Amazon Connect Cases helps organisations track customer issues that take multiple interactions or multiple owners to resolve.[1] That multi-contact, multi-owner nature is exactly why Cases is a goldmine for operational insight — a single case ties together every call, chat, task, and email about one issue, so the data tells you not just how long but how hard each type of problem is to resolve.
If you want the basics of what Cases is, start with our guide to Amazon Connect Cases. This post is about the next step: what to actually check for, and what each signal is telling you.
Where to look: Cases has a dedicated performance dashboard (a single place for case operations metrics), plus case-driven metrics on the historical metrics reports and via the API. The dashboard lets you compare current vs prior-period performance and drill down by template, status, queue, or agent.[2][4]
Case volume by type (cases created)
The Cases created metric counts all cases created, and can be filtered by case template and status.[3] Reading it by template is the insight — it shows you which issue types drive the most work.
Why it matters: a spike in one case type is an early warning (a product fault, a billing error, a confusing change). High steady volume for a type is a candidate for self-service deflection or a process fix upstream.
Average case resolution time
Average case resolution time measures the average time taken to resolve a case over the interval.[3] Unlike average handle time (one contact), this spans the whole issue end to end.
Why it matters: long resolution time for a case type signals a slow or stuck process — maybe it waits on a back-office team, a third party, or a handoff that keeps dropping. It's the metric that exposes slow journeys, not just slow calls.
SLA achievement rate
The performance dashboard surfaces SLA achievement rate — how often cases are resolved within their target.[1][4]
Why it matters: this is your promise to the customer (and often a contractual one). A falling SLA rate for a template tells you exactly where you're breaching commitments before the complaints arrive.
First-contact resolution percentage
The dashboard includes first-contact resolution percentage — the share of cases resolved in a single interaction.[1][4]
Why it matters: FCR is one of the strongest drivers of both cost and customer satisfaction. A low FCR for a case type means customers are having to come back — find out why that issue can't be closed first time.
Reopen rate (cases reopened)
The Cases reopened metric counts how many times cases are reopened.[3] Track it as a rate against cases created.
Why it matters: a high reopen rate is arguably the sharpest quality signal Cases gives you — it means cases are being closed before the issue is actually fixed (premature closure, incomplete fix, or a recurring root cause). Every reopen is a customer who had to come back.
Related contacts per case
Average case related contacts measures the average number of contacts — calls, chats, tasks, and emails — for cases created in the interval.[3]
Why it matters: this quantifies effort. A case type that takes six interactions to resolve is a broken or complex process; one that takes one is efficient. High related-contacts counts are prime targets for workflow redesign, better agent guidance, or self-service.
Backlog & ageing of open cases
Beyond throughput, watch the stock of open cases by status — how many are open, and how old they're getting. Status is a filterable dimension on case metrics,[3] so you can monitor what's sitting unresolved.
Why it matters: resolution time only counts closed cases — a growing backlog of ageing open cases is the problem that averages can hide. Old, unresolved cases are where customer frustration and complaints build quietly.
Drill-Downs That Find the Problem
A headline number tells you that there's a problem; the drill-down tells you where. The Cases dashboard lets you break every metric down by:[1][4]
| Drill-down | What it reveals |
|---|---|
| Case template | Which issue type is slow, breaching SLA, or reopening — the most useful cut |
| Status | Where cases pile up in the lifecycle (e.g. stuck "awaiting third party") |
| Assigned queue | Which team is the bottleneck or the outlier |
| Assigned agent | Who may need coaching or support (used fairly, with context) |
The pattern: start with the headline metric, drill by template to find the problem issue type, then by status/queue to find where it's stuck.
Turning Signals Into Improvements
As with any analytics, the point is action. Each Cases signal maps to a specific move:
- High volume for a case type → deflect it with self-service, or fix the upstream cause so the issue stops happening.
- Long resolution time → find the wait in the process (back-office, third party, handoff) and remove it.
- Low SLA achievement → re-balance capacity, re-prioritise, or renegotiate an unrealistic target.
- Low FCR / high related-contacts → give agents what they need to close first time (knowledge, authority, better workflow / step-by-step guides).
- High reopen rate → tighten closure criteria and fix the recurring root cause, not just the symptom.
- Growing backlog / ageing cases → surface and work the oldest cases before they become complaints.
Pair Cases with Tasks and conversational analytics. Cases tells you an issue type is slow or reopening; Tasks is where the follow-up work to fix it gets routed and tracked; and conversational analytics tells you what's actually happening in those interactions. Together they move you from "this is slow" to "here's why, and here's the fix."
Frequently Asked Questions
What metrics does Amazon Connect Cases provide?
Amazon Connect Cases provides case-driven metrics including cases created, average case resolution time, average case related contacts (the number of calls, chats, tasks, and emails per case), and cases reopened. The Cases performance dashboard adds SLA achievement rate and first-contact resolution percentage, with drill-down by case template, status, assigned queue, or agent.
What should I check for in Amazon Connect Cases?
Check case volume by type (cases created by template), average resolution time, SLA achievement rate, first-contact resolution percentage, reopen rate, the number of related contacts per case (how many interactions it takes to resolve), and backlog and ageing of open cases. Drill down by template, queue, and agent to find where the problems concentrate.
What does a high reopen rate in Amazon Connect Cases mean?
A high reopen rate usually means cases are being closed before the customer's issue is actually resolved — premature closure, an incomplete fix, or a root cause that keeps recurring. It is one of the strongest signals of poor-quality resolution, because a reopened case means the customer had to come back.
Why track related contacts per case?
Average case related contacts counts the calls, chats, tasks, and emails tied to a case. A high number means issues of that type take many interactions to resolve — a sign of a complex or broken process, a candidate for better self-service, clearer workflows, or process fixes that reduce the back-and-forth.
Conclusion
Amazon Connect Cases is most valuable when you read it as a lens on how hard your issues are to resolve, not just how many you get. Check volume by type, resolution time, SLA achievement, first-contact resolution, reopen rate, related contacts per case, and the ageing backlog — then drill by template, status, and queue to find exactly where to act. Watched this way, Cases stops being a ticket list and becomes a map of what to fix.
Sources
- AWS, Cases performance dashboard (cases created, avg resolution time, SLA achievement, FCR, drill-downs)
- AWS, Amazon Connect Customer launches performance dashboard for Cases
- AWS, Connect Customer Cases metrics (resolution time, related contacts, cases created, reopened; filterable by template/status)
- AWS, Amazon Connect Customer Cases (multiple interactions/owners)
Content was rephrased for compliance with licensing restrictions.