Issues not to be touched with a 10-foot cyberpoll

Should a major telecommunications company apologize for the unscrupulous actions of one of its employees? Yes or No?

Your hesitation to answer is understandable. This question — which I will explain in more detail below — involves a number of grey areas that demand more information. Yet

this is precisely the kind of question that is frequently asked in multiple choice Web site polls all over the Internet, and an Australian company called Telstra has shown how inadequate they can be.

The company was running two polls about its own Internet Provider Service, BigPond. The nature of the questions — “”Does Telstra’s BigPond Internet Service provide value for the money?”” and “”Should Telstra compensate its customers for regular service interruptions?”” suggests all was not well to begin with. Only 26 people responded, but the third-party provider of the poll discovered a “”bot”” had been released to boost the number of answers to an unprecedented 287 respondents in half an hour — all of them favourable to Telstra.

The bot, Telstra confessed last week, came from an IP address within its own company. This was not good news for a firm whose price hikes for Internet access have already attracted the attention of a competition watchdog for possible breaches of the country’s Trade Practices Act. But here’s the real punch line: of the initial 26 who had responded, only one said Telstra provided value for the money, and that respondent claimed to have hit “”yes”” by mistake after the poll-rigging came to light.

Telstra has yet to formally apologize for its actions, claiming they were not endorsed by the company and it cannot control the individuals it employs. Perhaps we should sidestep my original question and wonder why Telstra was using a cyberpoll to gain the customer feedback it would presumably use for decision-making purposes.

Cyberpolls became ubiquitous as soon as newspapers and magazines moved much of their content online. In the mid-90s, this was considered a shining example of the interactivity the Web brings. Everyone has one. They’re great for sports trivia: right now NHL.com is asking, “”Which team with a newly-hired head coach will finish higher in the standings this season, the Stars or the Devils?”” On ITBusiness.ca’s sister site, Mochasofa.ca, the editors ask visitors what they do to stay young, and offer four choices.

When we launched ITBusiness.ca I did not include a cyberpoll, because I have never clicked on one and have never found them to be of any value. Let’s admit it: sometimes the frequency of the polls and laziness of the creators makes for some inane questions. On Tuesday, for example, Canoe.ca was asking, “”Are you surprised that suspected members of al-Qaida were located in Canada?”” With only Yes/No type responses, what good would the results of such a poll be?

My arm has since been twisted, but I’ll admit that it has come in handy on occasion. When we covered the closure of @Home.com last year and its effects on Rogers, Shaw and Cogeco subscribers, I was wanted to get a sense of how bad the problem was. We asked through the poll, and it didn’t take long for us to receive hundreds of votes from disgruntled users.

This is what cyberpolls are good for — assessing the zeitgeist, of seeing how many people share an opinion or experience. This week we ask how many people see the need for a device like the Palm i705 in their organization. This was based on an interview I did Monday when I was told by a Palm executive that demand was low.

As the intellectual equivalent of junk food, cyberpolls are mostly harmless. But as a method of evaluating service, customers would tell you they prefer a more open-ended, direct approach. Just ask them.

sschick@plesman.com

Share on LinkedIn Share with Google+