Key takeaways
- Trust is low: only 37% of people said they trust estate agents in the Ipsos Veracity Index cited by the government[1].
- Test the calls that matter most: valuation requests, viewing requests, rental enquiries and out-of-hours calls.
- Score behaviours you can coach (questions asked, next step booked) rather than vague impressions.
- Share results as team patterns first and individual coaching second, or staff will see the programme as a trap.
Why test your own phones
Managers rarely hear the calls that go wrong. The negotiator who forgets to ask about a buyer's position, the call that rings out at lunchtime and the lettings enquiry that ends with "check the website" all pass without anyone noticing, until the vendor instructs elsewhere or a complaint arrives. Mystery shopping your own branch gives you a sample of what callers actually experience.
There are good reasons to look. Only 37% of people told Ipsos they trust estate agents[1]. Contact with the Ombudsman is rising too: it received 23,987 phone and email enquiries in 2025, up from 19,793 the year before[2]. And poor, unclear or delayed communication was the first issue listed in lettings disputes[3]. Much of that communication starts on the phone.
Source: The Property Ombudsman, 2025[2]
Show the data
| Item | Value |
|---|---|
| 2024 | 19,793 |
| 2025 | 23,987 |
Choose scenarios that reflect real demand
Pick four to six scenarios and run each several times across different days and times. Volume tells you where to focus: Propertymark branches registered an average of 69 new prospective buyers in August 2026[4], while letting branches averaged 121 new tenant registrations[5]. A lettings-heavy branch should weight its tests that way.
- Seller asking for a valuation, with a property to sell in three months and an onward purchase.
- Buyer wanting to view a specific listing, in rented accommodation with a mortgage agreed in principle.
- Rental applicant asking whether a flat is still available and what is needed to apply.
- Landlord asking about fully managed services for a single flat.
- Tenant reporting a non-urgent repair.
- A call at 12:45 or 17:40, when the office is stretched.
The scorecard
Score what you can coach. "Friendly" is too vague to help anyone improve. "Used the caller's name and stated a next step" is specific. Keep it to around ten criteria so assessors can mark consistently.
| Criterion | What good sounds like | Points |
|---|---|---|
| Answered promptly | Within five rings, or a clear message if busy | 1 |
| Greeting | Branch name and the person's name | 1 |
| Captured contact details | Name, number and email, read back | 1 |
| Qualifying questions | Position, timescale, finance or, for lettings, move date and household | 2 |
| Product knowledge | Correct facts about the property or service, or an honest "I'll check" | 1 |
| Compliance | Advertised rent only; no promises on price; fees explained accurately | 1 |
| Next step booked | Viewing, valuation or callback with a time | 2 |
| Follow-up sent | Confirmation by email or text within the hour | 1 |
For lettings tests, add a specific check that the negotiator quotes only the advertised rent and treats each applicant the same way, which is now a legal requirement in England. The rules are covered in rental viewings under the Renters' Rights Act.
Running the programme fairly
Tell the team
Announce that test calls will happen this quarter and share the scorecard. The aim is better calls, not catching people out.
Use realistic callers
Friends of the business or a specialist firm, briefed with a full persona. Never use real customers' details.
Respect data rules
Test callers should use contact details they control, and test leads should be removed from the CRM after scoring.
Record the calls
If your system records calls, mark test calls so they can be reviewed against the scorecard.
Score in pairs
Two people scoring the same call keeps marking consistent.
Report patterns
Share branch-level results first: what most calls did well and the two biggest gaps.
Coach individually
Play back one strong call and one weak call privately with each colleague.
Writing a persona brief
A test call only works if the caller sounds like a real customer. Give each mystery shopper a one-page brief so they can answer follow-up questions without hesitating, and so calls in the same scenario are comparable.
- Who they are: name, household, current home and why they are moving.
- Their position: chain, mortgage agreed in principle, cash, or renting with a move date.
- What they want: the listing or service, preferred viewing times, budget.
- Two questions to ask: for example, the council tax band and whether the seller has found a property.
- One curveball: a request the branch may not expect, such as a viewing at 7pm or a pet in a rental.
- What to do at the end: accept the next step offered, then cancel politely by email the next day.
