Last updated: September 12, 2026
What apartment mystery shopping buys
Apartment mystery shopping buys a handful of staged inquiries per property per year, scored by a contractor against a rubric. Scoring recorded leasing calls covers every real inquiry instead of a sample of staged ones. The two methods do different jobs, since a shop can test the tour and what happens after it and call scoring cannot.
A shop is a staged inquiry. Somebody trained calls your property as a prospect, works from a scenario, takes the agent through the greeting, the qualifying questions, the price conversation and the ask for the tour, then scores what happened against a rubric your regional manager signed off on. Weeks later the report arrives as a PDF, sometimes with the audio attached, scored by a person who has never heard a real call at your property.
Nothing about that is broken. The scenario holds steady across properties, so a regional manager ends up with a number they can bring into a review. A staged call is also the cleanest way to watch an agent handle somebody with no history at the property.
The gap is coverage. Four shops in a year sit beside several hundred real inquiries that nobody scored, and the calls that produced your actual signed leases are all in the second pile.
What a shop report usually contains
Grace Hill, MarketStat and Reality Based Group are the names that come up when operators talk about who runs their shops. Layouts differ between vendors, and most reports carry the same eight things.
| Field | What it carries |
|---|---|
| Property and date | Which property was shopped, plus the date and time of the call |
| Scenario | The story the shopper worked from, such as a one bedroom for a move six weeks out |
| Agent | Who picked up |
| Scored criteria | The rubric, usually the greeting, discovery, the presentation, objection handling and the close |
| Score | A percentage or a point total against that rubric |
| Narrative | The shopper's written account of the call in their own words |
| Audio | A recording of the staged call, on programs that include one |
| Comparison | The property against the portfolio, and sometimes against its own last shop |
The narrative is the part a manager reads twice. Say a property comes back at 78. The number says there is a problem somewhere, and the paragraph underneath it saying the agent quoted a price for a floor plan the caller never asked about is the thing that gets fixed on Monday.
The coverage arithmetic
Run the numbers on a portfolio you can hold in your head. Say twelve properties, four shops at each one per year, and 60 inbound inquiries per property per month. Swap in your own figures as you read, because the point survives whatever you put in the slots.
Four shops at twelve properties is 48 staged calls for the year. Sixty inquiries a month at twelve properties is 8,640 real ones. The shop program covers 48 of 8,640, which is a little over half a percent.
At the property level it reads worse, because four calls out of 720 is the sample your regional manager is reasoning from when they decide whether that leasing team can sell. Nobody listened to the rest of it.
What half a percent can and cannot tell you
A shop program is good at catching a systemic script failure. If the agent never asked for the tour on any of the four calls, that is real, you found it, and no larger sample was needed to find it.
What it cannot see is the agent who is fine on four calls a year and poor on four hundred. With four shops, one bad call is a quarter of that property's annual quality record, so a single agent having a rough Tuesday moves the number as much as a genuine coaching problem does.
A staged call is also the one call an agent has any chance of recognizing as staged. Though a good shopper is hard to spot, and the agents who catch one tend to be the people already doing the job well, so call that the smallest of the three problems.
The cost comparison, with your own numbers
Cost cells stay yours to fill in. Quotes move with the market, the scenario, and whether the program includes an on site visit, and a made up number would be the least useful thing on this page.
| What you buy | Calls it covers in a year | What it evidences | Your cost |
|---|---|---|---|
| Four shops at each of twelve properties | 48 staged calls | How an agent handled a caller with no history at the property, and the tour when the program includes one | Your shop invoice |
| Managers listening to recordings when they get a minute | About 170 calls at two percent, roughly 30 hours of manager time | Whatever the managers had time to open, held mostly in their memory | Manager hours |
| Scoring every recorded inquiry | 8,640 real calls | What every agent said on every inquiry, with the recording and the timestamp attached to the score | Your quote |
That middle row is the one most operators are actually running, whatever the org chart says. Manual review across the industry lands somewhere around one to three percent of calls, per Verint and similar write ups, and nobody sets out to review two percent. It is just what fits in a week.
What a shop does that call scoring cannot
This is the part a vendor page usually skips, so here it is up front. A recording carries the words and none of the building.
The shopper can tell you the model unit smelled like paint, that the leasing office was locked at 2 p.m. on a Wednesday, that there was nowhere to park, and whether the email they were promised ever showed up. They can tell you whether they would have rented there, which is a judgment no rubric produces and no transcript contains.
A shop also runs the same scenario at every property, so the comparison across your portfolio is clean. Real inquiries are not controlled. One property gets a morning of callers asking about a floor plan you have eleven of, another gets six people asking about the one unit you do not have, and the scores move for reasons that have nothing to do with the leasing teams.
So call scoring replaces the phone portion of a shop program and leaves the rest alone. Anybody telling you it replaces the whole thing has not walked a property.
What call scoring does that a shop cannot
Every inquiry, every agent, every day, scored against the same rubric, with the recording sitting next to the score. A month of that is a few thousand calls at a portfolio of twelve, which is enough to see one agent's pattern instead of one agent's Tuesday.
The consistency question is the one your risk people care about. Whether every caller got the same qualifying questions is answerable only when every call was scored against one list, and a sample of four cannot answer it either way. That sits in full under fair housing consistency on the phone, and the phrasing your agents use is a scored criterion on the leasing call scorecard. Your counsel decides what the question set should be, since nothing here is legal advice.
