nesaku
Talk to us Initial conversation
Writing

Real estate systems 29 August 2026 7 min read

How to Measure Real Estate Lead Handling: Source, Response, Contact and Viewing

Define source, response, contact, qualification and viewing metrics for a Dubai real-estate workflow without hiding delays behind averages.

By Alejandro at Nesaku

Measure real-estate lead handling as a sequence of timestamped events: receipt, capture, assignment, first human attempt, two-way contact, qualification, viewing, and outcome. One response-time figure cannot show whether an enquiry waited at the portal import, in a routing queue, or with an assigned agent.

This measurement model sits inside the wider Dubai real-estate lead-management workflow. It gives managers a way to locate a failed handoff without importing an internet benchmark or treating a fast acknowledgement as a completed sales conversation.

Key Takeaways

  • Keep the source timestamp and every later operating event.
  • Separate an automated acknowledgement, a human attempt, and two-way contact.
  • Publish the denominator and exclusions beside every rate.
  • Use median, 90th percentile, and target attainment rather than an average alone.
  • Connect response measures to qualification, viewing, and recorded outcomes without claiming causation.

Why does one speed-to-lead number fail?

A single clock combines several responsibilities. The source may deliver an enquiry late. An integration may wait before creating the CRM record. A routing rule may leave it unassigned. An agent may receive it promptly but act later. Each delay needs a different owner and correction.

Observed fact: Property Finder documents that its Enterprise API can pass leads into an external CRM (Property Finder, CRM integration with PF Expert 2.0, updated 2026-02-19). That capability supports direct capture, but its public page does not define a universal lead payload or prove when a particular agency received an event.

Recommendation: retain the source’s original event time as received_at. Add a separate captured_at when the operating record is created. If those values are overwritten by the CRM creation time, the dashboard will hide import delay.

An automatic WhatsApp acknowledgement belongs in the event history, but it should not close the human-response clock. It confirms that a system sent something. It does not prove that an agent understood the request, that the customer replied, or that the enquiry was qualified.

What event timeline should the agency record?

Use a compact event trail:

received_at → captured_at → assigned_at → first_human_attempt_at → first_two_way_contact_at → qualified_at → viewing_requested_at → viewing_booked_at → viewing_completed_at → outcome_at

Not every enquiry reaches every event. A wrong number may be disqualified before contact. A valid future buyer may enter nurture without a viewing. Missing later timestamps should remain visible as open or terminal states, not disappear from the report.

For each event, store the record ID, event type, timestamp, source system, actor, and relevant owner. Preserve corrections as new history rather than silently changing the original event. This makes one enquiry traceable when a monthly total looks wrong.

The earlier chapter on qualifying a Dubai property lead defines the evidence needed for qualified_at. Measurement should consume that decision, not invent a second qualification rule inside a dashboard.

How should source volume be counted?

Report raw enquiries, unique enquiries, and excluded enquiries separately.

  • Raw enquiries are all source events received before identity resolution or validity checks.
  • Unique enquiries are the operating enquiry records after a documented deduplication rule.
  • Excluded enquiries are tests, spam, invalid records, or duplicates removed under a published rule.

Keep original source immutable. If a person first enquires through Bayut and later continues on WhatsApp, the acquisition source remains Bayut while the latest conversation channel becomes WhatsApp. Replacing source with the latest channel makes the earlier acquisition report unreproducible.

Bayut describes Profolio reporting across Email, SMS, Phone, WhatsApp, and Chat, including listing context and export (Bayut, Best Features of Profolio, accessed 2026-08-29). This is evidence of source-specific reporting within that platform. It is not evidence that separate sources have already been reconciled into one person or one opportunity.

What counts as response and contact?

Define four different intervals:

MeasureCalculationOperational question
Capture delaycaptured_at - received_atDid ingestion delay the record?
Assignment delayassigned_at - captured_atDid routing delay ownership?
First human responsefirst_human_attempt_at - received_atWhen did an agent first make a relevant attempt?
Time to contactfirst_two_way_contact_at - received_atWhen did a real exchange begin?

