A property enquiry can arrive through a portal, email and a phone call within the same afternoon. In a Zanus AI integration evaluation, check whether those contacts become one coherent request or three competing tasks. Deduplication should remove repeated work without erasing genuine interest in different properties.
Define what counts as the same enquiry
Use the customer, property reference and purpose together. Two messages about one viewing may belong to the same request, while questions about two listings should remain distinct. Keep the original source identifiers so staff can trace where each message came from.
Ask the vendor which identifiers the proposed integrations actually expose. This guide does not assume a particular property portal connector exists. Demonstrate the exact route your agency intends to use before relying on it.
Prepare a small matching worksheet
| Incoming contact | Expected handling |
|---|---|
| Same person, same listing, same viewing request | Attach to the open enquiry or flag for review |
| Same person, different listing | Preserve the separate property interest |
| Shared email, different person | Avoid an unverified identity merge |
| Old enquiry followed by a new purpose | Keep the new request visible |
A useful match includes a reason and a way to reverse an incorrect merge.
Send the events out of order
Submit a test portal enquiry, then call to change the preferred viewing time. Deliver the older portal event again after the call. Check whether it overwrites the newer request or creates another callback.
Next let two agents claim what appears to be the same lead. Inspect ownership and notifications. Staff should not discover duplicate ownership only when the customer receives two calls. Keep a queue for uncertain matches instead of forcing every event into an automatic merge.
Measure saved work and lost context
Track duplicate tasks avoided, incorrect merges, separate property interests preserved and staff minutes spent repairing identity or ownership. Reconcile a sample from the source messages through to the final customer callback.
Keep the pilot bounded to a connector and enquiry type that staff can inspect easily. Expand only after newer updates survive late events and agents can explain the final request history. An apparently tidy CRM is not an improvement if it hides what the customer asked for.
Questions and answers
Should every contact from the same email address be merged?
No. Shared addresses, different properties and new request purposes can require separate records or staff review. Define the matching rule before enabling automatic merging.
Mapping the full enquiry journey before configuring any merge rule
Before an agency decides which contacts should collapse into one record and which should remain separate, it helps to trace the complete path a property enquiry travels from first contact through to a confirmed viewing or offer. This mapping exercise reveals the points where duplication is most likely to create confusion, and it surfaces decisions that belong to a business policy rather than a software configuration.
Consider a fictional scenario to illustrate the kind of journey worth mapping. A prospective buyer named Jordan Chen sees a listing on a portal and submits a viewing request on a Tuesday morning. That afternoon, Jordan calls the branch directly to ask whether Saturday afternoon is available, giving a mobile number that differs from the one entered in the portal form. Later that evening, the portal automatically re-sends an acknowledgement email because Jordan did not click the confirmation link. By Wednesday morning the agency's inbox may show three contact events that are, in substance, one person with one request about one property. Alternatively, they may show two people sharing a business email address, or one person genuinely pursuing two separate listings through channels they do not realise are connected.
The point of the mapping exercise is not to declare which scenario is more plausible. It is to force the agency to write down, before touching any configuration, exactly which identifiers it will treat as sufficient evidence of the same enquiry. That written rule becomes the acceptance criterion against which any deduplication behaviour is later evaluated.
Choosing and weighting the identifiers your agency will rely on
Different identifiers carry different reliability, and the weight a business assigns to each one should reflect the realities of its own customer base rather than a generic recommendation. Below are the main identifiers typically available in a property enquiry context, with the reasoning a business might apply when deciding how much to trust each one.
Email address
Email is often treated as a primary identifier because portals capture it systematically and it persists across sessions. The complication is that households, couples and flatmate groups frequently share one address, and many buyers use a different address for portal registrations than for direct agency correspondence. An agency serving a high proportion of joint purchasers should treat email alone as suggestive rather than conclusive and require at least one corroborating identifier before an automatic merge.
Phone number
Phone numbers are reliable for confirming identity during a call but are inconsistently captured elsewhere. A portal form may record a mobile number that the same caller never volunteers again when phoning the branch. An agency that wants to use phone as a matching signal should decide in advance whether it will treat mobile and landline numbers for the same contact as equivalent, and how it will handle a situation where a number has evidently been reassigned to a different person since a previous enquiry.
