Phone enquiries

AI receptionist vs AI employee for letting agencies

AI receptionist or AI employee for a letting agency? Message-taking compared with completing tasks such as viewing requests and repair logs.

Ava research teamPublished 6 min read

On this page
  1. Two different jobs
  2. Side by side
  3. Why message-taking falls short in lettings
  4. Where an AI employee earns its keep
  5. Fair treatment applies to both
  6. The landlord's view
  7. What the AI must never do
  8. Choosing between them
  9. Example: one evening, two set-ups
  10. Frequently asked questions
  11. References

Key takeaways

  • An AI receptionist answers and takes messages; an AI employee completes defined tasks and hands over a finished record.
  • Letting agencies feel the difference most: letting branches averaged 121 new tenant registrations each in August 2026[1].
  • Poor, unclear or delayed communication topped the issues in Ombudsman lettings disputes in 2025[2].
  • Whichever you choose, keep people in charge of offers, referencing decisions, complaints and emergencies.

Two different jobs

The phrase "AI receptionist" covers a wide range of products. At one end is a service that answers the phone, takes a name, number and message, and emails it to the branch. At the other is something closer to a junior colleague: it answers, understands what the caller needs, completes the parts of the job it is allowed to, and passes the rest to the right person with everything they need. In this guide we call the first an AI receptionist and the second an AI employee.

For a sales-only branch, the gap between them matters, but for a letting agency it is much larger. Lettings calls are frequent, repetitive and often urgent, and many of them end in a task: book a viewing, log a repair, send an application link. Propertymark letting branches averaged 121 new tenant registrations each in August 2026[1], with about ten applicants per available property[3]. Each registration is a call or message that someone has to deal with.

Side by side

AI receptionist and AI employee compared for letting agencies
Lettings taskAI receptionistAI employee
Applicant asks if a flat is availableTakes a message for a callbackConfirms status and the advertised rent from approved details, takes the applicant's details
Applicant wants to viewTakes a messageTakes the property, preferred times and household details as a viewing request; negotiator confirms
Landlord asks about managementTakes a messageCaptures the property, number of units and current arrangement, routes to the lettings manager
Tenant reports a leakTakes a message, may not spot urgencyLogs the fault and access details, flags urgency, alerts the out-of-hours contact
Caller asks for a named colleagueTakes a messageWarm transfer if free, detailed message if not
After the callEmail with a short noteRecording, transcript, summary and contact record for the CRM

Why message-taking falls short in lettings

A message is a promise that someone will do the work later. In lettings, later is often too late. An applicant who leaves a message about a flat at 7pm has usually called two other agents by 7.15pm. A tenant reporting water coming through a light fitting needs action, not a note in an inbox for the morning.

Messages also lose detail. "Call back re flat" does not tell the negotiator which flat, when the caller wants to move or whether they have pets. The callback becomes a second call to gather what the first one should have captured. That is where communication breaks down, and poor, unclear or delayed communication was listed first among the top issues in Property Ombudsman lettings disputes in 2025[2]. Lettings disputes resolved by the Ombudsman rose 65% to 3,119 in the same year[4].

Landlords in England using agent services, 2024Managed landlords generate the repair and access calls
Landlords in England using agent services, 2024Column chart. Agent letting services: 43%; Agent management services: 18%.0%20%40%60%Agent letting services: 43%43%Agent letting servicesAgent management services: 18%18%Agent management servicesLandlords in England using agent services, 2024Column chart. Agent letting services: 43%; Agent management services: 18%.0%20%40%60%Agent letting services: 43%43%Agent lettingservicesAgent management services: 18%18%Agent managementservices

Source: Ministry of Housing, Communities and Local Government (GOV.UK), 2024[5]

Show the data
ItemValue
Agent letting services43%
Agent management services18%

Where an AI employee earns its keep

Machine learning tools tend to land first on administrative work. Businesses using machine learning data processing most often report impacts on administrative or clerical roles (41%)[6]. In a letting agency, that admin sits around the phone: logging, routing and chasing. The useful question is which tasks you are comfortable handing over in full, and which you only want captured.

  • Hand over in full: answering availability and fact questions from approved details, routing, logging repair reports with access details, sending confirmations.
  • Capture for a person to complete: viewing requests (until diaries are connected), landlord enquiries, applications.
  • Keep with people: holding deposit decisions, referencing outcomes, choosing between applicants, rent negotiations, complaints and anything involving vulnerability.

Ava is an AI Employee designed for this middle ground: it answers in your branch's name, takes viewing requests for your negotiator to confirm, logs repairs and flags urgent ones, and hands over a recording, transcript and summary after every call.

Fair treatment applies to both

Whatever answers your phones has to follow the same rules as your staff. In England, rental adverts must state a specific rent and agents must not encourage offers above it[7], and discrimination against renters with children or who receive benefits is unlawful[8]. An AI service must give the advertised rent only, never hint that higher offers would help, and ask every applicant the same questions in the same order. Check this in testing by calling as different applicants and comparing the transcripts. More on the rules in rental viewings under the Renters' Rights Act.

