A well known aspect of the service recovery literature is that customers who have encountered a service failure and complained and in turn, had their complaint well managed are more satisfied than customers who have not encountered a service failure. In business-to-business markets, solving individual customer’s complaints has been a common endeavour. This raises the topic of anonymity and the disclosure of an individual respondent’s comments arising from marketing research.

Certainly in this researcher’s experience, it is markedly more common for business-to-business marketers to conduct data collection via internal contact centres or self-administered surveys than it is in customer markets. This is particularly so when it comes to customer satisfaction research. In such cases where the organisation conducts its own data collection there is no shield of anonymity provided to the respondent.

When a respondent opts for disclosure this means that they authorise the researcher to release their responses to the sponsor of the research. As a result, the organisation commissioning the marketing research is able to view the individual results of specific customers for actionability. The organisation may find this useful to drill down on a handful of priority customers and assess their feedback relating to say, satisfaction with a particular product or relationship.

Disclosing Customers are More Satisfied

An important question to consider here is should one expect that the levels of satisfaction found amongst those respondents who disclose their response to be reflective of those who choose non disclosure? The answer is no. Across a number of studies it has been found that the average level of satisfaction between respondents who choose disclosure versus non disclosure is significantly different. Perhaps not surprisingly, it is the non disclosure group who are significantly less satisfied than those who are willing to be identified.

Indeed, several independent research studies have shown that respondents who participate anonymously are more likely to provide candid, objective feedback than those who are not anonymous.

Consequences of Removing Anonymity

This is an important point to consider given the practice of organisations collecting their own data using say, their own outbound contact centres. By default then, all participants are identified or disclosed. Withdrawing the offer of non-disclosure in this way in effect removes from the respondent the shield of anonymity. Such a ‘shield’ is more often sought by those with bad news to report than those who come to praise.

Not providing the option of non-disclosure then comes with two problematic consequences:

1. The possibility of artificially high satisfaction levels because the respondent is hesitant to be truthful when bearing bad news;
2. Possible non-response bias resulting from those preferring non-disclosure refusing to participate.

Supplier Power Changes Disclosure Rate

Following on from this finding it makes sense to consider what the disclosure rates are across the different industries and whether or not they differ. Generally speaking, disclosure rates tend to differ depending on the relationship the respondent holds with the organisation. In banking where the bank might hold security over an individual’s assets such as the small to medium enterprise segment, the disclosure rate can be as low as 50%. This is significantly lower than in other markets where there is a more typical supplier / buyer relationship. In such cases, the disclosure rates are closer to 70% or better. In cases where routine feedback is sought and a constant loop back of action arises from the research, 90% disclosure is observed.

Once a respondent does disclose a service failure, it frequently becomes their expectation that the organisation will in turn respond to the information. Not responding raises the risk of further lessening the respondent’s satisfaction. Alternatively, responding provides the opportunity to raise the customer’s satisfaction above the levels of those who had not encountered a service failure.

Generally amongst very large corporations in business-to-business customer satisfaction research, experience indicates that respondents want and expect disclosure. Respondents believe that when a supplier seeks feedback that a cycle of improvements will follow.

Revisiting Finite Populations

In business-to-business research where the concentration of sales is very high (i.e. a few clients represent a large proportion of sales), the size of the customer set, or population, can be rather small. Take for example, a fast moving consumer goods (fmcg) manufacturer who might have only 12 wholesale customers in Australia. For such a small population it would make sense to include all of the 12 in the research, that is, conduct a census. But what if not all of them wanted to participate? What if only ten agreed to be involved? The question that must then be answered is, ‘is this sample large enough to be able to draw valid and reliable inferences?’ That is, is this a sample from which qualitative findings can be drawn or are we able to draw quantitative findings from a mere ten respondents?

One way to answer this is to call on the fpc (finite population correction factor) adjusted sample size formulae. This takes into account the small population from which the sample is to be drawn and adjusts the required sample size accordingly. As a result, in this example, where the population is 12 and the required confidence and precision level is set at 95% and 5% respectively, the sample size needed is in fact, only nine (assuming a mean and variance of 7.5 and 0.93 on an 11 point Likert rating scale).

In business to business where customer sets can be small, the fpc factor allows an organisation to make assessment with known statistical dimensions.

Friend's Email:
Your Name:
Your Email:

Submit