Property reference
The property reference is essential context but is not an identity signal on its own. Its purpose in a matching rule is to prevent contacts from different people about the same listing from collapsing into one record. A useful matching rule always includes the property reference as a scoping condition: even if two contacts share an email address and a phone number, if they concern different properties they represent separate interests that should each remain visible and actionable.
Request type and purpose
A viewing request and a maintenance query from the same landlord are not the same enquiry even if they share every contact detail. Agencies managing both sales and lettings, or both landlord and tenant relationships, should include the purpose of the contact as a matching dimension. Omitting it creates a risk that a new commercial request is silently appended to an old resolved thread where it may not receive appropriate attention or routing.
Time window
Some matching logic uses a time window to decide whether a second contact is likely to be a follow-up to a recent one or an independent new enquiry. A business evaluating this approach should define the window deliberately and review whether it suits the patterns of its own enquiry volumes. An agency with a slow rural market may find that a buyer circles back after several weeks; a city agency may see follow-up calls within hours. The window is a business decision, not a technical default to accept without review.
Designing a readable matching rule card for front desk staff
A deduplication configuration that only the system understands offers limited operational value. Staff who cannot explain the matching logic cannot spot errors, cannot handle exceptions gracefully and cannot tell a customer why two enquiries were treated as one or kept separate. A practical tool is a short matching rule card that can sit at a reception desk or be printed inside a staff handbook.
The card should answer three questions for every common contact scenario: should this contact be attached to an existing record, should it create a new record, or should it be held for a human to decide? The third category is the one most agencies underinvest in. An unresolved queue that staff review daily is operationally slower than full automation, but it is far less damaging than a merge error that causes two prospective buyers to receive the same callback about a property only one of them enquired about.
A sample card structure might look like this:
| Scenario | Recommended handling | Who decides |
|---|---|---|
| Matching email, matching phone, same property reference, arrived within the same day | Attach to existing open enquiry; do not create new record | System, with staff review of merge log |
| Matching email, different phone, same property reference | Flag for staff to confirm identity before attaching | Front desk |
| Matching email, same phone, different property reference | Preserve as separate property interest under same contact record | System |
| Matching email, no phone, enquiry purpose changed from viewing to offer | Keep new purpose visible; do not close old thread as resolved | Responsible agent |
| Shared business email, different names provided on each contact | Do not merge; hold for manual identity verification | Senior staff or manager |
| Portal re-send of an event already received and processed | Detect as duplicate event; ignore or log without creating task | System, with idempotency check |
The card is most useful when it is reviewed by the staff who will actually act on exceptions, not only by whoever configured the system. Frontline staff often identify edge cases that a desk-based configuration review misses entirely.
Running an exception-handling rehearsal before going live
A useful pre-launch exercise goes beyond the standard out-of-order event test described in the evaluation guidance. It focuses specifically on the scenarios where a purely automatic merge would cause a visible customer experience failure, and it gives staff practice in recognising and resolving those failures before they encounter them in a live situation.
In one rehearsal scenario, a coordinator plays the role of a buyer named Sam who submits a portal enquiry for a two-bedroom flat and later calls asking about a three-bedroom house on the same street. The exercise tests whether the two property interests remain distinct and whether the agent who picks up the second call can see the first enquiry without concluding it is already handled. The coordinator then submits a second portal enquiry using Sam's email but a slightly different first name, simulating a typo or a formal-versus-informal name difference. The question for the exercise is whether the system presents this as a possible duplicate, a definite duplicate or a new record, and whether staff can locate both contacts when searching by phone number alone.
In a second rehearsal scenario, two agents are both assigned a contact that arrived through a portal at the same time as a phone call about the same listing. The exercise checks how quickly each agent can see that the other has ownership of part of the enquiry, what the handover process looks like and whether the customer is at risk of receiving two calls from different people who each believe they are the sole point of contact.
Document the outcome of each rehearsal scenario, including any steps that required improvisation, and use those notes to update the matching rule card before the system handles real enquiries.
Ownership clarity as a distinct problem from identity matching
Deduplication of contact events addresses whether records should be merged. Ownership assignment addresses who is responsible for acting on the merged record. These are related but different problems, and an agency should evaluate them separately rather than assuming that resolving the first one automatically resolves the second.
