SAMPLE: a fictional business. Not a client of Adwaizer.
AI Audit report: Larchmere Home Comfort Ltd. (sample)
Prepared by: Adwaizer AI Consulting Inc. · Engagement: Adwaizer AI Audit (2 weeks, fixed price USD 1,500) · Report version: sample, October 2026
What this page is. This is the report a client keeps at the end of the Adwaizer AI Audit. Larchmere Home Comfort is invented to show the format and depth; the business, its people and its numbers do not exist. In a real audit, every figure comes from the client's own records or interviews, and every estimate says what it rests on.
How to read the numbers. Nothing in this report was measured. Figures marked (owner's estimate) are what the fictional owner told us in interviews. Figures marked (estimate) are ours and name their assumption next to them. Costs are in Canadian dollars (CAD), before tax.
1. Executive summary
Larchmere Home Comfort is a residential heating, cooling and plumbing service company in Ottawa with 24 people. Its office team of four spends much of the week on work that follows a fixed pattern: answering the same customer questions, typing technician notes into invoices, chasing maintenance-plan renewals and filling in rebate paperwork. The owner's concern is that the office cannot keep up in the two peak seasons (first cold snap, first heat wave) without hiring.
We mapped 10 workflows and scored 6 AI opportunities. Our recommendation:
- Start with technician job notes to invoice and customer summary (opportunity O2). It is low risk, uses tools Larchmere already has, and gives the office back an estimated 6 to 8 hours a week.
- Next, a maintenance-plan renewal assistant (O3), which protects recurring revenue the owner believes is leaking today.
- Then an after-hours intake assistant (O1), once the job notes and renewal workflows have shown the team that AI drafts can be trusted and checked.
We advise against automated quoting and AI-written rebate submissions for now (section 5). The 90-day plan in section 6 delivers the first opportunity in weeks 1 to 6 and starts the second in weeks 7 to 12. Estimated cost of the three recommended opportunities: CAD 9,000 to 16,000 one-time plus CAD 1,500 to 4,200 a year to run (section 7), against an estimated time saving of 12 to 18 office hours a week in peak season once all three run.
2. The business (fictional)
| Item | Detail |
|---|---|
| Name | Larchmere Home Comfort Ltd. (invented) |
| Location | Ottawa, Ontario; serves the city and nearby towns |
| Sector | Residential HVAC (furnaces, heat pumps, air conditioning) and plumbing: repairs, installs, maintenance plans |
| People | 24: owner-operator, office manager, 2 dispatchers / customer service, 1 bookkeeper, 2 estimators, 14 licensed technicians, 3 apprentices |
| Customers | About 3,000 active households (owner's estimate); about 900 on an annual maintenance plan (owner's estimate) |
| Revenue | Not disclosed for this sample |
| Current tools | A field-service app for scheduling, dispatch and invoicing; QuickBooks Online; Google Workspace (Gmail, Calendar, Drive, Sheets); a shared spreadsheet for maintenance plans; a phone system with voice mailbox; paper or phone-photo forms for rebate paperwork |
| Who we spoke with | Owner, office manager, both dispatchers, bookkeeper, one estimator, three technicians (eight 30 to 45 minute interviews) |
3. Current-state findings
F1. The office is the bottleneck in peak season, not the technicians. In the first cold week of the year the dispatchers answer about 120 calls a day against about 50 on a normal day (owner's estimate). Calls go to the voice mailbox when both lines are busy, and after-hours voice messages are returned the next morning.
F2. Job notes are re-typed. Technicians write notes in the field-service app in their own shorthand. The office manager rewrites them into invoice line items and a customer-facing summary, about 4 minutes per job, longer for installs (owner's estimate), on about 150 jobs a week averaged over the year and 35 to 45 jobs a day in peak season (owner's estimate).
F3. Maintenance-plan renewals live in a spreadsheet. Nobody owns the renewal list. The owner believes 15% to 20% of plans lapse each year without anyone asking the customer (owner's estimate).
F4. Quotes for installs take 3 to 5 working days (owner's estimate) because estimators write them after hours from site notes and photos.
F5. Rebate paperwork is error-prone. Utility and government rebate forms for heat pumps and furnaces are filled in by hand from invoices. The bookkeeper reports two to three forms a month sent back for missing details (owner's estimate).
F6. Data is in reasonable shape for a business this size. Customer, equipment and job history are in one system (the field-service app). Maintenance-plan data is the exception (spreadsheet, inconsistent dates).
F7. No AI policy exists. Two staff already paste customer messages into a free consumer chatbot to draft replies. That puts customer names and addresses into a tool the company has not reviewed.
Readiness (our scoring, 1 = weak, 5 = strong)
| Dimension | Score | Why |
|---|---|---|
| Data | 3 | Job and customer data in one system; maintenance plans in a spreadsheet |
| Tools | 4 | Cloud tools with export and integration options already paid for |
| People | 3 | Office team open to help; technicians wary of "more typing" |
| Process | 2 | Peak-season steps are informal and depend on two people |
| Governance | 1 | No AI policy; unreviewed consumer tool in use with customer data |
These scores are our judgement from the interviews, not a measurement.
