|
Survey-driven attitudinal analytics have exposed the hearts and minds of real website visitors, and the findings have started a whole new wave of conversations about the voice of the customer in corporate boardrooms. Gone are the days when customer relationship management meant only periodic telephone satisfaction surveys and bi-yearly focus groups. With online attitudinal reporting at their disposal, progressive decision makers are continuously attentive to the voice of their customers.
This flourishing industry is not without its critics, however. There are influential voices in the field of market research that have lobbied for a more minimalist and simplistic approach. In his seminal work on customer loyalty and brand evangelism, The Ultimate Question, consultant Fred Reichheld argues that satisfaction surveys are poor barometers when it comes to the true salient issues with which companies are grappling. Further, he dismisses the notion that high customer satisfaction, within the frame of feedback surveys, has any causal link to corporate performance and growth.
Reichheld’s work is not to be taken lightly. Indeed, he is incontrovertibly a pioneer in the movement to monetize customer loyalty. Yet, in many ways, he overstretches himself in this book and attempts to impugn the value of customer satisfaction surveys in ways that are, at best, misguided and, at worst, misinformed. All of his analysis is based on offline observations; indeed, despite being written within the last eighteen months, his book has a decidedly dated feel to it. Nowhere does he grapple with the nascent phenomenon of survey-based web analytics and the way in which this type of analysis can yield insights that his monochromatic approach never could.
Yet given his stature and the reputability of the company he works for, Reichheld’s arguments have often been cited as reasons why online customer satisfaction surveys are not the way to go. Therefore, this paper will do more than just meet and discredit Reichheld’s objections—it will propose 10 reasons why satisfaction surveys absolutely do not fail.
People like knowing that their opinions are valued, but they are always conscious of the demands being placed on their time. Reichheld contends that when “surveys grow to thirty or forty questions or more…response rates drop, and the sample size shrinks,” a pattern that, according to him, “introduces sample bias and makes scores volatile and unreliable.”1Gratuitously long telephone surveys can certainly become tedious very quickly, especially if there are young children needing attention in the background or dinner boiling on the stovetop. Responding to more than 10 questions in a situation like that might be an extreme exercise in patience. Yet with an unobtrusive online survey, delivered after the site visit, there are surprisingly low leakage rates, even as the number of data points collected exceeds 30.
Consider the following example, taken from data collection for a major electronics OEM. Structurally, our survey allows for drop-off only after the first two sections, which, in this case, comprise 25 data points. As the graphic below demonstrates, even as the number of data points collected rose to 30 and then 35, the abandonment rate never exceeded 8% of the total sample during this five-month collection period. A robust enough sample base, devoid of bias, was maintained at all times.
A similar pattern emerges across industry verticals, such that low samples sizes are most frequently attributable to low site traffic, rather than wholesale survey abandonment. While response rates are not dissimilar to those observed in offline market research, completion rates regularly trend above 90%. People like keeping their word and seeing a survey through to its end. By soliciting responses without obstructing the site visit or taxing the visitors’ time, it is that much easier for them to do so. Further, by employing a 3-month cookie, visitors who have already taken or refused the survey are not then solicited again. Experience has borne out the fact that indiscriminate surveying only yields respondents upset at being queried for the third, fourth, or fifth time. Minimizing response fatigue and making sure not to over sample—these are two small but significant ways in which the survey experience is made that much more visitor-friendly.
|
