The evening enquiry
Most enquiries arrive when the office is shut. The ones answered next morning are often already viewing with someone else.
See how the agents workExample · Real estateEnquiry and viewing scheduler
The enquiry at 21:40, answered before nine
An agent that answers questions about a listing from the record, books the viewing inside the diary rules, and passes anything about price, offers or eligibility to the agent.
Illustrative scenario — not a real client result
Listing enquiries · 21:40
- Answered
6 enquiries answered
21:41 · Rent, deposit and available date from the listing record
- Answered
Two viewings booked
21:43 · Saturday slots the agency opened, 20 minutes each
- Person
“Would you take an offer?”
21:44 · Price and offers · with the negotiator
Nothing about price or offers comes from the agent
Enquiry and viewing scheduler · Real estate
- Sourced facts
- The agent negotiates
- Listing record only
01 — A typical day
Three small failures that add up.
What real estate owners tell us goes wrong in an ordinary week.
A detail guessed from memory
“Is the parking included?” The answer is in the listing, but it is repeated from memory, and it is wrong.
See what the agent may sayScreening decisions nobody audits
An automated score decides who gets a viewing. It was never tested for who it excludes.
See the fair-housing cases01 — A typical day · the evidence
What the numbers actually say.
Each figure opens its source, year, sample and limits. Click a badge to check it.
01
Source
NAR 2025 Technology Survey, methodology reported by HousingWire
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source20%
of REALTORS® say they use AI daily
NAR 2025 Technology Survey: 1,241 responses from 49,233 invited members, July 2025.
NAR 2025 Technology Survey · 202502
Source
NAR 2025 Technology Survey, methodology reported by HousingWire
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source46%
say AI has had no noticeable impact on their business
The same survey. Adoption is not the same as a result, and this page does not claim one.
NAR 2025 Technology Survey · 202503
Source
NAR 2025 Technology Survey, methodology reported by HousingWire
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source66%
name saving time as the reason to adopt new tools
Same survey. A stated motivation, not measured hours saved.
NAR 2025 Technology Survey · 202502 — What the helper does
From your records to a decision you own.
The agent reads the enquiry against the listing record: rent or price, deposit, available date, viewings and what is included. If a detail is not in the record, it says so.
It books the viewing inside the diary rules the agency sets, sends the confirmation and directions, and logs the enquiry against the listing.
Anything about price, offers, eligibility or a complaint goes to the named negotiator. The agent does not screen applicants and does not advise on a tenancy.
6 steps · 2 decisions you own
Reads from
- Portal leads
- SMS
- Listing record
Text version of the flow
- 1. Read the enquiry — Which property, what question.
- 2. Answer from the record — Only what is written down.
- 3. Book the viewing — Agency slots only. Human checkpoint: You set the viewing rules.
- 4. Send the practical detail — Where, when, what to bring.
- 5. Stop at price and offers — Offers, eligibility, negotiation. Human checkpoint: A negotiator handles price and offers.
- 6. Log it against the listing — One enquiry, one record.
02 — What the helper does · a simulated example
One simulated example, step by step.
Use the buttons or the left and right arrow keys. Nothing here is a real customer.
Simulated example
Aldergrove Lettings, a six-person agency
Portal lead and SMS · 21:40
An agency with 40 managed properties. The agent answers portal leads outside office hours. Every message is simulated; no real applicant data is used.
This is Aldergrove Lettings' AI assistant. I can answer questions about the listing and book a viewing. Reply HUMAN for the office.
03 — What you decide
The helper prepares. You decide.
The agent answers from the record. The negotiator decides.
The listing is the only source
If a field is empty, the agent says it is unconfirmed instead of filling the gap.
No screening decisions
The agent does not score applicants or decide who may view. Eligibility goes to a person.
A negotiator takes over
Offers, reductions, negotiation and complaints go to the named negotiator with the thread.
You can pause it
Pause per listing, per office or entirely; enquiries fall back to the shared inbox.
Without your approval · No valuation, no offer handling, no eligibility decision, no advice on a tenancy. Anything outside the listing record is a question for a person.
04 — What changes when a safeguard is missing
Documented cases, not scare stories.
Three documented cases with their sources: two about automated decisions in housing, one about an automated valuation. None is a result claimed by an agency.
A screening score that excluded voucher holders
Applicants alleged that a tenant-screening score leaned on credit history, did not account for housing vouchers, and gave Black and Hispanic voucher holders low scores used in rental decisions. Fair Housing Act claims proceeded and a settlement received final approval in November 2024 — a settlement, not a finding of liability at trial.
What the helper would do differently
The agent makes no screening decision at all. It books viewings, answers listing questions and passes eligibility to a person, so no score ever stands between an applicant and a viewing.
