Estate agency technology has improved considerably over the last decade. CRMs are more capable. Portal integrations are smoother. Automated follow-up sequences reduce the manual work of staying in contact. Reporting dashboards give principals a clearer picture of what their teams are doing.
What none of this technology does is tell an agency which sellers are actually close to instructing.
That is not a criticism of the vendors. It is a description of what the technology was designed to do. The tools that dominate the UK estate agency market were built to record activity, organise contacts, and automate communication. They do those things well. Intent identification is a different problem, and it was not the problem these tools were built to solve.
Understanding why requires looking at what each category of tool actually measures, and what it cannot see.
The activity layer
The core of most estate agency technology stacks is the CRM. In the UK market, platforms such as Reapit, Jupix, and Alto sit at the centre of how agencies manage their vendor relationships. These systems are built around records: contact records, property records, applicant records, valuation records.
What they track is activity. Calls logged. Emails sent. Valuations booked. Follow-up tasks created and completed. Viewings arranged. Offers recorded. The CRM is fundamentally a ledger of what has happened between the agency and its contacts.
Portal integrations extend this further. Rightmove and Zoopla feed enquiry data into the CRM, creating records for inbound contacts. The activity layer then begins: automated acknowledgements are sent, contacts are added to follow-up sequences, tasks are assigned to negotiators.
Automated communication tools sit on top of this. Sequences triggered by time elapsed, by enquiry source, by property type. The logic is activity-based throughout: if X days have passed since last contact, send Y. If the enquiry came from portal Z, apply sequence W.
Reporting tools close the loop. Calls made per negotiator. Response times. Conversion rates from enquiry to valuation. Instructions won per month. All of these metrics measure what the agency has done, and how often.
This is the activity layer. It is comprehensive, well-built, and widely adopted. It tells an agency a great deal about its own behaviour. It tells it almost nothing about seller intent.
Why activity is not intent
Activity and intent are related but they are not the same thing, and the distinction matters commercially.
Activity is what has happened. A seller submitted an enquiry. An agent sent a response. A valuation was booked. These are facts. They are recorded accurately by the tools designed to record them.
Intent is what is about to happen. Whether the seller who submitted that enquiry is ready to instruct, still deciding, or gathering information with no near-term plan to move. This is not a fact that gets recorded anywhere, because no existing tool in the standard estate agency stack is designed to assess it.
The gap between the two shows up most clearly at the moment of first contact. When a seller enquiry arrives, the activity layer captures it accurately: source, time, contact details, message content. What it does not capture is what the message reveals about the seller’s situation, their readiness, their motivation, or how close they are to making a decision.
An agency with a sophisticated CRM and a well-configured follow-up sequence will respond to that enquiry quickly and consistently. But it will respond in the same way regardless of whether the seller is ready to instruct next week or is twelve months away from being in a position to move. The activity layer cannot distinguish between them.
Activity tells you what has happened.
Intent tells you what is about to happen.
No existing tool in the standard estate agency stack measures both.
The missing layer
The technology gap is not in the activity layer. That layer is well served. The gap is in what comes before the activity starts.
Before a seller enters a CRM record, before a follow-up sequence begins, before a negotiator picks up the phone, there is a moment where the enquiry could be read for what it contains. The seller has written something. That message contains information about their situation. How motivated they are. Whether they have context that suggests they are close to a decision. Whether the language of the enquiry reflects someone preparing to act or someone exploring a possibility.
Current technology does not read that message for intent. It reads it for data points: name, email, property address, enquiry source. The content of what the seller has written is treated as unstructured text, passed to a negotiator, and assessed informally if it is assessed at all. The consequence of that informality is a specific kind of failure — one that most AI systems in the market have not solved, and some have made worse.
The result is that intent assessment, where it happens at all, is performed inconsistently by individuals working from instinct rather than structured information. One negotiator reads an enquiry and recognises that the seller sounds serious. Another reads an equivalent enquiry and files it in the standard follow-up sequence. The outcome for the seller depends on who happens to read their message, not on what the message actually contains.
That inconsistency is not a training problem. It is a structural absence. There is no layer in the standard technology stack that exists to assess intent systematically before the activity layer begins.
What measuring intent changes
Adding an intent layer to the technology stack does not replace anything that already exists. The CRM still records activity. Follow-up sequences still run. Reporting still tracks what the team has done.
What changes is the information that the activity layer acts on.
Instead of every enquiry entering the same queue and being processed in the order it arrived, enquiries are assessed for instruction likelihood before the activity begins. The sellers who are closest to instructing are identified at the point of first contact. The negotiators who call them first are calling with a reason: not because it is Tuesday and the automated sequence triggered, but because the enquiry itself indicated that this seller warrants immediate attention.
The activity layer becomes more effective because it is acting on better information. The same calls, the same follow-up sequences, the same reporting infrastructure, all informed by an assessment that currently does not exist in any estate agency technology stack.
That is the gap. Not in the tools that record what agencies do, but in the layer that should exist before they do it.