A property viewing is deliverable only when the customer, agent, property and access arrangement agree. For a Zanus AI booking evaluation, test these constraints together. A free calendar slot alone does not show that someone can enter the property or that the person conducting the viewing is available.

Write the viewing constraints explicitly

Record the listing reference, authorised viewing windows, assigned agent and any access approval needed. Keep the public listing description separate from restricted access instructions. Ask who may view or disclose each part of the record in the proposed workflow.

If staff must confirm access with an occupier or keyholder, label the customer’s request as pending. The assistant should not call it a confirmed viewing while that step remains open.

Use a booking record staff can trust

Record itemWhat to verify
Property referenceThe correct listing, including unit where needed
Appointment stateRequested, awaiting access approval or confirmed
Agent assignmentA staffed owner who can attend
Customer contactA verified way to communicate changes

Keep any sensitive access detail within the agency’s approved handling process. A general confirmation message need not reveal internal instructions.

Test a late change at the property

Create a test viewing and then block access to the property. Ask to move the appointment and inspect the original event, replacement request and customer confirmation. Check whether the assigned agent sees the same final state.

Repeat with a customer enquiring about two similarly named listings. Have the caller correct the listing reference during the conversation. The final booking should reflect the last agreed property, not the first one mentioned.

Review the day as an agent would run it

Give the agent the day’s saved viewings and ask whether they can fulfil each appointment. Check travel gaps, ownership, access approval, customer contact and the status of changes. Record missing information and calendar repairs.

Track attended viewings separately from booking requests and confirmations. Before expanding, confirm that cancellations release the right time and that repeated messages do not create duplicate events. A useful booking workflow creates an accurate working diary for the agency.

Questions and answers

Can a viewing be confirmed before access is approved?

Only if the agency’s authorised process already permits it. Otherwise keep the request pending and explain the staff confirmation step clearly.

Mapping the Viewing Appointment as a Chain of Distinct Decisions

A viewing appointment in residential or commercial property involves at least three separate decisions that should not be collapsed into one booking action: whether the customer is eligible or ready to view, whether the property can be accessed at the proposed time, and whether a specific agent is both available and authorised to conduct that viewing. An AI-assisted call workflow can gather information quickly, but the agency still needs to design which of those decisions the system may resolve automatically and which must pause for a staff member. Mapping that chain before configuring any booking tool is the practical first step, and it is one that changes meaningfully depending on property type, tenure and occupier status.

Consider a fictional illustrative scenario involving a lettings enquiry for a flat where the current tenant is still in residence. A caller requests a viewing at a time that appears free on the agent's calendar. On the surface, the slot is available. However, the occupier has requested a minimum notice period and has indicated certain days are unsuitable. The property access record and the agent's calendar are two different documents, and a booking tool that reads only the calendar will see an open slot where the agency's operational record would show a constraint. This scenario illustrates why the evaluation of any AI calling workflow should test the access record separately from the calendar, and why those two sources need to be reconcilable at the point of confirmation.

Defining What a Confirmed Viewing Actually Means for Your Agency

Different agencies draw the confirmation line in different places, and that choice has real operational consequences. Some agencies consider a viewing confirmed once the customer has agreed a time and the agent is assigned. Others require a separate access confirmation from a keyholder or property manager before they will use the word confirmed in any message to the customer. A third group run a staged process in which a provisional acknowledgement goes to the customer immediately, followed by a separate confirmation once internal checks are complete.

Before evaluating any calling tool, the agency should write down its own definition in plain language and then check whether the workflow being tested can reflect that definition without ambiguity. A useful test is to ask a staff member who was not involved in configuration to read a sample confirmation message and state what they believe the appointment status to be. If the message could be read as a firm booking when it is actually a pending request, the workflow needs adjustment regardless of the technology behind it. Ambiguous language in confirmation messages generates customer complaints and, more practically, causes agents to arrive at properties that are not ready for viewing.

A Worked Fictional Enquiry: From Inbound Call to Agent Diary

The following fictional walkthrough is intended as an evaluation template rather than a description of any specific product's behaviour. Agencies can adapt it to their own testing process.

A prospective buyer calls to request a viewing of a semi-detached house. The property is occupied by the vendor, who has agreed to allow viewings on weekday mornings only. The assigned agent is shared across three listings that week.

The call captures the caller's name and a contact number, the property reference they mention, their preferred day and approximate time, and whether they have any access requirements that the agent should know about. Before a booking record is created, the workflow should verify that the property reference matches the active listing, not a withdrawn or similar-sounding alternative, and that the proposed time falls within the vendor's agreed viewing window.

