Surveys influence customer loyalty and buying habits
It's not just unethical practices like "push-polling" and "sugging" that influence survey respondents. (Push-polling is where untrue ideas are presented to a survey respondent knowing that an element of doubt will be created - "If I were to tell you that politician X was a child molester would that alter your view of him?" Sugging is an abbreviation for selling under the guise of researching and is another dubious practice which is illegal in America where the study referred to here was done.)
There's an interesting article in the 1 May 2002 edition of Harvard Business Review by Dholakia and Morwitz that shows that surveying satisfied customers just once can increase profitability over a year. They conducted genuine customer satisfaction surveying on just under 1000 customers of a financial services institution and found a huge difference between that group and an un-surveyed control group. The surveyed group were three times as likely to have opened new accounts, were half as likely to have defected, and were more profitable than the control group. It should be said that the survey respondents were mostly highly satisfied at the end of the survey anyway, but they had been selected at random by highly respected academic researchers in the first place.
Interesting! The full academic version is to be published later this year, when more details may well become available. Let us know if you're interested in hearing more!