TNS Boss Questions Veracity of Online Data Research

05 September 2007

LONDON: It's a common ploy to divert attention from a disappointing set of financials with a fuzzy but controversial sub-plot. However, the world's third largest market research company needs no such subterfuge.

The latest set of numbers from Taylor Nelson Sofres stand robustly on their own two feet, delivering a healthy 13% uplift in first-half pretax profits to £33.9 million ($68.36m; €50.11m).

Nonetheless, the company's report for the period to June 30 was suitably spiced by chief executive David Lowden, who aired his concerns over the veracity of data culled from online panellists.

The amount of TNS business via online methodology rose 22% during the first half of 2007, prompting Lowden to voice his concern at the emergence of "professional" web respondents.

Many of whom sign-up to several different online research panels - a practice Lowden fears could skew internet-garnered data.

Over-questioning of online panellists, especially in the light of of the web's low comparative cost, is also seen as a dangerous temptation. TNS, however, was quick to disassociate itself from such practices.

"Our response rates to online questionnaires tend to be 40%-50%. But there are companies where it is less than 5%, because they simply send too many questions," said Lowden.

"We have strict rules on how we use online respondents. That is not the case with all suppliers.

"We continue to emphasise to clients the benefits of using the internet as a quick, cost-effective method of data collection and to build on our global internet access panel capability. Our goal is for this to increase to 40% within 3-5 years."

Lowden added: "Quality standards in internet research are becoming increasingly important for our clients ... and we are taking a pro-active role in the industry response to this issue."

Data sourced from Financial Times; additional content by WARC staff