Conversation Funnel Analysis
Conversation Funnel Analysis maps the stages a support conversation passes through - from initiation through triage, assignment, response, and resolution - and measures conversion rates between each stage. It identifies where conversations stall, get lost, or are abandoned.
Intercom metric
Conversation Funnel Analysis maps the stages a support conversation passes through - from initiation through triage, assignment, response, and resolution - and measures conversion rates between each stage. It identifies where conversations stall, get lost, or are abandoned.
Full guide: definition, formula, and benchmarksWhy Conversation Funnel Analysis matters for Intercom users
Support processes often have hidden inefficiencies - conversations sit unassigned, get routed to the wrong team, or fall through the cracks between handoffs. Funnel analysis makes these invisible failures visible and quantifiable.
For Intercom teams, understanding the conversation funnel reveals whether automation is helping or hindering, whether routing rules are effective, and where human intervention is most needed. It provides the blueprint for process optimisation.
Driver
Conversion rate
Outcome · 58% contribution
Revenue
Understand and act on Conversation Funnel Analysis with KPI Tree
Extract conversation state transitions from Intercom into your warehouse and model funnel stages in KPI Tree. Visualise each stage as a child metric beneath overall resolution rate in your metric tree.
Assign RACI ownership to the support operations team and use funnel data to prioritise process improvements that will have the highest impact on resolution rate.
Get started with your Intercom data
Pull metrics from Intercom directly through the Model Context Protocol.
Connect your existing warehouse where Intercom data already lands.
Our professional services team can build you turn-key AI foundations in a matter of weeks. Data warehouse on Snowflake/BigQuery, ELT with Fivetran, all modelled in dbt with a semantic layer.
Related Intercom metrics Ready to add to your trees.
Conversation Resolution Rate
Customer SupportResolution Rate = Conversations Resolved (Not Re-opened) / Total Conversations × 100
Conversation Resolution Rate measures the percentage of support conversations that are closed and remain closed - without being re-opened by the customer within a defined window. It distinguishes genuine resolutions from premature closures that leave the customer's issue unaddressed.
View metric
Conversation Abandonment Rate
Customer SupportAbandonment Rate = Abandoned Conversations / Total Conversations × 100
Conversation Abandonment Rate measures the percentage of support conversations where the customer stops responding before the issue is resolved. It indicates friction, frustration, or perceived futility in the support experience. High abandonment often correlates with long wait times or unhelpful initial responses.
View metric
Support Ticket Escalation Rate
Customer SupportEscalation Rate = Escalated Conversations / Total Conversations × 100
Support Ticket Escalation Rate measures the percentage of conversations that require escalation from first-line agents to senior specialists, managers, or other departments. High escalation rates indicate gaps in first-tier training, documentation, or tooling that prevent frontline resolution.
View metric
Average Resolution Time
Customer SupportResolution Time = Conversation Resolved Timestamp − Conversation Created Timestamp
Resolution Time measures the total elapsed time from when a customer opens a conversation in Intercom to when it is marked as resolved. It encompasses first response time, back-and-forth exchanges, internal investigation, and any waiting periods. It is a primary indicator of support efficiency and customer experience.
View metricExplore Conversation Funnel Analysis across integrations
All Intercom metrics
Empower your team to understand and act on Intercom data
Map what drives your metrics, measure progress at any grain, prove what works statistically, and deliver personalised action plans to every team member.