At this stage the record state should read something like awaiting access check rather than confirmed. The agent assigned to that listing receives a notification showing the request, the time, and the outstanding access step. A staff member contacts the vendor, receives approval, and updates the record. Only at that point does the appointment move to confirmed state, and only then should a confirmation message go to the customer with the agent's name and the agreed time.

When evaluating a workflow against this scenario, observable evidence to look for includes: whether the interim record is visible to the right staff member, whether the customer receives a message that honestly describes the pending state, and whether the final confirmed record matches the time that was actually approved rather than the time originally requested if those differed.

Handling the Caller Who Names the Wrong Property

Property enquiries regularly involve naming confusion. Streets with similar names, development block letters that callers misread, and multiple units within the same address all create the possibility that a caller describes one property while intending another. This is not a rare edge case; it is a practical daily risk in any busy lettings or sales office.

A useful evaluation exercise is to script a call in which the caller initially states an incorrect or ambiguous property reference and then corrects it partway through the conversation. The test question is simple: which property appears in the final booking record? The answer should always be the property agreed at the end of the call, not the one first mentioned. Evaluators should check the raw record, not just the confirmation message, because a message might summarise correctly while the underlying record retains the first property reference and creates a conflict when the agent checks their diary.

A secondary check is whether the workflow can handle a caller who is genuinely uncertain between two properties and asks for information about both before deciding. In that case, no booking record should be created until the caller makes a choice, and any information disclosed about access or timing should reflect only what the agency has authorised for disclosure to prospective viewers of each listing separately.

Access Instructions as a Separate Category of Information

Access instructions for a property, such as keysafe codes, alarm information, or specific entry procedures, occupy a different category from the general listing description that appears on a portal. The general description is public. Access instructions are operationally sensitive and should be disclosed only to the confirmed attendee, at the right time, through an authorised channel.

When designing a booking workflow, the agency should explicitly decide which information the system may share during a call, which should appear only in a message to a confirmed and verified attendee, and which should never be communicated through automated channels at all. This is a governance decision for the agency, not a default that any tool resolves on the agency's behalf.

A practical way to test this is to configure a sample property record with clearly labelled public fields and clearly labelled restricted fields, then run a test call and review exactly what information was disclosed and at what point. The evaluator should note whether the restricted fields remained inaccessible and whether the caller could not extract them by rephrasing the question. If the workflow does not make this distinction robustly, that is a design gap that needs addressing before the agency uses it with live listings.

Agent Assignment and Ownership of the Appointment

Assigning a viewing to an agent is not the same as the agent accepting and owning that appointment. In a small agency, the distinction may seem trivial, but in offices where multiple staff share listings or where a viewing agent differs from the listing agent, ownership of the appointment determines who bears responsibility for attending, who communicates with the customer about changes, and who updates the record if something changes at the property.

A booking record is more useful when it identifies a named individual as the responsible owner of each confirmed appointment, rather than recording only the agency name or a shared resource. When evaluating a workflow, ask whether the agent sees the appointment in their own view of the diary, whether they can identify which property it concerns without opening a separate document, and whether they receive any alert when the appointment is changed or cancelled. An agent who learns about a cancellation only when they arrive at a property has been let down by a process failure, not a technology failure.

Consider also the case where the assigned agent becomes unavailable after confirmation. The record should allow a manager to reassign the appointment to another agent and ensure that the customer receives an updated confirmation reflecting the change, without creating a duplicate event or leaving the original agent's diary showing an obligation they no longer hold.

Reviewing the Full Day Before It Runs

A practical acceptance exercise for any viewing booking workflow is to generate a realistic fictional day of appointments and then review it as an agent would before leaving the office in the morning. The review should answer several questions without the agent needing to open multiple systems or make additional calls.

  • Is each appointment associated with a specific property address and unit reference where relevant?
  • Does each appointment show a confirmed access state rather than a pending one?
  • Are travel gaps between appointments realistic given the property locations?
  • Is there a contact method for each customer in case the agent needs to communicate a delay?
  • Is there any appointment that shows a changed time in the record but an older time in the confirmation sent to the customer?

If the agent cannot answer all of those questions from the diary alone, the workflow is producing records that require supplementary checking. That supplementary checking is itself a risk, because under time pressure it may not happen consistently. The evaluation should note each gap and decide whether it is something the workflow configuration should address or whether it reflects an inherent limitation that the agency needs to accommodate through a separate process.

