Customer Services – Meeting Demand

The Background

Customer Service agents found themselves swamped, increasingly unable to cope with the service requests from customers, with growing backlogs of unopened emails, and unanswered calls. The situation was compounded by the nature of the requests - agents could not always resolve the issue themselves, they needed to invoke support from third parties. This created a feeling of helplessness; the customer journey was not always a simple first-touch.
The process design in place at the time supported multiple hand-offs to other teams, but internal reporting governance provided limited visibility of where the customer request was at any point in time, and no visibility of when, or even whether the problem would be resolved.  This was resulting in customer frustration and customer loss.
The CEO felt no longer able to commit to customers that their problems would be dealt with effectively and efficiently in a timely manner.

What We Did.

We undertook a rapid diagnosis, design and mobilisation approach, involving coordination flow mapping, scheduling alignment, and MI visibility.

We closely observed and measured the behaviours and activities, then  deconstructed the As-Is processes. Despite existence of written process flows, there was no unified, consistent approach to dealing with customer issues. Further, the effectiveness of the processes were measured in terms very closely related to an ‘off-my-desk’ approach, which reinforced behaviours that kept the issue moving through different teams without achieving resolution. Ownership of the issue was not a factor in the processes.
Having mapped the Coordination Flow (read more here), we reconstructed the activities in a simple, innovative process design ensuring that the customer - and the customer’s issue – was genuinely kept at the heart of the workflow.

In parallel, we used the information obtained during work monitoring to construct a resource model based on uncertain demand. A customer issue could take five minutes to resolve, or half an hour; or would need to be despatched to another team or external service provider. Volumes could be high one day, low the next. We built a scheduling model based on the heijunka approach which gave better control of meeting uncertain demand.

Finally, pulling the coordination flow and scheduling design together, we implemented case tracking MI to show where each case was at any point in time, and where intervention was needed to move customer cases along.

The Outcomes

Call abandonment rate  improved significantly, customer resolution has improved from <50% to 77% and continues to improve, with the genuinely active cases with third parties being managed to regulatory SLAs.

Understanding how to meet variable demand in uncertainty will take care of backlogs.



Meeting customer demand is not as linear as is commonly expected. Nevertheless there is still a tendency to think in terms of ‘daily demand = x, therefore staff needed = y’. Of course, there is a level where this works, but it is inefficient, fails to cater for unpredictable demand or uncertain staff availability (special project allocation, ‘burning platforms’ and so on).
Look at this model, the loss of one resource impacts service delivery from  90% to  35%.

This is the start of a backlog, which, when not properly understood, can grow out of control. Here’s a study where backlogs and scheduling directly relate to efficiency and customer satisfaction.

Meeting demand 2
Logo 2018 v1

Process Coordination Specialists