The honest limit sits on the other side of the scorecard. Criteria written as rules hold up well, and anything that needs a read on tone gets shakier, which Aircall and StackAI both note about automated scoring. Every flag carries a timestamp into the transcript and a confidence score, so a supervisor spends their hour on the low confidence ones and skips the rest. A scorecard still will not tell you whether the agent sounded like somebody you would want to rent from.
Your recordings come out of whatever your phone system or CRM already keeps, which is why scoring calls without a phone system project is the version that starts this month rather than next quarter.
Mapping your shop rubric onto a scorecard
The criteria on your shop report mostly survive the move, which matters, because a regional manager who has reported against the same rubric for four years should not have to explain a new standard to owners in the same quarter the measurement changes.
Weights are deliberately absent from this table. The point values live on the leasing call scorecard, which is the one place they are maintained, so the two pages cannot drift apart.
| What the shop report grades | The scored criterion on a recorded call | Auto fail |
|---|---|---|
| Greeted the caller and named the property | Opened with the property name from the script | no |
| Asked how the caller heard about the property | Captured the lead source before the price conversation | no |
| Asked qualifying questions | Asked from the qualifying set the property uses for every caller | yes |
| Presented availability and pricing | Quoted availability and price from the current rent sheet | no |
| Described the floor plan and community | Described the floor plan the caller asked about | no |
| Handled the objection | Answered the price or availability objection without ending the call | no |
| Asked for the appointment | Asked for the tour | no |
| Confirmed the appointment | Read the day, the time and the address back to the caller | no |
| Captured contact information | Captured a callback number and read it back digit by digit | no |
| Rapport and enthusiasm | Handled the caller as a person rather than as a form | no |
| Sent the email afterward | No criterion. The recording ends when the call ends | no |
| Curb appeal, model unit and tour | No criterion. None of it happens on the phone | no |
Those weights are ours. If the report you already send to owners weights these differently, keep its weights, since the argument here is about coverage rather than about rubric design. The mapping downloads as a CSV with no form in front of it, and the script the criteria came off sits at a leasing call script that doubles as a scorecard.
Two rows on that table have no criterion, which is the mapping telling you where the shop budget still belongs.
Running both for a quarter
Cancelling the shop program on the strength of a blog post would be a strange way to make a budget decision. Run them side by side for a quarter instead.
Score every recorded inquiry at the properties you pick, keep the shops on their normal schedule, then put the two outputs next to each other at the end of it. The question to answer is narrow: did the shops surface anything the scored calls missed, and did the scored calls surface anything four staged inquiries could never have reached. One quarter of that is a better argument with your owners than any comparison table, including the one above.
The two properties to start with
Pick the two with the widest gap between traffic and signed leases. High inquiry volume and a soft conversion number means the calls are arriving and something is happening to them, and those are the properties where a few hundred scored calls pay for themselves fastest.
Keep shopping those properties, though fewer times, and point what is left at the tour. The thing a shop buys that nothing else does is the experience of a prospect with no history at your property, walked through the model unit by an agent who does not know they are being measured. That is a read on how your agent handles a person, which is different from how your agent handles a script, and it is worth paying for on its own terms. The phone portion was never worth four calls a year, and that is the line to move.
A first pass on your own audio scores a handful of your own recordings against your rubric so you can see what a month of them would look like, and call QA for property management covers what that turns into once it runs every day. Pull last month's recordings for those two properties and score them against the same criteria your shop reports already use.
Frequently asked questions
How much does apartment mystery shopping cost?
Shop programs price per shop, and the number moves with the scenario, the market, and whether the program includes an on site visit alongside the phone call. Bundled annual pricing across a portfolio is common. Get a current quote rather than working from a figure on somebody's blog, since we are not publishing one we cannot source.
What is included in an apartment mystery shopping report?
Most reports carry the property and date, the scenario the shopper worked from, the agent's name, the scored criteria, a total score, a written narrative, the audio when the program includes it, and a comparison against the rest of the portfolio. The table earlier on this page lays out all eight fields.
How often should you mystery shop a property?
Quarterly is the common cadence and some operators buy monthly, though neither is a standard anybody publishes. Whatever you pick, set it against your inquiry volume before you decide what it proves. Four shops a year against 720 inbound calls is a different claim than four shops a year sounds like.
Is mystery shopping worth it for apartments?
For the tour, the model unit, and the experience of a prospect with no history at the property, yes, and nothing else tests those. On the phone the sample is too small to tell you much about an individual agent, and the recordings of every real inquiry are already sitting in your phone system.
What is an alternative to mystery shopping?
Scoring your recorded leasing calls against the same rubric, on every inquiry instead of a sample. It covers only what happened on the phone, so it sits alongside a reduced shop program rather than replacing one.
Sources
- Mystery shopping as the incumbent QA method in multifamily, covering under one percent of calls, and Grace Hill, MarketStat and Reality Based Group as the vendors operators name. ICP expansion brief, 2026-09-08. Client research, unverified independently.
- Manual QA review covering roughly one to three percent of calls. Verint, Gistly and ViciStack. Third party, unverified, used as setup only.
- Rule based criteria scoring more reliably than tone and empathy. Aircall and StackAI. Third party, unverified.
- The twelve property arithmetic, the four shops a year, and the 60 inquiries a month. Editorial hypothetical, stated as such and meant to be replaced with your own numbers.
- Mystery shop pricing. No verified figure turned up in the research pass, so this page publishes none.
- Trinsic feature sheet and call notes, 2026-08-27 and 2026-09-10. Verified internally.
- The point weights in the mapping table. Our own weighting for this page.