Home / Case studies
Case studies
What a full review actually
turns up
One measured deployment, reported in full including what we have not measured — plus the scenarios we would examine in other industries, clearly marked as scenarios rather than dressed up as clients we do not have.
How to read this page. The case study below is a real 30-day deployment at an Australian business, anonymised because we do not have written consent to name them, and every figure in it was counted on their calls. The section after it sets out what a review examines in other industries and the patterns that come up.
Measured deployment
30 days, 2,349 calls, one Australian business
A corporate travel agency. Every recorded call in scope was read, and 45 were raised to a manager. Then we checked what the phone system's own sentiment score had said about those same 45.
The finding that mattered was not the count. It was the concentration: 36 of the 45 were the same category of failure, and they traced overwhelmingly to one internal system that kept letting customers down the same way. Not forty-five bad calls. One problem, surfacing forty-five times.
What a sentiment threshold would have done instead
Acting on scores below 40 would have surfaced 3 of the 45 — and returned a clean bill of health on the other 42. The calls that needed a manager sat at a median of 82 out of 100, because a capable agent stays warm and professional while the business quietly fails the person on the phone.
What we did not measure
Precision was checked against the client's own verdicts, call by call. Recall was not measured — the share of all real problems the review caught. Measuring it honestly means a human listening to a large random sample of the calls that were not flagged, and we have not done that. Any vendor quoting you a recall figure without describing that exercise is guessing.
Client not named: we do not have written consent to identify them. No call content, customer names or booking references are published anywhere on this site. Figures are as measured over the 30 days to mid-2026.
By industry
What a review looks for in other industries
The calls a review examines in each sector, and the pattern that usually emerges — so you can judge whether it would find anything in your business before you spend a minute on it.
🏠Property management
Where this applies. Suburban agencies with a few hundred to a few thousand properties, several property managers and a shared reception line.
What a review would examine:
- Every call where a repair is promised, then checked against whether a job was raised
- Tenants who ring three or more times about one matter inside a fortnight
- Faults described calmly that meet the urgent-repair definition
- Calls where the tenant ends up with no owner and no date
What usually surfaces. Repeat contact tends to cluster on a handful of properties and a handful of trades, not across the whole portfolio. That turns a vague sense that “maintenance is chaotic” into a specific list to fix on Monday.
🔧Trades & home services
Where this applies. Plumbing, electrical, HVAC and similar businesses running a fleet with one or two people answering the phones.
What a review would examine:
- Quotes promised on a call, checked against whether one was actually sent
- Appointment windows missed with no outbound call beforehand
- Back-ordered parts where the customer chased rather than being told
- Second and third visits for the same fault
What usually surfaces. The most common finding is not rudeness or bad work. It is quotes discussed warmly on the phone that never left the office — revenue that was already won and then quietly lost.
🩺Medical & allied health
Where this applies. General practices, dental and allied health clinics, single or multi-site, with a reception team taking the calls.
What a review would examine:
- Patients told they would be called with results, checked against whether a call happened
- Referrals nobody can locate when the patient rings
- Appointment changes made at short notice
- Distressed callers who end the call without a booked next step
What usually surfaces. Administrative failures are experienced by patients as being forgotten, and they are almost always polite on the phone. They are close to invisible to any monitoring that looks at tone.
🚗Automotive service
Where this applies. Dealership and independent service departments taking bookings, progress updates and pick-up calls.
What a review would examine:
- Vehicles returning a second time for the same complaint
- Quotes approved on the phone and not actioned
- Promised pick-up or completion times missed without an outbound call
- Customers chasing a status update more than once
What usually surfaces. Repeat visits for one fault are the cheapest quality signal a service department has, and almost nobody counts them, because each visit is logged as a separate job.
Being straight with you
Why there is only one real case study here
Because there is only one measured deployment to report.
WiseSentry is a young product from a telephony business, not a decade-old platform with a wall of logos. We could fill this page with invented companies and invented percentages, and you would have no way to check any of it — which is exactly the problem with most vendor case studies.
The alternative we prefer: run it over your calls for a month, and let the case study be yours. You will see every flag, and every call we deliberately left alone.
Make the next case study yours
A month of your own recordings, reviewed in full. If it finds nothing worth acting on, that is a useful answer too.