What can AI do for a UK property management or lettings business?
Yes, and some of it is already being used by agencies and landlords who've tested it on real portfolios. AI can triage maintenance requests so urgent issues get prioritised properly, handle routine tenant correspondence without your team typing the same answers repeatedly, and track compliance certificates so nothing expires without warning. The question is what's ready now versus what's being sold by people who've never managed a property portfolio.
Property management is a reactive business with regulatory pressure. Your team spends substantial time on work that doesn't grow the portfolio: answering tenant queries, chasing contractors for quotes, tracking down expired gas certificates, updating landlords on routine matters, and documenting everything for compliance. The bit landlords value you for (keeping tenants happy, keeping properties maintained, keeping them legal) takes up far less time than the admin that surrounds every property.
Where the time actually goes
Maintenance requests and contractor coordination. Tenant reports an issue (boiler not working, leak under sink, window won't close). Someone has to read the report, work out how urgent it is, decide whether it needs a contractor or if the tenant can fix it themselves, get quotes if it's expensive, get landlord approval, book the contractor, update the tenant, and chase the contractor if they don't show up. Simple maintenance is straightforward. Vague reports ("something's wrong with the heating") or out-of-hours emergencies take multiple calls and emails to resolve.
Tenant correspondence and routine queries. "When's my tenancy renewal?" "Can I get a reference letter?" "The bins weren't collected, what do I do?" "Can I paint the bedroom?" Your team fields these constantly. Most have standard answers, but each one still takes time to read, look up the specific tenancy details, write a reply, and file it. The queries that need a decision (can the tenant have a cat, will the landlord agree to redecoration) take longer because you need landlord input.
Compliance certificate tracking and renewal chasing. Every property needs gas safety certificates annually, electrical certificates every five years, EPCs valid when you let it, and various other compliance documents depending on the property type. Your team tracks when they're due, books contractors to renew them before they expire, chases contractors who are running late, and updates your records when the new certificates arrive. Miss one and you're letting illegally. The tracking is tedious but essential.
Landlord updates and reporting. Landlords want to know what's happening with their properties: maintenance completed, inspections done, tenant issues, compliance status. Your team compiles updates (monthly reports, ad-hoc emails when something significant happens, annual statements), pulls information from various places (your property management system, email, contractor invoices), and formats it so it's clear. Landlords with multiple properties want consistent reporting. Landlords with one property want personal attention. Both take time.
Calls you miss, and calls that interrupt. Tenants ring about repairs, and they ring when the problem happens rather than when it suits you. Your team is out at a property, on another call, or gone for the day. Some tenants leave a voicemail. Plenty just ring again later, or ring the landlord directly, or let a small problem become an expensive one. The calls you do answer arrive in the middle of something else, so whatever your team was doing stops while they take the details. Out of hours is the worst of it: a leak at 9pm either reaches your on-call person or turns into a complaint and a bigger repair bill.
Three things AI can genuinely do here
1. Triage maintenance requests and prioritise them by urgency automatically
What it does: Tenant submits a maintenance request through your portal or by email. AI reads it, identifies what the issue is (heating, plumbing, electrical, structural, cosmetic), assesses urgency based on keywords and context (no heating in winter is urgent, squeaky door isn't), checks if it's the tenant's responsibility or the landlord's, and routes it accordingly. Emergency issues go straight to your duty person and an approved contractor. Routine maintenance gets added to the queue with a suggested priority. Your team reviews the triage, adjusts if the AI got it wrong, and proceeds.
Why it works for property management: Maintenance triage is pattern-matching with consequences. "Boiler not working, no hot water, temperature outside is 2 degrees" is clearly urgent. "Kitchen tap drips slightly" isn't. AI can do that initial assessment faster and more consistently than someone reading emails as they arrive, especially out of hours. Your team spends their time on the actual coordination and contractor management, not on reading every request to work out what's urgent.
Roughly what it takes: This needs tenants to submit maintenance requests digitally (through a portal, email, or app) rather than by phone. If most of your maintenance requests come in by phone, you'll need to shift that behaviour first. The AI itself is straightforward once you have digital requests, but you'll need to train it on what counts as urgent for your portfolio and your landlords' expectations. And before any of it, the data questions below.
2. Handle routine tenant queries with templated responses, mostly automation
What it does: Tenant emails asking when their tenancy is up for renewal, or requesting a reference letter, or asking about bin collection day. AI reads the query, identifies what they're asking, pulls the relevant information from your property management system (tenancy end date, property address, their tenancy history), and sends a response using your standard templates. Queries that need a decision or aren't routine get flagged to your team instead of getting an automated response.
Why it works for property management: Many tenant queries are asking for information you've already got in your system. Tenancy dates, rent payment history, how to report maintenance, what's covered in their tenancy agreement. A human doesn't add anything to "your tenancy ends on 15th June 2027" except the time it takes to look it up and type the email. And because this is mostly automation rather than AI (looking up data and filling in templates), it's reliable and doesn't need constant retraining.