4. Workflow map and scored AI opportunities
4.1 Workflows mapped (10)
| # | Workflow | Hours a week, office (owner's estimate) | Error risk | Data sensitivity | Fit for AI |
|---|---|---|---|---|---|
| W1 | Answering calls and booking jobs | 40+ in season | Medium | Medium (names, addresses) | Medium |
| W2 | After-hours voice message follow-up | 4 | Medium | Medium | High |
| W3 | Job notes to invoice and summary | 8 to 10 | Medium | Medium | High |
| W4 | Maintenance-plan renewals | 3 (when someone gets to it) | High (missed renewals) | Low | High |
| W5 | Install quotes | 6 (estimators, after hours) | High (pricing) | Low | Medium |
| W6 | Rebate paperwork | 4 | High (rejections) | High (owner details, account numbers) | Low for now |
| W7 | Online review replies | 1 | Low | Low | High |
| W8 | Supplier invoice entry | 3 | Medium | Low | Medium |
| W9 | Technician scheduling | 10 | High | Low | Low (the field-service app already does this) |
| W10 | Hiring and onboarding apprentices | 2 | Low | Medium | Low |
4.2 Scored opportunities
Scoring: Value (time or revenue protected), Effort (how hard to build and run; 5 = easiest), Risk (what could go wrong for customers or the business; 5 = safest). Each is 1 to 5; total out of 15. Scores are our judgement, with the reasoning in the last column.
| # | Opportunity | Value | Effort | Risk | Total | Reasoning |
|---|---|---|---|---|---|---|
| O2 | Job notes → invoice lines + customer summary. AI drafts both from the technician's notes and photos; the office manager checks and sends. | 4 | 4 | 4 | 12 | High volume, a human checks every draft, data stays in reviewed tools |
| O3 | Maintenance-plan renewal assistant. Clean the plan list, then send each customer a personalised renewal reminder drafted by AI and approved in a batch. | 4 | 4 | 4 | 12 | Protects recurring revenue; low data sensitivity; small build |
| O1 | After-hours intake assistant. A web and SMS form, with an AI agent that answers common questions from Larchmere's own service pages and books a call-back; urgent words (gas smell, no heat with infants or seniors) go straight to the on-call technician. | 5 | 3 | 3 | 11 | Biggest peak-season gain; safety routing must be tested hard |
| O4 | Review reply drafts. AI drafts replies to online reviews for the owner to approve. | 2 | 5 | 4 | 11 | Quick win, small value; good first exposure for staff |
| O5 | Quote first drafts. AI turns estimator site notes and photos into a quote draft from Larchmere's price book. | 4 | 2 | 2 | 8 | Valuable, but pricing errors cost money; price book not yet structured |
| O6 | Rebate form pre-fill. AI fills rebate forms from invoices and equipment records. | 3 | 2 | 1 | 6 | Sensitive personal and account data; program rules change; errors delay customer money |
Top 3 for Larchmere: O2, O3 and O1, in that order. O4 is a quick win that can ride along with O2 at almost no extra cost.
4.3 What could go wrong, per top opportunity
| # | Main risk | How we would contain it |
|---|---|---|
| O2 | A draft adds a part or charge the technician did not use | Every draft is checked against the job's parts list before sending; nothing is sent automatically |
| O3 | A reminder goes to a customer who already renewed or moved | Clean the list first; batch approval by the office manager; easy opt-out in every message, per Canada's anti-spam law (CASL) |
| O1 | An urgent safety call is treated as routine | Fixed keyword and question rules route to the on-call technician before any AI step; the agent never gives repair advice; tested with 50 scripted calls before launch |
5. Do not do yet
| Idea | Why not now | Revisit when |
|---|---|---|
| Automated quoting sent straight to customers (beyond O5 drafts) | Pricing errors cost money and trust; the price book is not structured | The price book is in one structured list and O5 drafts have been right for 3 months |
| AI-filled rebate submissions (O6) | Sensitive personal and account data; program rules change often | A reviewed, paid AI tool and a written AI policy are in place |
| An AI voice agent answering the main line | Customers in a no-heat emergency want a person; risk to reputation | After-hours intake (O1) has run one full season |
| Replacing the field-service app | It works; switching costs far exceed any AI gain | Not recommended |
6. 90-day roadmap
| Weeks | Work | Owner (Larchmere) | Measure of success |
|---|---|---|---|
| 1 | Write a one-page AI use policy; move staff off the free consumer chatbot to a reviewed business tool | Owner | Policy signed by all office staff |
| 1-2 | O2 design: collect 50 past jobs, agree the invoice and summary format, set up the drafting step inside the current tools | Office manager | Format signed off |
| 3-4 | O2 pilot on real jobs, every draft checked; O4 review drafts added | Office manager | 80% of drafts need only small edits (target, not a forecast) |
| 5-6 | O2 live for all jobs; time per job tracked for 2 weeks | Office manager | Time per job down from the baseline the team records in week 1 |
| 7-8 | O3: clean the maintenance-plan list into one structured source | Bookkeeper | Every plan has a customer, start date and renewal date |
| 9-10 | O3: renewal reminders drafted and approved in weekly batches | Office manager | Renewal rate tracked against last year's list |
| 11-12 | O1 design and safety test plan (build starts after day 90) | Owner, on-call lead | 50 scripted test calls written and agreed |
Measures are set by the client in week 1. We do not promise a saving; we agree how it will be measured before work starts.