In a scenario where a portal enquiry is automatically routed to one agent and a phone call about the same property is manually logged by a second agent, a correct identity merge may still leave two people uncertain about who is expected to make the callback. The customer has one expectation: that someone who knows the full context of their enquiry will contact them. The agency needs a clear rule about how ownership transfers or consolidates when a merge occurs after initial assignment.
Possible approaches include assigning ownership to whoever first logged a contact, to whoever most recently interacted with the customer, to a dedicated enquiry coordinator who then assigns to a viewing agent, or to a pool queue that any available agent can claim. Each approach has different implications for accountability, for the customer experience when their named contact is unavailable and for the way commission or credit is attributed internally. The evaluation of a deduplication configuration should include a question about how ownership is displayed and transferred, not only about how records are matched.
Maintaining an audit trail that staff can actually read
A merge that cannot be explained to a customer or a manager after the fact is a merge that creates liability rather than reducing effort. An agency should define, before configuring automatic merging, what the merge record will show and whether that record is accessible to anyone other than a system administrator.
A readable audit trail for a property enquiry merge might include the original source of each contact event, the identifiers that triggered the match, the timestamp of the merge decision, the name of the rule or process that applied and whether a human reviewed or overrode the automatic decision. It should also include a straightforward way to reverse the merge if a staff member subsequently determines that the contacts belonged to different people or different purposes.
In a practical acceptance exercise, ask a member of staff who was not involved in the configuration to locate a specific merged record, explain why the contacts were combined and demonstrate how they would separate them if the merge turned out to be incorrect. If that exercise takes more than a few minutes or requires administrator access, the audit trail is probably not operationally useful for day-to-day error correction.
Handling enquiries that span the sales and lettings boundary
A buyer who places an offer on a property may also enquire about a short-term let as a bridging arrangement. A landlord who is selling a property may have an active tenancy query on a separate listing. In agencies that handle both transactions, the risk is that deduplication logic designed for one service line creates unexpected behaviour when the same contact appears in another.
Before configuring shared identity matching across service lines, consider whether a contact record that combines a sales enquiry and a lettings query will route correctly to the right team. A match that is technically correct may still cause operational confusion if it surfaces to a sales negotiator who has no visibility of the lettings context, or vice versa. The evaluation question is not only whether the identity match is accurate but whether the combined record is useful and visible to the right people.
Agencies that decide to keep sales and lettings records separate should document that decision explicitly and configure the matching rules to scope each match within a service line. This prevents a correct identity match from inadvertently creating a cross-team ownership dispute or hiding a relevant piece of the customer's history from the person making the callback.
Reviewing deduplication quality on a regular basis
A matching rule that works well when it is first configured may behave differently as contact volumes increase, as portal connectors change, or as staff habits around data entry shift. Building a lightweight regular review into the agency's operations calendar, rather than treating the initial configuration as a permanent solution, allows errors to be caught before they accumulate into a significant backlog of incorrect or missing records.
A practical monthly review might involve a staff member sampling a small number of enquiries that were automatically merged during the period, checking that each merge was correct and that the original source records are still traceable. It might also involve checking the unresolved queue to confirm that items are being cleared promptly and that none have been sitting unactioned for longer than the agency's defined service standard. The review does not need to be exhaustive to be useful; its value comes from regularity and from the habit of treating deduplication as an ongoing operational process rather than a one-time configuration task.
Agencies that log the outcomes of these reviews build a body of evidence that is genuinely useful when evaluating whether to expand the configuration, adjust matching rules or raise a question with the vendor about specific behaviour they have observed in their own data.
Clear matching rules written before any configuration begins, a staff-readable audit trail, a deliberate approach to ownership handover and a scheduled review of merge quality give an agency the operational foundation to use contact deduplication in a way that genuinely reduces effort rather than merely redistributing confusion from the inbox to the record history.
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 CRM integration and lead handoff. To compare providers using the same requirements, explore Sysevo integrations or book a Sysevo demo. Sysevo publishes this guide and sells its own business software.
Related workflow guides
- Zanus AI CRM and Voice AI: Match Calls to the Right Record
- Zanus AI CRM for Estate Agents: Separate Buyer and Seller Leads
For current Sysevo subscriptions and usage options, compare plans and pricing.


