GrabIT

Description

We were part of a team that developed a next-generation customer satisfaction measurement by creating a comprehensive Customer Experience Score (CES) for every customer. Departing from traditional metrics like Net Promoter Scores and satisfaction surveys, we aimed to capture the complete customer experience across all interaction channels. By leveraging AI models to score each interaction based on its impact, the solution generated individual CES scores for each channel and aggregated them to provide an overall score. This approach enabled detailed analysis of customer satisfaction at various levels, which then enables segmentations by demographics, channel-specific satisfaction assessments, and scores for specific stores, regions, and employees. This not only provided a transparent view of user satisfaction but also empowered actionable insights for improving customer experiences, addressing technical issues, and enhancing user satisfaction across different touchpoints.

Solution

A customer has many interactions with the telecommunications company (channels), through retail stores, call centers, the website and mobile apps, upgrades and downgrades of packages, to actions that the customer does or can happen to him, such as the amount of data usage, number of dropped calls, slow internet connection, low queue lines at the stores, long wait times at the call center, interaction with the live chat for support and more. The goal is to capture every single interaction for every customer, create operational KPIs and with the use of AI models score them from 0 to 100 based on the positive or negative impact of that interaction. Slow internet connections, long queues, transfers between agents may generate negative experience scores, whereas positive interactions will generate high scores on the ranking. With this we create separate CES scores for every channel a customer might interact with the provider, as well as we can aggregate to get the overall user satisfaction score of that customer. After this not only that we can analyze the satisfaction of our customers, but we can also separately analyze the satisfaction of each channel (e.g. retail stores, internet reliability, call center) separately. We can also perform segmentations and drilldowns, for example the CES scores of certain demographics. This platform also provides us with a robust framework where we can also analyze customer satisfaction of single stores (where we combine the CES of the customers that visited that store), CES across regions, CES of the employees (indirectly via the CES of their customers that they helped) and more.

Results

With the development and implementation of the Customer Experience Score, we have created a glass-box approach to gauging user satisfaction, where all component ratings that build up the CES score are available. We can look at both overall and individual experience ratings, perform hierarchical drilldowns, stratifications, as well as many additional analyses based on the data. Using these techniques the company can then perform actionable tasks. For example, if the CES score in a particular region is low for dropped calls, that might be a cause to fix some technical issues. If the CES score is low in a particular store, or for a particular agent, actions can be taken to improve user satisfaction for the users interacting with those channels. The implementation of the Customer Experience Score not only provided a transparent view of user satisfaction but also empowered actionable insights for improving customer experiences, addressing technical issues, and enhancing user satisfaction across different touch points.