The work code the call actually earned
Disposition data drives staffing, product decisions and reporting — and in most contact centres it is entered in the two seconds before the next call connects. Xperia reads the code off the conversation instead.
Derived from the call, not from the dropdown
Xperia determines the work code from the content of the conversation itself. Because it is derived from what was said rather than what was clicked, it is independent of the agent's selection — which is what makes it useful for more than saving a few seconds of wrap-up time.
Your list, or one it writes itself
Two modes. Restrict Xperia to your existing work-code list, so output stays consistent with the codes your reporting already depends on. Or let it generate a short free-form code per call, which is the better choice when you suspect your current list does not cover what customers are actually calling about. Switching between modes is a confirmed action, since it changes how every subsequent call is categorised.
Comparing it against what the agent chose
If your telephony platform or CRM passes the agent's selected disposition through as metadata, you have two independent answers for the same call: what the agent recorded, and what the conversation contained. Where those disagree systematically — one agent, one queue, one code that is always chosen — you have found a reporting problem that no amount of averaging would have surfaced.
Dead calls are separated out
Recordings containing only IVR audio, hold music, or a call that never reached an agent are marked as dead calls rather than being categorised and scored. Without that, they land under whatever code is first in the list and quietly distort both your disposition mix and your quality averages.
Quality per work code
Once codes are derived consistently, quality can be cut by category. Which call types score worst, which generate the most fatal incidents, which have the longest handle time. That is the view that turns disposition data into an operational decision rather than a monthly count.
Ask about it directly
The AI Assistant answers work-code questions in plain language — the current code list, call volume by category, quality by code, which categories are growing. Useful when the question is one-off and building a report for it would not be worth the effort.
See it working
Short walkthroughs
Questions
Frequently asked
How does Xperia decide a call's work code?
It determines the code from the content of the conversation itself. You can restrict it to your existing work-code list so output stays consistent with your current reporting, or let it generate a short free-form code per call.
Can we check whether agents are selecting the right disposition?
Yes, if your telephony platform or CRM passes the agent's selected disposition through as metadata. You then have two independent answers for the same call — what the agent recorded and what the conversation contained — and can identify where they systematically disagree by agent, queue or code.
Can Xperia use our existing work-code list?
Yes. Restricting it to your configured list keeps output consistent with the codes your reporting already uses. The alternative free-form mode is more useful when you suspect the current list does not reflect what customers are actually calling about.
What happens to calls that never reached an agent?
They are marked as dead calls and excluded rather than categorised and scored. IVR-only audio, hold music and unanswered calls otherwise land under whichever code appears first and distort both the disposition mix and quality averages.
Can we see quality broken down by work code?
Yes. Once codes are derived consistently you can compare scores, fatal incident rates and handle time across call categories, which is what makes disposition data operationally useful rather than just a monthly count.
Related
AI Call Scoring
Xperia scores every call against 48 quality dimensions across 10 categories. Weightings, pass criteria and the parameters themselves are configurable per department.
Analytics & Dashboards
Score trends, distributions, category and parameter drill-downs and an agent leaderboard — filterable by agent, department and date range, with bulk export to PDF or CSV.
Try it on your own calls
Every account starts with $3 of free credit — around 75 minutes of call analysis. No card required.
Start Free