Quality and analytics
Insights and trends
How conversations are grouped to show what is changing, what customers are really asking about, where your team can be coached, and which customers need action next.
On this page
Your customers tell you what is wrong with your business every day, one conversation at a time. Read one at a time, those conversations are service requests. Read together, they are evidence: about your documentation, your processes, your products and your people. The Insights view of the console groups conversations to show what is changing, what is driving contact, and what should happen next, so you can fix causes as well as answering questions.
A signal, not just a service problem
If hundreds of buyers in one tower all ask the same question, that is not a service problem. It is a signal about your documentation, your sales process, or the project itself. The same is true when hundreds of policyholders ask about one exclusion, or when guests at one hotel keep asking how to reach the beach.
Each conversation answered well still matters. But the pattern tells you something no single conversation can: that a handover letter is unclear, that a policy wording is being misread, or that a booking confirmation leaves out the thing guests most want to know. Fixing the cause helps every customer who would otherwise have had to ask.
Topics and how they change
Topic clustering is configured during implementation. It groups conversations by what they are actually about, including subjects nobody thought to create a category for, and tracks each group over time. A topic that appears suddenly, or grows week on week, is highlighted, so an emerging problem is visible before it turns into complaints.
Topic clustering works alongside reason-for-contact tagging. Reasons for contact use the categories you agreed, so they are stable and comparable month to month.
Topics are found in the conversations themselves, so they catch what the categories miss. When a topic keeps appearing, it is often worth adding as a new category. See Outcomes and reasons for contact.
Coaching insights for your team
Where conversations held by your own team are captured into the customer record, coaching insights are configured during implementation. They compare high-scoring and low-scoring conversations against your quality standard and show what the best ones have in common: how they open, how they explain a delay, how they handle a customer who is unhappy with an answer.
Insights are presented as patterns and examples for team leaders to use in coaching, not as rankings. The point is to make your team better at the conversations that need them, with evidence from your own best work. See Conversations your team handles and Quality review on every conversation.
What to do next for a customer
A record of what happened is most useful when it tells you what to do. Next-best-action recommendations are configured during implementation. For each customer, they suggest what your business should do next, based on the customer record and the rules you set: call back a buyer whose snagging item has been open longer than your target, offer a renewal conversation to a policyholder whose claim closed recently, or check on a guest whose complaint was resolved.
Recommendations appear in the customer record and in the lists your team works from. A person decides whether to act. Where a recommendation leads to outbound contact, it follows the same consent, opt-out and calling hours checks as any other outbound message. See Contact rules for outbound messages and calls.
Customers close to a decision
In sales conversations, some customers are ready to decide and the rest are not yet. Buying intent detection is configured during implementation. When a conversation shows the signals that someone is close to a decision, such as asking about reservation fees, payment plan dates or availability of a particular unit, the right person on your sales team is notified, or the conversation is routed to them, under the rules you set.
Buying intent works alongside lead qualification, which scores each enquiry against your own qualification criteria. See Property.
In practice
A Dubai property developer uses the agent for post-sale questions on a tower approaching handover.
- In the Insights view, Mariam, the head of customer care, sees a new topic: "Parking allocation at handover". It grew from a handful of conversations to more than a hundred in three weeks.
- She opens the conversations. The handover letter tells buyers their handover date and their unit, but not their parking bay, and buyers are asking the agent, the call centre and their sales agents.
- The agent has been answering from the parking allocation list, but buyers still have to ask. Mariam asks for the bay number to be added to the handover letter.
- She adds "Parking allocation" as a reason for contact, so it is reported alongside the others from then on.
- The following month, she checks the count for the new reason against the weeks before the letter changed.
What your team controls
- Which insights are configured, and for which departments.
- The rules behind next-best-action recommendations and buying intent routing.
- Who is notified, and who can see insights, by role.
- Which topics become reasons for contact.