Contact centre contact reason analysis
A total-volume forecast treats all contacts as interchangeable. They are not. A billing enquiry has a different handle time, arrival pattern, and demand elimination opportunity than a technical fault call. Disaggregating volume by reason makes every downstream WFM output more accurate.
Where contact reason data comes from
Wrap codes (disposition codes)
Data quality
High — agent-applied per contact
Limitation
Quality depends on code discipline. Agents under occupancy pressure choose the first plausible code rather than the most accurate. Code lists that are too long (50+ codes) or too granular produce unreliable data.
How to use it well
Audit wrap code usage monthly. Flag contacts tagged with umbrella codes ('Other', 'General enquiry'). Run coding calibration sessions where WFM and Operations review the same call and independently apply codes.
Speech/text analytics
Data quality
Very high — system-applied, not agent-dependent
Limitation
Requires investment in analytics tooling and model training. Accuracy improves over time as the model is trained on a larger sample. Initial categorisation may require human validation.
How to use it well
Use as a cross-check against wrap code data in the first 6 months. Once validated, use analytics categorisation as the primary source and wrap codes as a secondary validation layer.
IVR/digital channel intent data
Data quality
High for self-served contacts; moderate for contacts that abandon the IVR and reach an agent
Limitation
Captures the customer's stated intent at IVR entry, which may differ from the actual reason for contact if the IVR options do not match the customer's need.
How to use it well
Use to forecast the volume reaching each IVR routing destination. Do not assume IVR intent = contact reason for contacts that abandon the self-service flow.
Manual sampling
Data quality
Low — sample size too small for statistical reliability
Limitation
A 50-call sample per week cannot produce reliable reason-level trends. Suitable for initial hypothesis formation but not for operational forecasting.
How to use it well
Use only for initial audit of wrap code accuracy or for investigating specific volume anomalies. Do not use manual sampling as a primary data source for WFM forecasting.
Five WFM uses for contact reason data
AHT modelling by contact type
How reason data helps
Disaggregates total AHT into component AHTs per contact reason. Staffing model applies the correct AHT to each volume component before combining. Prevents the blended AHT from masking spikes in complex contact types.
Without it
Total-volume AHT forecasts are systematically inaccurate when the mix of contact types shifts — and mix shifts are common after product changes, billing events, and marketing campaigns.
Interval-level volume forecasting
How reason data helps
Different contact reasons have distinct intraday arrival patterns. Technical fault contacts peak Monday 08:00–10:00. Billing contacts peak month-end 09:00–11:00. Disaggregating and recombining at interval level produces more accurate interval forecasts than forecasting total volume.
Without it
Total-volume interval forecasts average out peaks that only appear in specific contact types. A 10% spike in complex technical contacts may not be visible in the total volume line until it has already damaged SL.
Event-driven volume forecasting
How reason data helps
Identifies which contact types are triggered by specific events (billing runs, system outages, product launches, marketing communications). The WFM function can forecast the event-driven component separately and layer it onto the baseline only when the triggering event is scheduled.
Without it
Event-driven volume spikes are unpredictable at total level. At contact-reason level they are highly predictable — the same billing event generates the same billing-enquiry volume spike within a narrow range every month.
Demand elimination targeting
How reason data helps
Identifies which contact types represent avoidable contacts (confused customers, failed self-service, post-process notification contacts that could have been proactive) vs. essential contacts (genuine customer need). Demand elimination investment (better self-service, proactive comms, IVR deflection) should target avoidable contact types with high frequency.
Without it
Demand elimination initiatives are aimed at total volume without knowing which contact reasons are driving it. Reduces the probability of targeting the highest-opportunity reasons and increases the risk of deflecting valuable contacts.
Impact measurement for change initiatives
How reason data helps
When a product change, process improvement, or communication change is made, the WFM function can measure whether the targeted contact reason decreased — and by how much — without it being masked by unrelated volume movements in other categories.
Without it
A 5% reduction in a specific contact reason is invisible in a total volume trend if unrelated contact reasons increased by the same amount. Without reason-level data, the impact of change initiatives cannot be isolated and proved.
Contact reason analysis questions
How does contact reason analysis improve WFM forecasting?
Three mechanisms: (1) Disaggregated AHT — different contact types have different handle times; blending them into a single AHT hides mix-shift effects. Applying per-reason AHT to per-reason volume produces more accurate staffing requirements. (2) Distinct arrival patterns — billing enquiries peak at month-end; technical faults peak Monday mornings. Forecasting by reason and combining at interval level produces sharper interval predictions than total-volume forecasting. (3) Event-driven components — some contact types are reliably triggered by specific events (billing runs, maintenance windows, marketing sends). These are unpredictable at total level but highly predictable at reason level once the event calendar is known.
Related guides
Volume forecasting
Building the total volume forecast
Forecasting methods
Statistical models for volume forecasting
Wrap code design
Designing a reliable reason taxonomy
Demand management
Reducing avoidable contact volume
AHT guide
Managing AHT by contact type
Forecasting review
Monthly forecast accuracy review
Deflection ROI calculator
Model the saving from deflecting avoidable contact reasons
Forecast accuracy calculator
Measure WAPE by contact reason to target forecast improvement