Managing Cancellations and Preventing Duplicate Records

Cancellations carry two practical risks in a booking workflow: that the released time slot is not correctly freed for other use, and that a customer who cancels and then re-books creates duplicate records that confuse the agent's diary. Both risks are worth testing explicitly rather than assuming the workflow handles them correctly.

A useful test sequence for cancellations runs as follows. Create a confirmed appointment. Cancel it through the workflow. Confirm that the time slot is now shown as available. Check whether the original record is preserved in a cancelled state for reference, rather than simply deleted. Then have the same customer call to rebook a different time for the same property. Confirm that the result is a single new appointment record, not a duplicated entry alongside the cancelled one.

For agencies that track viewing history for compliance or vendor reporting purposes, the cancelled record should remain accessible to staff rather than being removed from the system entirely. That is a configuration and data handling question the agency should raise with the vendor directly, rather than assuming a default behaviour.

Separating Individual Customer Identity from the Booking Request

A booking request describes what is wanted: a property, a time, an agent. Customer identity describes who is making the request. These should be held and handled as separate pieces of information, particularly where the agency may share booking confirmations across multiple channels or where a single household might have multiple individuals making enquiries.

Consider a scenario where one member of a couple calls to book a viewing and a second member calls later to check or change the same appointment. The workflow needs to establish whether the second caller is associated with the same booking before disclosing its details or allowing changes. This is a customer service matter as much as a data handling matter: disclosing appointment details to an unconnected third party who happens to know the property address would be a process failure with real consequences.

The agency should define its verification standard for callers seeking to access or change an existing booking. That standard does not need to be elaborate, but it should be consistent and documented so that staff applying it manually and any automated step in the workflow apply the same threshold.

Comparison Criteria When Assessing a Viewing Workflow Tool

When comparing tools or configurations for viewing appointment calls, the following criteria are worth weighting explicitly because of how they affect daily operations in a property context specifically.

Criterion Why it matters in property viewings
Access state as a distinct booking field Property access and calendar availability are separate conditions; conflating them causes agents to attend properties that cannot be entered
Ambiguous property reference handling Booking the wrong unit or similar address causes wasted journeys and customer frustration
Pending versus confirmed status transparency Customers and agents need to know whether a booking is firm before planning around it
Named agent ownership in the record Shared or unassigned appointments are missed; named ownership creates accountability
Cancellation record retention Viewing history supports vendor reporting and helps identify patterns in no-shows
Restricted field handling Access instructions must not be disclosed through the same channel as general listing information
Caller verification for booking changes Appointments should not be altered or disclosed based on knowledge of the property address alone

Handover Arrangements for Out-of-Hours Calls

Viewing enquiry calls do not arrive only during office hours. An agency using an AI calling workflow needs to decide what happens to a request received outside staffed hours: whether the system accepts it into a pending queue, whether it communicates a realistic expectation to the caller about when a human will review and confirm it, and whether the on-call arrangement, if any exists, is reflected in the workflow.

The handover arrangement matters most for late changes. If a caller cancels a viewing at an early hour on the morning of the appointment, the assigned agent needs to know before travelling. A workflow that queues that message for morning office review rather than routing it as an urgent alert will result in a wasted journey. The agency should test what happens in precisely that scenario and decide whether the outcome is acceptable or whether the handover rule needs adjustment.

A question worth asking any vendor during evaluation is how the workflow distinguishes between a routine out-of-hours booking request and an urgent same-day change, and what control the agency has over how each is handled. The answer will affect whether the agency can rely on the system for active listings with same-day viewing slots.

The most durable viewing booking workflow is one that accurately represents the state of every appointment to every person who needs to act on it, whether that is a customer waiting to hear if their request is approved, an agent checking their morning diary, a manager reassigning a viewing, or a colleague covering an absent colleague's day. Building that accuracy into the record design from the start, before testing begins, makes the evaluation itself more revealing and any subsequent configuration decisions more grounded in how the agency actually runs its viewings.

Sources and next steps

This independent Sysevo guide proposes evaluation steps, not hands-on Zanus AI results. Check the official Zanus AI Front Office overview and ask the vendor to demonstrate your exact workflow and confirm its current availability.

For related context, read Zanus AI appointment-setting and calendar tests. To compare providers using the same requirements, explore Sysevo scheduling or book a Sysevo demo. Sysevo publishes this guide and sells its own business software.

Related workflow guides

For current Sysevo subscriptions and usage options, compare plans and pricing.