the two systems, with an exceptions queue for the accounts officer when something doesn't match, instead of silent overwrites.
Review requests
One request to every customer after the job is complete and paid, with no incentive and no filtering out of unhappy customers. Google's review policy prohibits selectively asking for positive reviews, and the ACCC warns against suppressing negative ones.
Operations dashboard
Enquiries by source, time to first response, time to quote, quote win rate, value of open quotes, integration exceptions and reviews received, refreshed daily for the owner and operations manager.
Delivery stages
Indicative timings.
Diagnostic and process map
Time with the office and two technician ride-alongs to map every enquiry and quote path, audit the price book and integration options, and agree the opportunity portfolio, baseline and targets.
2 to 3 weeks (indicative)
Intake and follow-up
One enquiry record, acknowledgements, quote follow-up and job reminders go live first, because they need the least change on site and show results quickly.
3 to 4 weeks (indicative)
Xero integration
Build and test against a copy of the Xero file, run both ways in parallel for one month-end, then stop manual re-keying.
4 to 6 weeks (indicative)
AI quote drafting
Price-book clean-up, then a pilot on one job type such as split systems, comparing AI drafts with the estimator's own quotes before extending to solar and electrical work.
6 to 8 weeks (indicative)
Reviews, dashboard and handover
Review requests switched on, dashboard live, written procedures for the office, and a 90-day check against the baseline.
2 to 3 weeks, plus a 90-day review (indicative)
The arithmetic
Assumptions first, then the calculations they produce. Change an assumption and the result changes with it, which is the point.
| Assumption | Value |
|---|---|
| Jobs completed per month | 600 today, growing 10% next year if nothing changes (assumption) |
| Office time re-keying each job between systems | 8 minutes (assumption) |
| Share of re-keying removed by the integration | 75%, with the rest handled as exceptions (assumption) |
| Install quotes written in the office per month | 120 (assumption) |
| Time per install quote | 60 minutes today; 35 minutes to check and approve an AI draft (assumption) |
| Loaded hourly cost (wage plus on-costs) | Office $45, estimator $70 (assumption) |
| Install quote acceptance rate | 35% today; 38% with consistent follow-up (assumption) |
| Average accepted install quote | $4,500 ex GST (assumption) |
Re-keying load today, and the do-nothing baseline
Today: 600 jobs × 8 min ÷ 60 = 80 hours a month. Next year with no change: 660 jobs × 8 min ÷ 60 = 88 hours a month.
80 office hours a month, rising to 88 (modelled)
Office hours released by the Xero integration
600 jobs × 8 min × 75% ÷ 60
60 hours a month (modelled)
Estimator hours released by AI-drafted quotes
120 quotes × (60 − 35) min ÷ 60
50 hours a month (modelled)
Value of capacity released
(60 h × $45) + (50 h × $70) = $2,700 + $3,500 = $6,200 a month; $6,200 × 12
About $74,400 a year of capacity, redeployed to faster responses, more quotes and debtor follow-up rather than cut (modelled)
Extra work won through consistent follow-up
120 quotes × (38% − 35%) × $4,500 = 3.6 quotes × $4,500 = $16,200 a month; $16,200 × 12
About $194,400 a year in accepted quote value, before job costs (modelled; revenue, not profit)
What we'd measure
- Median time from enquiry to first response, by channel
- Median time from site visit to quote sent
- Quote acceptance rate, and value of open quotes older than 14 days
- Share of AI-drafted quotes approved with minor edits, and the price changes the estimator makes
- Office hours on re-keying and integration exceptions, from a weekly time sample
- Google reviews received per month, and the share of completed jobs that received a request
Risks and how we'd manage them
An AI draft gets the price or scope wrong
The estimator approves every quote before it is sent, the assistant works only from the price book and the site record, and edits are tracked so recurring errors are fixed at the source. We pilot one job type before extending it.
Site information is captured inconsistently
A short in-app checklist with required photos, tested with two technicians before rollout. Poor capture means poor drafts, so we measure checklist completeness weekly and coach where it slips.
The integration corrupts accounting records
Testing happens on a copy of the Xero file, both systems run in parallel for a month-end, and any mismatch goes to an exceptions queue for the accounts officer rather than overwriting a record.
Messages annoy customers or breach the Spam Act
Consent recorded at enquiry, the business identified in every message, a working unsubscribe on commercial messages, frequency limits, and sequences that stop the moment a quote is accepted or declined. We check against the rules; your lawyer confirms anything uncertain.
The takeaway
Keep the platform that works and connect what sits around it. In this example the gains come from one intake, quotes drafted for an estimator to approve, and follow-up that never depends on someone remembering.
Services in this example
- Business process automationAutomate intake, quoting, follow-up, onboarding and reporting across the systems you already use. Measured in hours released, with people on exceptions.
- System integrationsConnect your CRM, Xero, job system and email so nobody re-keys data. APIs, webhooks and sync rules, plus monitoring that alerts you to failures first.
- Email & marketing automationLifecycle email and SMS flows from enquiry to win-back, built around Spam Act consent rules and authenticated for deliverability. See what's included.
- AI-augmented workforceHelp your existing team do more valuable work with AI: role-based training, approved tools, prompt libraries and adoption tracking. Capacity, not cuts.
- Reporting & dashboardsOne source of truth for finance, sales, marketing and operations: agreed definitions, automated data pipelines and AI commentary that explains the numbers.
Recognise your business in this?
A 30-minute call is enough to tell whether the same moves would pay off for you.
Sources
Third-party facts on this page, checked 14 October 2026.
- Avoid sending spam, Australian Communications and Media Authority. www.acma.gov.au/avoid-sending-spam
- Spam email and texts, Australian Communications and Media Authority. www.acma.gov.au/what-spam
- Prohibited and restricted content (Maps user-generated content policy), Google. support.google.com/contributionpolicy/answer/7400114?hl=en
- Managing online reviews, Australian Competition and Consumer Commission. www.accc.gov.au/business/advertising-promoting-your-business/managing-online-reviews
- Small business (Privacy Act coverage), Office of the Australian Information Commissioner. www.oaic.gov.au/privacy/privacy-guidance-for-organisations-and-government-agencies/organisations/small-business