Customers have always been important to companies, but what are the best metrics to measure the success of customer-related activities, and how well companies meet customer expectations?
At the beginning of 2012 I carried out benchmark research into customer relationship maturity. In it I asked respondents what were their most important customer-related metrics. The results show the three most important metrics were how much revenue each customer had generated (71%), customer satisfaction scores (50%) and how much it costs to provide customer service for each customer (49%). As they were quite new at the time, not surprisingly customer effort (fifth out of nine), net promoter (seventh) and influencer (eighth) scores came further down the list of priorities. The most surprising insight was that companies on average use only two to three metrics, and the majority use just one or two.
More recently I finished a benchmark into customer feedback management where I asked a similar question but with slightly revised options. The top three metrics were customer satisfaction scores (71%), first-contact resolution rates (31%) and revenue generated by each customer (30%). To keep the results in perspective, net promoter scores (NPS) was fourth out of nine, customer effort scores (CES) was seventh and influencer scores ninth. In a remarkably similar result to my findings in customer relationship maturity, the average number of customer feedback metrics used by companies remains between two and three, and once again the majority use just one or two.
Without reading too much into the two sets of results, because the options were slightly different, I found:
I went one step further in the customer feedback benchmark so that I could determine how well organizations are performing against key metrics. Far more of the respondents (approximately 70%) knew how well their organization was performing in terms of customer satisfaction compared to customer value, NPS and CES, where only between 50 and 60 percent knew how their organization was performing. However, the results are not encouraging, with only a third beating the CSAT target, 20 percent beating customer value, 15 percent NPS and 10 percent CES, and in all cases about a quarter being on target.
One contributing factor to poor performance is that many interactions flow through a company’s contact center, and my research into the agent desktop shows that most agents face considerable hurdles handling customer interactions. The first and foremost is the state of their desktops, which the research found to be cluttered, with several systems to handle different communication channels, multiple business applications, dashboards and, increasingly, collaboration systems. Added together, these make the agents’ lives difficult, which decreases agent satisfaction and makes it only half as likely they will meet targets for these key metrics. In an effort to improve, we recommend organizations take a close look at their agent desktops and determine what they can do to make them easier to use and smarter in supporting agents.
In both these and other benchmarks I have also sought to discover how organizations produce their metrics. The most popular tool remains spreadsheets, which are not good at combining information from multiple systems, are labor-intensive, can’t produce results in near real time and require a lot of skill to develop formulas to automate the calculation of the more complex metrics. They also can’t process unstructured data, so organizations are missing out on insights locked into call recordings and text-based interactions (emails, letters, surveys, text messages, IM scripts, web scripts, social media posts). Unlocking these insights requires using advanced voice and text analytics, which only the most mature companies have deployed.
As companies try to improve customer-related activities and business outcomes, we recommend companies start by reviewing the metrics they use in order to gain a better balance between efficiency and effective (outcome) metrics. Organizations should evaluate new analytics tools available to process structured and unstructured data, and above all they should share the results more widely so that everyone in the organization can see how what they do impacts customers.
Regards,
Richard J. Snow
VP & Research Director