A West Wing and a prayer

Joe TwymanHead of Political and Social Research (EMEA)
May 16, 2011, 3:15 AM GMT+0

Recently I was watching an old episode of The West Wing from the very first season.

Now I know that not everybody spends their time watching repeats of American political dramas, and even those people seldom get excited when almost an entire episode is devoted to polling, but for me (and quite possibly only me) it was a reminder of just how much things had changed in ten years.

During this particular episode the characters spent days speculating, debating and generally sweating over the approval ratings for President Bartlett they were eagerly waiting to receive.

One of the fictitious White House staffers even asked why it takes so long to get the results. Because that’s how long these things take, they are told, before it is added that only one in four people who is called actually takes part.

Telephone polls would have been lucky to get a one in four response rate back then. Now, ten years later, they would be really lucky to get anything close to that.

Tales from a garden shed

What makes this episode so interesting for me is that it was first broadcast back in May 2000, almost exactly the same time that YouGov launched, ten years ago this year, from a shed in a garden in Great College Street, Westminster.
Whereas back then telephone polls would take days, we now have internet polls that can deliver results in 24 hours or even, in the case of the election debates, minutes. However, ten years ago using the internet for nationally representative polling was almost entirely unheard of in Britain.

Back in 2000 there were rumours that internet research might be popular within other research agencies, though little had made it into the outside world at that stage. At the time the internet was becoming good for getting illegal music thanks to Napster and was already good for geekery and pornography (sometimes at the same time), but it was still not a massively popular place for researching people’s opinions – at least in the UK.

Over in the United States things had moved on slightly further and, partly as a result, my previous company had tentatively dipped its toe in the water of internet surveys, but it remained within the domain of computer professionals. For the large established companies it was just too much to risk. Why put their reputations on the line by embracing online? What if things went wrong, as they frequently did during the early days of internet-based surveys?

Blazing the trail for internet research would instead need to be done by those with little or nothing to lose, but everything to gain. The small band of us at YouGov were able to capitalise on the situation and use it to our full advantage.

Getting things right

But of course it did help that we kept getting things right. In fact it was vitally important that we kept getting things right against the actual measurable outcomes that came our way. When we were virtually spot on predicting the 2001 General Election few looked up from their newspapers to notice. When we predicted Iain Duncan Smith would triumph in the ensuing Conservative leadership election more attention came our way. By the time we confidently went against supposedly-expert opinion and proclaimed that Will Young would triumph over Gareth Gates in the first series of Pop Idol, many were finally starting to take notice.

Our critics would claim that just because we can predict elections and things like Pop Idol, does not prove everything we do is accurate. And they were right, but we always preferred to approach things from a position of having got things right, than having got things wrong.

Once we had what Dragons’ Den would no doubt call ‘proof of concept’ we were able to sell the other advantages of accurate, online, nationally representative, research – the significantly great speed, the substantially lower costs, the tracking of respondents, the ease of use for respondents etc. etc.

From strength to strength

Riding on the back of such numerous advantages the company went from strength to strength. From the shed in Westminster to a basement in Farringdon, then it was off to an entire floor of an office in Smithfield before ending up where we all are now: two floors down the road from the Old Street roundabout.

At the same time that both our reputation and office space were growing, so was our panel. Our respondents were always our most treasured of assets and the close relationship we had with them was almost our top priority. As the size of the panel grew so did our commitment to maintaining this relationship. There was much rejoicing the day when we were sending out so many cheques to respondents each month we decided to purchase an automatic cheque printer.

Despite the change of scenery (and the number of windows) our eye for methodological detail and our commitment to accuracy has remained the same.
Closely coupled to this is our desire to innovate. Whereas previously YouGov was the new (and only) kid on the block now we have been joined by numerous others offering online nationally representative polls. However, as time has passed we have consistently expanded beyond the perceived boundaries of conventional research.

Into Iraq…and beyond

YouGov pioneered the principle of Daily Polling. Whereas receiving new data daily is now commonplace at YouGov, it was first actually used by ITN during the early days of the Iraq War, with our surveys launching every morning and delivering results for that evening’s news.

Our surveying in Iraq extended further. We were the first polling agency in the world to conduct a survey inside post-war Iraq and the first to offer commercial polling in that country.

From pioneering Daily Polling, we turned up the speed and moved even faster. During the recent General Election campaign we were the first agency to publish results in the immediate aftermath of each of the three leadership debates. Delivering over 1,000 respondents in around ten minutes.

BrandIndex is another example of YouGov pushing the boundaries of Daily Polling beyond the political realm for which we are most well-known. Five years since launching it now tracks 850 consumer brands tracked across 34 sectors in the UK (and thousands of brands worldwide) recording seven unique measures every single day.

More recently we developed TellYouGov, offering panel respondents an opportunity to tell us their views in real time without waiting to be asked in a survey, and for responses to be analysed in a quantitative manner, shifting the model away from conventional theories towards new territories.

One result of ten years of such activity is that a huge wealth of data is continually being collected. To take advantage of this we launched SixthSense, utilising our data to provide a new concept in comprehensive market intelligence, with reports that are updated as soon as anything changes in the market.

There will be many more such advances in the future, that much is certain, but what is less clear is exactly what they will be and how they will work. Nobody yet knows. However, while nobody can be entirely sure where things will go from here, I am sure that this first ten years is only the beginning.