A tenant calling at night gets an answer, not voicemail
A Hebrew voice agent that logs maintenance calls with building, unit and an exact description, tells an emergency from something that can wait, escalates to the on-call and records every request.
The short answer
How does a property manager cover tenant calls around the clock?
A property manager covers tenant calls around the clock when an AI agent answers every one in Hebrew and runs it to a conclusion: identifying the building and unit, asking exactly what happened, deciding whether it is an emergency by the rules you set, and escalating to the on-call or opening a work order with a service window. Faults happen at night and at weekends, precisely when no manager is near the phone — which is why an unanswered call turns into a bigger repair. Every request is stored with a recording, transcript, summary and escalation reason, giving you a clean record for tenants and owners alike.
What happens today
- A leak at three in the morning goes to voicemail, and by daylight it has spread to the flat below
- A tenant describes a fault without saying which building or unit, and the crew is sent to the wrong address
- The same incident is reported separately by five tenants and five work orders are opened for one fault
- One manager covers dozens of buildings and takes every call on a personal mobile
- Rent, lease and building-committee questions get mixed in with urgent faults on the same line
How it sounds
Hello, you have reached the management company. I am the digital assistant. I can log a fault, check an existing request, or leave a message for the office. Which building and which unit is this about?
What the agent captures
- Tenant name and a callback number
- Building address and the unit number
- Type of fault — water, electricity, lift, heating, damage to common areas
- Urgency: an active leak, no power, people trapped in a lift
- When the flat can be accessed and who will be home
- Whether the caller is a tenant, an owner or the building committee
Common use cases
Faults do not happen during office hours
The most expensive calls arrive exactly when nobody is answering: a leak at night, a lift stuck on a Saturday, no heating over a cold weekend. The agent answers at any hour, handles several tenants at once when a whole building reports the same fault, and runs the call through to a logged request or a transfer to the on-call.
- 24/7 answering on the management company's existing line
- Simultaneous calls when one incident triggers a wave of reports
- The same agent on WhatsApp, convenient for tenants sending details
- Transfer to the on-call or an extension by fault type and time of day
A work order with everything needed to dispatch a contractor
A partial work order is a wasted drive. The agent makes sure the building, unit, fault type, exact description, access window and who will be home are all captured — and only then opens the request. If the fault is in a common area rather than inside a flat, that is flagged, because it changes who pays and who approves.
- A fixed script per fault type
- Upload a list of buildings, units or vendors as CSV or Excel for filtering on the call
- Work orders opened in your system through a connected tool
- Requests pushed out over the REST API or signed webhooks
Emergencies reach the on-call, the rest goes in the diary
You define what counts as an emergency — an active leak, people trapped in a lift, a building with no power, a gas smell — and what happens then: an immediate transfer to the on-call extension or the emergency contractor. Everything else is booked into a normal service window, and the escalation reason is stored on the conversation so you can check later whether the rules are right.
A full record for tenants and owners
When there is a dispute about what was reported and when, the record is what settles it. Every call is stored with a recording, transcript, summary, sentiment and escalation reason. Tenants' identifying data is masked in the transcript per the policy you set, recording playback is limited by role, and erasure requests run through a full anonymization flow with an audit trail.
- Recording, transcript and summary on every request
- Identifying data masked in transcripts per company policy
- Recording playback gated by role permission
- Anonymization for erasure, with an audit log and reference number
Frequently asked
How does the agent decide what is an emergency and what can wait?
By rules you set in the call flow, not by free interpretation. You decide which faults require an immediate transfer to the on-call and which are booked into a normal service window, and the agent follows that and records the escalation reason on the conversation — so you can review and tune it later.
What happens when the whole building reports the same fault?
The agent answers all of them in parallel — there is no hold queue — and takes each tenant's unit number and what they can see, giving you the full extent of the fault. All the calls are summarized and appear together, so it is immediately clear this is one incident and not ten separate work orders.
Does the agent also answer rent and lease questions?
It answers what you approved in the knowledge base — procedures, payment dates, how to submit a request, who handles what — and hands any specific question about a lease, a debt or a settlement to the office. The agent does not interpret contracts and makes no commitment on an owner's behalf.
Can we produce a report of requests for owners?
Yes. Every call is stored as a record with a summary, sentiment and outcome, and the data can be exported or pushed out through the REST API and signed webhooks. That lets you show an owner exactly what was reported in their building, when, and what was done about it.
Call the agent now
A real call, in Hebrew. Exactly what your callers would hear.
Have the agent call me
Leave a number — the agent calls you in under a minute.
Updated: