Every company will agree that listening to the voice of their customers is highly important. There are obviously multiple ways to do this – you can ask your customers for feedback via multiple choice surveys (e.g. in customer satisfaction polls, or NPS surveys); collect interaction data (e.g. from sales orders, web stats, helpdesk tickets); or engage your customers through in-depth conversations (e.g. through contextual interviews and co-creation labs). A good mix of these approaches is usually the most successful, as they all have their own strengths and weaknesses.
Big Data is a relatively new ‘kid on the block’, in terms of listening to the voice of customers. It is an extension of interaction data, with the advantage that data collection is automised. However, as a result of this, the amount of data currently being collected is exploding. This creates interesting potential for insights and innovation, but also some confusion. For companies who are serious about harvesting useful and actionable insights from big customer data, two things are important to keep in mind:
1. Big Data collection alone is insufficient
Just amassing enormous amounts of data with no clear plan, and then expecting to magically uncover amazing insights is a naive and not very successful strategy (if we can call it a strategy at all). Industry-wide monitors have pointed out that 75 percent of companies are just collecting data, or only very lightly analysing it, without deriving many actionable insights (Alan Porter). These companies are clogging up a lot of customer data, but they are not actually listening to what their customers are saying to them, and therefore not learning much from it.
A survey in 2017 by Temkin Group among 200 large companies revealed that while most companies believe their efforts of listening to the voice of their customers are successful, less than one-quarter of companies consider themselves good at making changes to their business based on insights from analysis. Many companies reported to find their Voice of the Customer (VoC) programs to be primarily valuable for “identifying and fixing quick-hit operational issues” and only slightly valuable for “identifying innovative product and service ideas.” There seems to be a large untapped potential here (more about this below).
Many companies appear to be overwhelmed by the volume of data they’re getting in and the lack of a plan on how to make use of it. The most reasonable answer would be to strategically invest more resources in analysis and sense making. The companies with the most mature VoC programs are not only using more data sources, they are also doing more analytics, and dedicate more specialised staff to this. In 2013 only 2 percent of companies claimed to have transformed their business by using the insights they derived from the collected data. In 2017 this percentage of ‘transformers’ is still only 20 percent (Temkin), but at least it is an increasing trend across the industry.
2. Patterns in Big Data analysis need interpretation
When analysed properly, big data can point to relevant patterns and trends in customer behaviour and preferences. Usually these are general indications of something that is going on, but not yet an in-depth explanation of why this is happening, or how customers feel about it. For instance, are these deliberate, happy events, or error-based and frustrating occurrences? Big Data does not offer a one-stop-shop for in-depth customer insights. To derive meaningful insights from the collected data, companies need to be able to correlate the “what” and the “why”. They need to take a holistic approach and consider all types of customer feedback, including interaction data, as one unified data set.
Big data alone cannot tell you everything about your customers. A considered mix of quantitative and qualitative research is needed to develop an in-depth understanding of the underlying motivations, aspirations and concerns of patterns and trends in customer behaviour. For instance, you can roll up from issues that are repeatedly occurring in webstats or mentioned during recorded calls, to survey scores and more in-depth feedback through conversations with single customers. Sometimes the approach is quantitative first — finding patterns and then diving in to really understand them. And sometimes it is qualitative first — finding hypotheses and defining the data you should collect and analyse. By taking such a holistic and more strategic viewpoint of learning from the voice of customers, the odds of identifying actionable changes that may lead to valuable business transformation are considerably higher.
Top: Image from commons.wikimedia.org
Middle: Image from Temkingroup
Bottom: Image from Microsoft