7. Costs (CAD, estimates)
All figures are estimates for this fictional business. A real proposal replaces them with vendor quotes and a fixed Adwaizer price.
| Item | One-time | Yearly running | Assumption |
|---|---|---|---|
| AI Audit (this report) | USD 1,500 fixed | none | Adwaizer's published audit price; billed in USD |
| O2 job notes drafting (build and setup) | 3,000 to 5,000 | 600 to 1,500 | 3 to 5 days of build at a small-firm consulting rate; running cost assumes a business AI subscription for 2 office users plus model usage on about 8,000 jobs a year |
| O4 review reply drafts | 0 to 500 | included above | Rides on the O2 setup |
| O3 renewal assistant | 2,500 to 4,500 | 300 to 900 | 2 to 4 days to clean the list and set up drafting; running cost assumes a reminder or forms tool for about 900 customers |
| O1 after-hours intake (build after day 90) | 3,500 to 6,000 | 600 to 1,800 | 4 to 6 days including safety testing; running cost assumes a forms or agent tool plus SMS charges at peak-season volume |
| Total, three opportunities | 9,000 to 16,000 | 1,500 to 4,200 | Sum of the rows above; the audit fee is separate |
Value side (estimate). If O2 cuts the office manager's 4 minutes per job to about 1 minute of checking, on about 8,000 jobs a year, that is about 400 office hours a year, or 6 to 8 hours a week (estimate: 8,000 jobs from the owner's estimate of about 150 jobs a week; 1 minute of checking is our assumption, to be confirmed in the O2 pilot). If O3 brings back even 50 of the plans the owner believes lapse each year, that protects their yearly plan fees (estimate; the plan price is not given in this sample). These are the numbers the week-1 baseline will confirm or correct.
8. Tool choices
Larchmere already pays for a field-service app, QuickBooks Online and Google Workspace, and is not asking to switch. Every opportunity above is designed to work with those tools first.
| Need | Option we would put in front of Larchmere | Why it fits |
|---|---|---|
| O3 renewal tracking and O1 call-back queue, if the team wants one board for office work | monday.com (work management or monday CRM), as an option next to the existing spreadsheet and field-service app | One shared board for renewals and call-backs, with owners and due dates, without changing the field-service app |
| O1 after-hours intake form and simple answering agent | Jotform (Forms and AI Agents), as an option | Quick to stand up a mobile form and an agent that answers only from Larchmere's own service pages |
| O2 drafting from job notes and photos | A Claude or OpenAI build by Adwaizer, connected to the field-service app's export | Custom logic and checks against the parts list need a small build |
We're a monday.com partner, so we may earn a referral fee; the recommendation stands on fit. We are experienced with Jotform. Larchmere can stay on its current spreadsheet and field-service app for all of the above; the partner tools are options, not conditions.
9. What Larchmere would buy next
| Next step | What it is | Indicative scope |
|---|---|---|
| Build package 1: O2 + O4 | Job notes drafting and review replies, built, tested and handed over with a one-page runbook | Fixed scope and fixed price, about 2 weeks (estimate) |
| Build package 2: O3 | Renewal list clean-up and renewal reminder drafting | Fixed scope and fixed price, about 2 weeks (estimate) |
| Build package 3: O1 | After-hours intake with safety routing and the scripted test plan | Fixed scope and fixed price, 3 to 4 weeks (estimate), after packages 1 and 2 |
| Staff training | Two 90-minute sessions on safe AI use and checking AI drafts, built on the AI use policy from week 1 | Half a day (estimate) |
Each package is quoted separately after this audit. Larchmere can also take this report and build any of it with its own team or another provider.
10. How the work would be delivered
- Certified delivery: PMP and PMI-ACP certified; fixed scope and fixed price.
- Weekly progress: small deliveries every week, each shown in a live demo.
- No surprises: a risk and issue log shared with you, and written acceptance criteria you sign off.
Sample report. Larchmere Home Comfort Ltd. is fictional; any likeness to a real business is a coincidence. Adwaizer AI Consulting Inc. promises no savings or funding outcome; every figure above is an estimate or an owner's estimate as labelled.