Cancelling test appointments quickly matters. A negotiator who blocks out a Saturday slot for a buyer who never existed is wasting real capacity, and the trust in the programme drops fast if that happens often.
Coaching on the results
Start with what went well, then pick one behaviour to change. A negotiator who books viewings but never asks about a buyer's position needs one new question, not a lecture on the whole call. Role-play the fix immediately: "Before I book you in, are you in a position to proceed, or is there a property to sell first?"
Rule of thumb: one behaviour per person per month. Retest that behaviour, not the whole scorecard.
Use the results to fix systems too. If most failed tests were at lunchtime, the problem is cover, not people. If lettings callers keep being told to "check the website", the team may not have the property facts to hand.
Example: what one round revealed
Example (fictional): a two-branch agency runs 24 test calls in October. Average score is 7 out of 10. Every negotiator greets well and captures details, but only a third of valuation calls ask about the seller's onward purchase, and four calls after 5.30pm go unanswered. The agency adds the onward-purchase question to its call guide, sets up an overflow for evening calls and retests six weeks later. Valuation calls now ask the question every time and the evening calls are all answered.
Common failings and quick fixes
- No qualifying questions on viewing calls: add a two-question prompt to the call guide (position and finance) and display it by the phone.
- Valuation calls that never ask why the seller is moving: the motivation question shapes the whole appraisal, so make it the first question after the address.
- Lettings callers told to apply online with no next step: book the viewing on the call and send the application link afterwards.
- Facts got wrong, such as council tax band or tenure: keep an up-to-date property sheet for every live listing where phone staff can see it.
- No confirmation sent: use a short template that can be sent from the CRM in one click.
- Calls ringing out at busy times: set a hunt group so calls move to another desk or an overflow service after a few rings.
Most fixes are small. The point of testing is to find the handful of habits that cost the most, then make the right behaviour the easiest one.
Tracking results over time
A single round is a snapshot. The value comes from repeating the same scenarios each quarter and watching the scores move. Keep a simple sheet with one row per test call: scenario, date, time, branch, total score and the score for each criterion. After two or three rounds you will see which behaviours have stuck and which slip back when the branch is busy.
- Look at the criterion with the lowest average first, not the lowest individual score.
- Compare time bands: mornings, lunchtimes, late afternoons and Saturdays often differ more than people do.
- Compare sales and lettings separately, because the questions and the compliance checks are different.
- Note unanswered calls as a separate measure. A branch that scores well on answered calls but misses many is still losing business.
Include the hard calls
Most programmes only test friendly buyers. Add at least one difficult scenario each round: a frustrated tenant, a caller with limited English or someone who sounds confused about the process. These show whether the abusive callers policy and vulnerable customer guidance are actually used. The Ombudsman decided 61% of the cases it ruled on in 2025 in the consumer's favour[6], and the calls that lead to complaints are rarely the easy ones.
Messaging channels deserve the same treatment. Send a test enquiry by WhatsApp and through your website form, and score the response time and quality the same way.
Frequently asked questions
How many test calls do we need?
Enough to see patterns. Three or four calls per scenario per branch each quarter is a practical start; one call tells you very little.
Is it fair to test staff without warning?
Tell the team the programme exists and share the scorecard, but not the timing of individual calls. That keeps calls natural while keeping trust.
Can we use real enquiry recordings instead?
Yes, reviewing real recordings is useful alongside test calls, provided your privacy notice covers it. Test calls let you control the scenario and compare like with like.
What should we fix first?
The gaps that cost instructions or create complaints: unanswered calls, missing next steps and compliance errors on lettings enquiries.
References
Ministry of Housing, Communities and Local Government (GOV.UK) (2024). Home buying and selling reform (consultation) (opens in a new tab) ↑
The Property Ombudsman (2025). Annual Review 2025 (opens in a new tab) ↑
The Property Ombudsman (2025). Annual Review 2025 (opens in a new tab) ↑
Propertymark (2026). Housing Insight Report: August 2026 (opens in a new tab) ↑
Propertymark (2026). Housing Insight Report: August 2026 (opens in a new tab) ↑
The Property Ombudsman (2025). Annual Review 2025 (opens in a new tab) ↑