Recommendation: count a human attempt only when an agent calls or sends a relevant message intended to address the enquiry. Count contact only when the prospect and agency exchange information in both directions. A sent template, ringing call, delivery receipt, or read receipt is not contact.

Meta documents sent, delivered, read, played, and failed status events for WhatsApp messages (Meta, WhatsApp message-status webhooks, accessed 2026-08-29). Those events describe message handling. They do not establish qualification, a viewing, or a commercial outcome.

If the agency operates defined coverage hours, report both elapsed time and staffed-hours time. State the timezone and coverage calendar. Do not silently stop the clock overnight, and do not compare an elapsed-hours figure with a staffed-hours target.

Which rates need explicit denominators?

Every rate should name its denominator:

  • Contact rate: unique eligible enquiries reaching two-way contact divided by unique eligible enquiries received in the cohort.
  • Qualification rate from contact: qualified enquiries divided by contacted enquiries.
  • Qualification rate from source: qualified enquiries divided by unique eligible enquiries.
  • Untouched rate: eligible enquiries with no human attempt by the reporting cut-off divided by eligible enquiries.
  • Viewing booking rate: enquiries with at least one booked viewing divided by qualified enquiries.
  • Viewing completion rate: completed viewings divided by viewings due in the period.
  • Post-viewing follow-up coverage: completed viewings with recorded feedback and a next action within the internal target divided by completed viewings.

Report both qualification rates. One tests conversation quality after contact. The other shows the full source-to-qualification path. Calling both simply “conversion rate” prevents a manager from knowing which problem changed.

For viewings, separate requested, booked, completed, prospect-cancelled, agency-cancelled, rescheduled, and no-show outcomes. A calendar event whose end time has passed is not evidence that a viewing happened.

Why use median and 90th percentile?

An average can improve while a smaller group waits much longer. Report the median for the typical handled enquiry and the 90th percentile to expose the slower tail. Add the percentage handled within the agency’s own target so the team can see whether the operating promise was met.

Set that target from the agency’s coverage, source mix, baseline, and staffing. Avoid repeating generic claims that every property enquiry must receive a human response within a particular number of minutes. The useful question is whether the published internal rule is realistic, monitored, and connected to a fallback when missed.

Segment the distribution by original source, project or listing, received hour, day, assigned team, and staffed-hours status. Compare response bands with contact, qualification, viewing, and outcome. Describe any relationship as an association unless the agency has evidence for a causal conclusion.

The next chapter on Dubai real-estate lead routing rules explains how assignment and fallback should work. The later CRM stages guide defines the state changes that the measurement trail observes.

What should a weekly lead-handling scorecard show?

Use one page with source counts, timing distributions, stage outcomes, and exceptions. Include raw, unique, excluded, unassigned, untouched, contacted, qualified, viewing-booked, viewing-completed, nurture, lost, and won counts. Add median and 90th-percentile capture, assignment, response, and contact times.

Beside the totals, list the oldest records without an owner, next action, or expected event. A manager needs both the pattern and the cases that require action.

Illustration: suppose a fictional weekly report shows prompt capture but a long 90th-percentile assignment delay for enquiries received outside staffed hours. That observation points to coverage and fallback design. It does not prove that routing caused a lost sale. Trace several affected records from source to outcome before changing the rule.

The representative lead-routing and viewing-booking case study shows how this event trail can fit around portals, CRM records, calendars, and WhatsApp. It is an illustrative workflow, not a measured client result.

Start with the timestamp you cannot currently explain

Choose several recent enquiries from different sources and reconstruct their event timelines. Mark missing timestamps, overwritten sources, ambiguous contact events, exclusions without reasons, and stage changes without evidence. That review will show whether the first repair belongs in capture, routing, agent handling, viewing records, or reporting definitions.

If the agency wants that diagnosis turned into a controlled operating path, see Nesaku’s automation and lead-handling service for Dubai real-estate teams.

Bring one week of enquiry records and the report you do not trust. We can map the clocks, denominators, and ownership gaps before discussing automation.

Sources