People answer a lot of satisfaction surveys, whether it is on social media, in the streets or for customer feedback. Such surveys are great to better know customers, understand their motivations and improve their experience. This is why a lot of companies use them to improve Customer Satisfaction.
Improving Customer Satisfaction through data can be tough
We worked on public data from a railway company . Since millions of passengers take trains every year, understanding their needs and expectations to improve Customer Satisfaction is essential. Hence, they conduct regular surveys in train stations to get passengers’ feedback and suggestions.
Surveys are then analyzed by marketing research teams to deliver insights to the customer satisfaction teams. However, new traveling habits inherited from Covid and the environmental crisis impacted the railway. Analyzing the batch of surveys usually took them seven weeks, which is not convenient to deliver a fast action plan.
They conduct satisfaction surveys twice a year in 125 train stations on an average of 75,000 customers. It means that their data sets have around 600 rows and 30 columns per station, quite a Small Data set. Unfortunately, they were using traditional Machine Learning techniques for Big Data, which meant that they required heavy data cleaning and engineering to get the best predictive model possible – hence the seven weeks.
Insight discovery 50 times quicker with Explainable AI
Enter MyDataModels. The railway was looking for workflow improvement to get insights quicker without sacrificing accuracy.
To begin with, we created a predictive model based on the data from the surveys. We used ZGP, our proprietary algorithm to analyze and train the model on the raw data. The evolutionary approach for Small Data showed its results quickly, with results in one day instead of seven weeks.
There also was no loss of quality: accuracy remained at 71% and Mean Absolute Percentage Error (MAPE) at 2%.
Moreover, our Explainable AI approach allowed the railway company to understand what were the key factors of satisfaction and how they correlated with one another.
They also showed a lot of interest in our ability to build operational scenarios and test outcomes – in that case, how it would improve Customer satisfaction. Not only did they emphasize on what makes customers satisfied and reduce the impact of what caused discontent, but they also discovered new insights. Most importantly, they did it in a fraction of the time it used to take. It helped them deploy and get feedback on these new insights much quicker.
Small Data helps you grow your business
Covid and climate change concerns have changed our behaviors towards public transportation. As people are changing the way they travel and move, transportation companies such as railways and airlines need to adapt to a quickly changing landscape. Combining regular satisfaction surveys with Small Data Analytics helps them get fast actionable insights to ensure Customer Satisfaction and recommendation.
More generally, happy customers are loyal customers who turn into ambassadors, promote your brand and business. This is why keeping customers happy is the number one priority of most companies – and why they are constantly looking to improve customer satisfaction.
By helping brands get fast actionable insights and helping them test scenarios, Small Data analysis and Explainable AI play a big role in improving Customer Satisfaction. They can also help discover untapped insights, adapt a business strategy or confirm gut feeling – so feel free to reach to discuss your game-changing use cases!