Key takeaways
- Test what a stranger experiences: the phones at 12.45pm, the web form on a Friday evening, the email reply on a Monday.
- Use written scenarios and a fixed scoring sheet so results compare month to month and between teams.
- Tell staff that testing happens, never who or when, and use results to fix process before judging people.
- Close the loop within a week: one owner, one change, one retest.
What mystery shopping tells you that reports do not
Call statistics show that the phone was answered. They do not show whether the caller was told the firm handles their kind of matter, whether anyone asked for a callback number before transferring, or whether the promised email arrived. A mystery shop is a structured test of the experience from the outside, run by someone the team does not recognise, scored against the standards the firm has already set.
Prospective clients are testing you anyway. When choosing a main adviser, 25% of people in the Legal Services Board's Legal Needs Survey searched around, 15% obtained details of the service, 14% searched for reviews and 13% searched for prices[1]. In 2026, 42% of legal services consumers shopped around before choosing, and 69% made their choice within a week[2]. Each of those people is, in effect, a mystery shopper who does not report back.
Source: Legal Services Board, 2024[1]
Show the data
| Item | Value |
|---|---|
| Searched around | 25% |
| Got service details | 15% |
| Searched for reviews | 14% |
| Searched for prices | 13% |
Decide what you are testing
Pick the channels and the moments where you suspect problems, not the ones you are proud of. A useful first round covers each channel at least twice, at different times of day, across your two or three highest-volume matter types.
| Channel | When to test | Scenario | Main thing to observe |
|---|---|---|---|
| Phone, main number | Monday 9.05am | Buyer with an accepted offer asking about conveyancing | Greeting, price information offered, next step agreed |
| Phone, main number | Lunchtime | Parent asking about child arrangements after separation | Whether the call is answered, empathy, no advice given |
| Web form | Friday 7pm | Employee dismissed last week asking about a claim | Acknowledgement wording and time to human reply |
| Email to info@ address | Tuesday mid-morning | Executor asking what probate involves and costs | Plain English, link to published prices, booking offer |
| Missed call, no voicemail | Any | Hang up after the greeting | Whether anyone calls the number back, and when |
| Website review | Desk test | Look for price pages, SRA badge, complaints procedure | Whether a stranger finds them in two clicks |
Keep the first round small enough to finish in a fortnight. A dozen contacts, scored consistently, will show more than fifty contacts scored loosely. Include at least one test outside office hours, because that is where most firms assume the website or voicemail is doing the work and rarely check what a caller actually hears or reads. Include one test of a matter type you do not handle, too: how the team turns someone away says as much about the firm as how it books someone in.
Run the shop in five steps
Write the scenarios in full
Give each shopper a one-page brief: name, phone number and email they control, the facts of the matter, two questions they must ask and one they should push on, such as 'can you just tell me if I have a case?'. Fictional facts only; never use a real client's situation.
Choose shoppers the team will not recognise
Friends of staff in another office, a partner firm's practice manager doing a reciprocal swap, or a specialist agency. Avoid anyone whose voice reception knows.
Score immediately, against the sheet
The shopper completes the scoring sheet within an hour of each contact, while details are fresh, and notes the exact words used for anything marked as a fail.
Withdraw cleanly
If the firm books a consultation, the shopper cancels it promptly by the normal route, so no fee earner sits waiting and no diary slot is lost.
Collate and compare
The practice manager collects scores by channel, time and matter type, not by named individual, and looks for patterns across at least six contacts before drawing conclusions.
Tell staff in writing that the firm tests its enquiry handling from time to time, why it does so and how results are used. Do not tell them who or when. Testing people by surprise and then disciplining them destroys the trust the exercise depends on.
A worked scenario brief
Example brief (fictional): shopper name Priya Shah, using a pay-as-you-go mobile and a new email address. Situation: she and her partner are separating after nine years, they are not married, they own a house together and have a seven-year-old daughter. She has found the firm through a search and wants to know whether the firm can help and what a first meeting involves.
- Must ask: do you deal with this kind of thing, and how much is a first appointment?
- Should push on: 'I just need to know if I can stay in the house, can you tell me that now?'
- Must not volunteer: the daughter's school or any detail that would make the scenario identifiable as a test.
