Call centres — or contact centres, as they are now often known, to reflect the increasing number of channels for customer contact — are a vital part of the customer experience with many companies. In my previous note, I reviewed the path that got us to the modern contact centre.

Research suggests, however, that a number of companies may be missing out on some of the benefits of contact centres, because they are not being run efficiently and effectively.

Source of inefficiencies

There are a number of sources of inefficiencies in call and contact centres. For example, many centres quite often have the wrong number of staff available — for example, too few at busy times, resulting in long customer wait times, or too many at quieter times, resulting in wasted staff time. And while having too many staff sounds like a good thing for customers, because of shorter wait times, bored staff are unlikely to be improving customers’ experience through their enthusiasm and willingness to help.

Another issue is that there is still a big discrepancy between what the customers are really calling for and what the contact centre reports are showing to the management. According to a recent pilot project SAS has run, there is up to 20% deviation between the actual call reasons vs. reported.

This analysis was done using natural language processing techniques to automatically predict and classify the call reason based on a combination of structured data and the free text logs available in the contact centre application. These logs could be potentially typed by the agents themselves or could be converted text from a speech recognition system.

A wide range of issues to cover

The lack of visibility into the actual customer call reasons - customer’s voice, if you will - causes huge impact in terms of unnecessary spending to remedy issues that may not be there in the first place while leaving the customer even more frustrated than before.

This issue occurs typically due to the human factor in the equation and could be caused by training issues, contact centre application design, KPI-driven misclassification of issues, etc.

Sign up to this webinar series and learn about the practicalities of artificial intelligence and making AI business-smart

Centres may also have the wrong type of staff available. Contact centres have to deal with a wide range of issues, from complaints through to pre-sales and sales, as well as post-sales support. SAS research shows that companies are very bad at predicting the types of calls that they will receive, and therefore staffing correctly to meet type of demand, as well as volume.

These issues have a direct effect on customer experience, and are therefore vitally important. But there is another more hidden and subtle problem: Companies may fail to take advantage of the data generated by contact centres and call centres. This might not sound like too big a problem, but it can result in some lost opportunities, and particularly a failure to understand and get to know customers better.

Improving efficiency and effectiveness

Companies that are actively using and analysing data from their customer contact centres are finding huge benefits in doing so. In the first place, they have improved their contact centre operations, by reducing inefficiencies. For example, historical call data enables models to be built to forecast likely call volumes — and therefore demand for staff — across particular time periods. These models can easily take into account holiday periods. More advanced analytics can be used to build in the effects of future sales and marketing activity such as promotions and offers, or operational issues like price changes and outages. This means that staffing requirements can be forecasted fairly accurately, and staffing levels optimised for both type and volume of calls.

Perhaps more importantly, companies have been able to use contact centre analytics to improve their customer profile information, identify other inefficiencies as a result of information from customers, and improve their products and services. All these increase ability to target offers and products more accurately, and to increase customer satisfaction.

No distinction between channels

These companies are using analytical techniques like natural language processing, text and speech analysis, and concept creation to explore and exploit data generated routinely during contact centre interactions. They are also exploiting sentiment and topic analysis to understand customers’ problems, and whether these are getting more or less important, to give them better insights into customer behaviour.

Companies that are using analytics most effectively to improve contact centre operations are bringing together data from multiple sources, not just the contact centres themselves. It seems particularly important to look at data across all channels, and think about the customer journey. Customers do not distinguish between channels, or consider that an interaction via mobile is different from a phone call. Companies, therefore, should not do so either: They should focus on getting a picture of customer preferences, motivations and behaviours, regardless of channel. Being armed with this information will make them more competitive, and more likely to survive in the longer term.

Research from Harvard Business Review Analytics also found that control of customer experience needed to lie with the business. Often, the most important improvements were not driven by the c-suite, but by teams trying out analytical solutions, and finding that they worked. The results then became a normal part of ‘what we do’, and were incorporated into ‘business as usual’.

The bottom line

Companies using analytics to manage customer experience tended to have better financial outcomes, including profitability and market share, as well as improved customer retention figures. In other words, improving the customer experience, including through contact centre analytics, very definitely does pay. The effect on the bottom line suggests that it is well worth taking time to incorporate a strong analytics approach to customer experience management.