Roughly what it takes: This needs your property management system to have accurate, up-to-date information and to integrate with your email. If your data is patchy or your system doesn't connect to anything, automated responses will just give tenants wrong information faster. Fix your data quality first, then automate the responses. Start with the simplest queries (tenancy dates, how to report maintenance) and expand gradually. And before any of it, the data questions below.
3. Track compliance certificates and automate renewal reminders before they expire
What it does: Your system tracks every compliance certificate for every property: gas safety due dates, electrical inspection dates, EPC expiry, legionella risk assessments, whatever applies. At 60 days before expiry, it emails the landlord and books a contractor from your approved list. At 30 days, it sends a reminder. At 14 days, it escalates to your compliance person to chase. When the new certificate arrives, it files it, updates the expiry date, and stops chasing. Nothing expires without multiple warnings.
Why it works for property management: Compliance tracking is entirely rules-based and the consequences of missing it are serious. If the gas safety certificate expires in 30 days and you haven't booked a renewal, that's a problem you need to fix now. A human doesn't add anything to that reminder except the risk they might forget. Automation doesn't forget, doesn't go on holiday, and doesn't get distracted by other work.
Roughly what it takes: This needs your compliance certificate dates to be in your system and kept up to date. If you're tracking certificates in spreadsheets, on paper, or in people's heads, you'll need to centralise that first. Once the data's in your system, the automation is straightforward. The hard part is the initial data entry and making sure new certificates get logged when they arrive. And before any of it, the data questions below.
Before any of this touches a tenant's data
Everything above involves software reading correspondence from tenants: names, addresses, maintenance requests, complaints, and sometimes information about personal circumstances, vulnerability or health that tenants share when explaining why something matters to them. That is the data to think about first.
Propertymark published a piece in December 2025 on AI in complaints handling. Its position matches the shape of this page: AI can summarise a long complaint, draft a structured response and flag points to check against your records. Final decisions rest with a trained professional. It must not be used to assess legal risk. That is not formal regulatory guidance, but it is the professional body's own voice, and it lines up with what the ICO expects.
Two of the three opportunities on this page are worth distinguishing. Opportunity 2, templated responses to routine tenant queries, is automation. A trigger in your property management system sends a message; it does not read the tenant's file or send it anywhere. Opportunity 1, reading and triaging maintenance requests, does read the tenant's correspondence before categorising it. Names, addresses, and whatever else is in the message go to a tool. The ICO's questions apply to that tool: where processed, who has access, is the data retained. Opportunity 3, tracking compliance certificates and renewal dates, is a diary in your own system.
Tenant data also intersects with tenancy law in ways the ICO guidance does not cover. Correspondence records may be relevant to a dispute. Keep that in mind when deciding how long a tool retains what it reads.
Links: Propertymark, Stay in control as AI-generated complaints rise (December 2025); ICO, Artificial intelligence hub.
What it won't do
It won't replace the property manager's judgement on whether a maintenance issue is actually urgent. AI can read "boiler not working" and flag it as urgent. It can't tell you whether the tenant's just turned the thermostat down, or whether the boiler's actually broken, or whether this is the third time this tenant has reported the same issue that turned out to be user error. That's still a property manager's call, and sometimes you need to speak to the tenant to work out what's really going on.
It won't handle tenant disputes or difficult conversations. If a tenant's complaining about their neighbor, or disputing a deposit deduction, or threatening to withhold rent, that needs a person who can listen, de-escalate, and find a solution. AI can't do that, and trying to automate those conversations will make things worse. This handles routine queries and information requests, not relationship management.
It won't fix properties that are poorly maintained or landlords who won't spend money. If your portfolio is full of properties with recurring maintenance issues because the landlord won't fix the underlying problem, automation will just log the same issue repeatedly. If landlords ignore your compliance reminders until the last minute, automation will document that you reminded them but it won't make them act faster. This makes good property management more efficient, it doesn't fix bad property management.
It won't fix a phone system that loses calls. If tenants cannot get through, or calls ring out in the evening with no route to an on-call person, an AI receptionist is not the first thing to buy. The cheaper fix is usually the phone system itself: sensible call routing, a proper out-of-hours rota, voicemail that actually reaches someone who can act on it. Get calls landing somewhere reliable first, then decide whether you need something answering them automatically. Paying monthly for an AI receptionist to paper over bad call routing is paying forever for a problem you could have fixed once.
Worth noting: much of what gets sold as "AI" is actually straightforward automation, which is often better news (cheaper, more reliable, easier to maintain). If you're not sure what counts as AI and what doesn't, we've written a plain-English explanation that cuts through the marketing.
If you're not yet sure AI is worth looking at for your business at all, this is worth reading first.
Scan your business now
This page describes what works across the sector. Your business is specific: different property types, different landlords, different bottlenecks.
Run a free scan and get three AI opportunities written for your business, based on your website and what you actually do. Takes about a minute, no sign-up required, and if the best answer isn't AI the report will say so.
Run a free scan for your business
Free, takes about a minute, no email needed to read your report.