The landlord's view

Landlords judge you partly on how their tenants are treated, and partly on how quickly their own calls are answered. A landlord who calls about a void property, a rent review or a new instruction and reaches a message service may assume nobody is minding their portfolio. Landlords with several properties are the most valuable clients: 63% of landlords with five or more properties used an agent for letting, compared with 30% of those with one property[9]. Those are also the clients most likely to notice slow responses across several tenancies at once.

For landlord calls, the difference between the two set-ups is whether the call ends with "someone will call you back" or with the landlord's details, properties and question already with the right manager, and a callback time agreed. The second feels like an organised business.

What the AI must never do

  • Say a property is available when it is let agreed or under offer.
  • Quote any rent other than the advertised rent, or suggest higher offers would help.
  • Tell an applicant they have been accepted or rejected.
  • Give legal advice on notices, deposits or evictions.
  • Promise a contractor visit time it cannot see in a diary.
  • Keep a distressed or vulnerable caller in an automated loop.

These are the same boundaries you would set for a new junior colleague in their first week. Write them into the configuration and check them in testing.

A basic AI receptionist rarely breaks these rules, because it does very little. The risk grows as the system does more, which is why an AI employee needs clearer boundaries and better review, not fewer. Ask any supplier to show you, on a live test call, what happens when a rental applicant offers more than the advertised rent, when a tenant says they are leaving, and when a caller sounds upset. The answers tell you more than any feature list.

Choosing between them

  1. Count your calls by type

    For two weeks, tally lettings calls into availability, viewing, landlord, repair, rent and other. Tasks dominate in most agencies.

  2. Note what happens after each message

    If most messages trigger a callback just to gather details, a receptionist is only moving the work.

  3. Decide your boundaries

    List what you will let AI complete, what it may capture and what it must pass to a person.

  4. Check the handover

    Ask to see exactly what your team receives after a call and whether it lands in your CRM.

  5. Test with real scenarios

    Run the calls from your tally through a trial before going live.

Example: one evening, two set-ups

Example (fictional): between 6pm and 9pm on a Tuesday, a lettings branch receives seven calls: four applicants, one landlord, one tenant with no hot water and one contractor. With a message-taking service, the branch opens on Wednesday to seven emails and spends the first hour calling people back to find out what they wanted; two applicants have already viewed elsewhere. With an AI employee, the four applicants have given their details and preferred viewing times, the landlord enquiry is with the lettings manager with the property address and unit count, the hot water report reached the out-of-hours contractor the same evening, and the contractor was put through to the property manager's voicemail with a summary. The negotiator starts Wednesday by confirming viewings.

The same pattern repeats at lunchtimes and on Saturdays, when negotiators are out on viewings. Over a month, the difference is not one evening of messages but dozens of applicants who were either helped at the moment they called or left waiting until someone was free.

Two notes of caution. First, the comparison only holds if the AI employee works from accurate property details. Second, the team must actually act on the handovers; a perfect summary that nobody reads is just a better-written message. For a wider view of what AI phone answering can and cannot do, see AI phone answering for estate agents, and use the supplier checklist before you sign anything.

Frequently asked questions

Is an AI employee just a better AI receptionist?

It does a different job. A receptionist takes messages; an AI employee completes defined tasks and passes a finished record to your team.

Can an AI employee choose between rental applicants?

It should not. Choosing tenants, referencing decisions and holding deposit questions should stay with people.

Will tenants accept AI for repair reports?

Most callers care that the report is logged properly and urgent issues reach someone fast. Always offer a route to a person, as most adults worry about decisions made without human input[10].

What should we test first?

Your most common lettings calls: availability questions, viewing requests and repair reports, including an urgent one out of hours.

References

  1. Propertymark (2026). Housing Insight Report: August 2026 (opens in a new tab) ↑

  2. The Property Ombudsman (2025). Annual Review 2025 (opens in a new tab) ↑

  3. Propertymark (2026). Housing Insight Report, August 2026 (PDF) (opens in a new tab) ↑

  4. The Property Ombudsman (2025). Annual Review 2025 (opens in a new tab) ↑

  5. Ministry of Housing, Communities and Local Government (GOV.UK) (2024). English Private Landlord Survey 2024: main report (opens in a new tab) ↑

  6. Office for National Statistics (2026). Artificial intelligence in UK businesses: 2023 to 2026 (opens in a new tab) ↑

  7. GOV.UK (2026). Assured periodic tenancies: rental bidding (opens in a new tab) ↑

  8. GOV.UK (2026). Rental discrimination: a guide for landlords (opens in a new tab) ↑

  9. Ministry of Housing, Communities and Local Government (GOV.UK) (2024). English Private Landlord Survey 2024: main report (opens in a new tab) ↑

  10. Department for Science, Innovation and Technology (GOV.UK) (2024). Public attitudes to data and AI: Tracker survey (Wave 4) report (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo