AI use case · Property

WhatsApp AI for tenant enquiries and maintenance requests.

For property managers and facilities teams whose staff answer the same tenant questions on WhatsApp all day. An assistant answers routine questions from information you approve, captures maintenance requests as tickets and passes anything sensitive to a person.

Ahmed Khaled, FounderLast updated:

Discuss this use case

A short note is enough. Phone and email let us reach you.

About: AI & Intelligent Automation

Tell us about the workflow, who uses it and what is difficult today. Please leave out passwords, confidential records and sensitive personal information.

Add company and other details (optional)

After you send this, choose a time for a free 30-minute discovery call. If no time suits, we get back to you as soon as possible to arrange one.

What is a WhatsApp AI assistant for tenants?

It is an assistant on your WhatsApp Business number that answers routine tenant questions, such as payment dates, building rules and move-in steps, only from information you approve. It records maintenance requests with the unit, issue and photos as tickets, and hands disputes, payments and complaints to your staff with the conversation attached.

The problem

Tenant messages pile up in personal chats.

In Dubai, tenants expect to reach their property manager on WhatsApp. They ask when the next rent cheque will be deposited, how to get a parking card, whether visitors need a pass and why the AC has stopped. Many of these messages land on staff phones, mixed with personal chats and other buildings.

Most of the questions repeat, so the same answers are typed again and again, in Arabic and English. Maintenance requests are the bigger risk. A photo of a leak sent at night can sit unread, and nobody is sure later whether it was logged, who took it or what the tenant was told.

  • Repeated answers

    Payment dates, building rules, access cards and move-in steps are explained by hand to tenant after tenant.

  • Requests lost in chat

    A broken AC or a water leak is reported in a chat thread instead of a ticket, so there is no record of who owns it.

  • No shared history

    When a staff member is away, the conversation sits on their phone and the next person starts from nothing.

How it works

One tenant message, from WhatsApp to the right person.

An illustrative workflow. The channels, approved topics and the property system it connects to are agreed for each project.

  1. Message

    A tenant writes to your WhatsApp Business number or your website chat, in Arabic, English or a mix of both.

  2. Identify

    The assistant matches the phone number to a tenant and unit in your property system, or asks for the unit if there is no match.

  3. Answer from approved content

    Routine questions are answered only from the information you approved, such as building rules, payment schedules and move-in guides. If nothing approved covers the question, it says so.

  4. Capture a request

    For maintenance, it asks for the unit, the issue, the location in the unit and photos, then creates a ticket and sends the tenant its reference.

  5. Hand over

    Disputes, payment questions, complaints and anything the assistant cannot answer go to a named team queue with the full conversation attached.

  6. Staff follow-up

    Your team replies from a shared inbox, and every message, handover and ticket update stays on the tenant's record.

Who does what

What the AI does · What a person checks.

What the AI does

  • Answers routine questions from approved information only, in Arabic or English

  • Recognises the tenant and unit from the phone number where your records allow

  • Collects unit, issue, location and photos for maintenance requests

  • Creates the ticket and gives the tenant a reference

  • Detects disputes, complaints and payment questions and passes them on with the conversation

What a person checks

  • Every handed-over conversation, including disputes and complaints

  • Anything about money: rent changes, refunds, deposits and payment plans

  • Renewal terms and any notice a tenant gives or receives

  • The approved answers themselves, before they go live and whenever a policy changes

  • Urgent safety issues, which follow your emergency procedure, not the chat

First pilot

A first pilot on one building and real questions.

We start with the questions your tenants already ask. Scope, timeline and price are confirmed in a written proposal.

  • One building or community

    Choose a property where tenant messages are frequent and the rules are well documented.

  • Your past questions

    A sample of real tenant messages, shared safely after the first call, shows which topics are worth answering and which must go to staff.

  • An approved answer set

    Your team writes or approves every answer the assistant may give. Nothing outside it is answered.

  • Agreed measures

    Before launch, agree how answers, handovers and ticket quality are checked, and review a sample of real conversations with your team.

Data, hosting and Arabic

Where tenant data goes, and how Arabic is handled.

  • Hosting

    The application and stored data can be hosted on Azure UAE North, on private servers managed by Vloud, or in your own cloud account. The AI model is used through a provider agreed per project, and its processing region may be outside the UAE.

  • WhatsApp itself

    Messages pass through WhatsApp's business platform before they reach the workflow. Your team should decide which details tenants are asked to share there.

  • Arabic

    Arabic, English and mixed messages, including dialect and Arabic typed in Latin letters, are tested as their own sample with your team.

  • Access

    Staff see the conversations for the buildings they manage. Tenant data kept, its retention and who can export it are agreed with you.

Honest limits

Where it does not work yet.

  • It answers only from approved information. It never decides on rent, refunds, deposits, renewals or disputes.
  • It is not an emergency line. Fire, lift entrapment and similar cases must follow your emergency procedure and phone numbers.
  • Voice notes and unclear photos may not be understood well enough and are passed to a person.
  • We do not quote resolution rates or time savings before the pilot. Results are measured on your own tenant conversations.
  • If your property system has no way to receive tickets, requests are logged in a separate queue until an integration is agreed.

What drives cost and timeline. The number of buildings and approved topics, the channels used, the property system and its integration route, how handovers reach your team, Arabic dialect testing and hosting requirements. We don't publish a price list; scope, timeline and price come in a written proposal after the discovery call.

Evidence

What we can show you.

Every example is labelled: delivered work, our own product, a demonstrator, or an illustration with synthetic data.

  • Our product

    ArkanPM maintenance and work orders

    ArkanPM is our own property management product. Its maintenance and work orders module follows a tenant request from the first report to a completed work order, which is where tickets from the assistant can land. It is a product we own, not a delivered WhatsApp assistant.

Before you start

Questions buyers ask.

Does it work in Arabic?

Yes, in Arabic, English and mixed messages. Tenants often write in dialect or in Arabic typed with Latin letters, so those are tested as their own sample before launch.

Does it connect to our property system?

It can, if your system has a supported API or an agreed import route. We can also send tickets to ArkanPM. We review your system before confirming what the assistant can read and write.

What happens outside working hours?

The assistant can still answer approved questions and log maintenance requests. Anything that needs a person is queued for your team, and the tenant is told when to expect a reply from staff. Emergencies follow your own emergency numbers.

Can tenants still reach a person?

Yes. A tenant can ask for a person at any point, and the conversation is passed to your team with its history. The assistant also hands over by itself when a topic is outside its approved list.

Will it make promises about rent or renewals?

No. Rent, deposits, refunds, renewals and disputes always go to staff. The assistant can say which team will handle the question, but it never agrees terms.

How is tenant data protected?

Access is role-based by building, and stored data and retention are agreed with you. The application and stored data can be hosted on Azure UAE North, on private servers managed by Vloud, or in your own cloud account. The AI model is used through a provider agreed per project, and its processing region may be outside the UAE.

Who owns the code?

You own the source code and IP of the custom work we build for you. Our own products, including ArkanPM, stay ours, and any licence terms are agreed separately.

Next step

Test it on your own documents.

Tell us what the work looks like today. We reply with the questions we need answered before a pilot can be scoped.

Discuss this use case

A short note is enough. Phone and email let us reach you.

About: AI & Intelligent Automation

Tell us about the workflow, who uses it and what is difficult today. Please leave out passwords, confidential records and sensitive personal information.

Add company and other details (optional)

After you send this, choose a time for a free 30-minute discovery call. If no time suits, we get back to you as soon as possible to arrange one.