$2.275M
settlement, with injunctive relief
Housing ads delivered unequally
The Department of Justice alleged that Meta's housing-ad targeting and delivery could distribute housing ads discriminatorily, including through protected-characteristic and lookalike targeting. The settlement required a new system and government-approved monitoring.
What the helper would do differently
Where the agent is used for listing promotion, delivery outcomes are checked for who actually sees the ads — not just which settings were chosen.
$115K
civil penalty, plus a rebuilt ad system
A valuation model, and the capacity behind it
Zillow stopped buying homes after concluding its price forecasting was too unpredictable, reporting a $422 million loss in its homes segment and a $304 million write-down, alongside a backlog in renovations and closings.
What the helper would do differently
The agent never values a property and never promises a price. Estimates stay with a person, and the page says plainly that the agent answers from the listing record only.
$304M
write-down as iBuying was wound down
04 — What changes · the evidence
The rest of the evidence.
Every figure resolves to a source with its sample and limits. No figure without one.
20%
of REALTORS® use AI daily
NAR 2025 Technology Survey, 1,241 responses. Self-reported member survey.
Source
NAR 2025 Technology Survey, methodology reported by HousingWire
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source1,241
usable responses in the NAR survey
From 49,233 members invited, July 2025. The NAR release confirms the findings.
Source
NAR 2025 Technology Survey, methodology reported by HousingWire
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source46%
report no noticeable business impact from AI
Same survey. The page does not claim an impact it cannot show.
Source
NAR 2025 Technology Survey, methodology reported by HousingWire
2025
Method and sample are recorded in the source registry.
Read the source for its scope and limits.
Open the source
04 — What changes · your week
The same week, two ways.
Drag the handle, or use the left and right arrow keys when it is focused.
Typical week, beforeTypical week, after
Mon
Forty weekend leads in the inbox; the negotiator starts with the newest.
Tue
Two applicants arrive for a viewing that was cancelled on Saturday.
Wed
A parking question is answered from memory and turns out to be wrong.
Thu
An offer question sits unanswered for two days; the applicant goes elsewhere.
Mon 08:00
Weekend leads answered; viewings already in the diary.
Mon 08:05
Unlisted details are flagged to the negotiator instead of guessed.
Tue
Every viewing on today's list was confirmed the evening before.
Thu
Offer questions reach the negotiator the same day, with the thread attached.
Illustrative example · not a measured result
05 — Your version
Four questions about your own week.
Keep your answers. They are the brief for the first version we would build for you.
How many enquiries arrive per listing in a week?
What may be answered from the listing record, and what must come from the agent?
Which diary rules decide a viewing slot?
What always goes to the agent: price, offers, eligibility?
Next
Put your own figures into the calculator below. It runs in your browser, and nothing is sent or stored.
Put your week into the calculatorBring the four answers to the free one-hour call. We will use them to sketch the first version, and you keep the written plan either way.
Talk about your business05 — Your version · your numbers
Put your own week into it.
An estimate, not a promise. The model and its assumptions are printed below.
Your inputs
All listings, deduplicated by applicant.
count
Availability, rent, deposit, viewing times.
%
Reading, replying, arranging the viewing.
min
Negotiator or administrator pay plus employer costs.
per hour
Evenings and weekends.
%
From lead received to a human reply.
hours
Example defaults are typical for a small business. Replace them with yours. Nothing is stored, sent or shared.
Estimated result
IllustrativeNegotiator hours back each month
0–0
- Time released at your rate
- £0–£0
- Leads answered outside office hours
- 0–0%
- Time to first reply
- Already immediate
How we calculated this
Routine work handled by the helper: 70% of the share you marked routine.
Time saved per item: 80% of the minutes you entered.
Monthly volume: weekly figures × 4.33; monthly figures used as entered.
Cost: hours saved × your staff cost per hour, in pounds as an illustration.
Reply time: routine items move within the hour while the helper runs. Complex work still waits for a person.
Questions
The things people ask next.
No. Valuation, offers, reductions and negotiation go to the negotiator. The agent gives no figure and makes no promise about a landlord's decision.
No. It does not score, rank or exclude anyone. Eligibility and referencing are handled by a person, and the agent books viewings for whoever asks.
The listing record only. If a field is empty — parking, bills, a date — the agent says it is unconfirmed and flags it to the office.
Yes. The first message says it is the agency's AI assistant and that a person is available by replying HUMAN.
It acknowledges the complaint and stops. Complaints go to the named person, with the thread, and the agent sends nothing further on that thread.
Through the CRM's API where one exists, otherwise a scheduled export. Listings refresh before the agent answers, so it never quotes a stale price.
About two weeks: listing feed, diary rules, approved wording, and a parallel run on one office before the rest.
Start here
Show us a week of leads.
One hour with last week's enquiries and your diary rules. We will show you how many the agent would have answered, and where it would have stopped.