- Ends the call by: accepting whatever next step is offered, then cancelling by email the following morning.
- Records: the time the call was answered, the name given, whether price information was offered, the exact words used when she pushed for advice, and when any follow-up email arrived.
The push question is the most revealing part of most shops. A strong answer acknowledges the worry, explains that the solicitor will look at exactly that question in the meeting, and offers the earliest appointment. A weak answer either gives a view on the house, which is advice from someone not qualified to give it, or says 'we can't advise on the phone' and stops there, which leaves the caller with nothing. Both are fixable with a short script, and the guide to callers who want free legal advice has wording to start from.
The scoring sheet
Score each item yes, partly or no. Keep the list stable so results compare across quarters. Base it on what consumers say builds trust: being a specialist (92%), being regulated (91%) and explaining things clearly (91%) were the factors most likely to increase trust in a legal professional in 2026[3].
- Answered or acknowledged within the firm's own published standard.
- Greeting gave the firm name and the person's name, at a pace the caller could follow.
- The person established the type of matter within the first minute.
- The caller was told clearly whether the firm handles this matter, and if not, where to go.
- No legal advice or view on prospects was given, but the caller was not made to feel brushed off.
- Price information was offered or a route to it explained, including the published price page where one exists.
- A specific next step was agreed: a booked time, a callback window or a referral.
- Contact details were taken and read back before any transfer.
- Any written follow-up arrived when promised and matched what was said.
- Language was plain throughout, with legal terms explained when used.
Plain language deserves its own line on the sheet. 41% of adults who used a legal adviser said communication could be improved by using plain language[1], and receptionists often absorb the firm's jargon without noticing.
Reviews: look, do not plant
A shop often includes a look at the firm's online reviews, because prospects read them. Read what reviewers say about first contact and compare it with your findings. Do not respond to weak results by asking staff, friends or the shoppers themselves to post reviews. Under UK consumer law businesses must not write or commission fake reviews, and businesses that publish consumer reviews must take reasonable and proportionate steps to prevent and remove fake and concealed incentivised reviews[4].
Acting on the results
Sort every failed item into one of three causes before deciding what to do. Most failures in a first round turn out to be process or capacity, not individual performance.
- Process: there is no rule, or the rule is unclear. Example: no one owns missed-call callbacks, so they happen only when someone notices. Fix the rule and name an owner.
- Capacity: the rule exists but cannot be met at that time. Example: lunchtime calls ring out because reception and the secretary take lunch together. Fix the rota.
- Skill or habit: the rule is clear and there is capacity, but it is not followed. Coach privately, with the exact words from the shop and a better version.
Pick one change, give it an owner and a date, and retest that item within a month. Report the before-and-after scores to the whole team, without names. Repeat the full shop quarterly, varying scenarios so the team cannot learn them. If you also want to see how other firms handle the same enquiry, set clear limits first; the guide to benchmarking competitors' client service ethically covers what is fair.
Feed the results into your regular enquiry handling KPIs report as a single quality score alongside speed and booking rate. Numbers show the trend; the shopper's verbatim notes show what to change.
Frequently asked questions
Is it fair to mystery shop your own staff?
Yes, if staff are told in advance that the firm tests enquiry handling, results are reported by channel and time rather than by name, and the aim is to fix process first.
How many mystery shops should a small firm run?
At least two contacts per channel across your main matter types for a first round, then a smaller quarterly repeat with fresh scenarios.
Can we use a real client's situation as a scenario?
No. Use fictional facts. A real client's details would be a confidentiality problem and could identify them to staff.
Should shoppers book and attend consultations?
Test up to the point of booking, then cancel promptly by the normal route. Testing the consultation itself wastes fee earner time and is better reviewed through client feedback.
References
Legal Services Board (2024). Legal Needs Survey 2023: Summary report (opens in a new tab) ↑
Legal Services Consumer Panel (2026). How consumers are choosing legal services (Tracker Survey 2026) (opens in a new tab) ↑
Legal Services Consumer Panel (2026). How consumers are using legal services (Tracker Survey 2026) (opens in a new tab) ↑
Competition and Markets Authority (GOV.UK) (2025). Short guide for businesses: publishing consumer reviews and complying with consumer protection law (opens in a new tab) ↑