Web polls are fun and entertaining but there’s no denying their limitations when it comes to offering any form of reliable and serious scientific market research.
They have a number of issues which undermine their credibility as reliable indicators of public opinion… but where there’s a problem there’s an opportunity, or so we reckon.
In principle the Internet is a great medium for reaching people – the best. In the UK, the ONS found that 88% of UK adults (45.9 million) had used the internet in the last 3 months with only 10% having never used the Internet. Internet use is fastest growing among older and the youngest generations, and that growth seems set to continue.
So where and why do web polls fall short of the mark?
Web polls are rarely limited to the target audience you actually want to hear from. An Internet poll enables you to reach a vast global audience, but this is unhelpful for building a fair representation for the demographic you’re polling.
Participation in a web poll is determined by those who happen to come across it, and who then go on to participate. As such, a web poll is already constrained by the demographic profile of the audience that visits the website it’s hosted on. Unlike conventional market research polls, participants are not preselected in order to fit a predetermined cross-section of the target audience.
- No context or background
Web polls tend not to require any registration in order to take part. This means the poll captures no background data about the person, and hence there’s no way to determine any context and meaning for the voter’s decision.
- Duplication & manipulation
While basic controls on repeat voting are in place on most web polls, they are fairly easy to work around. This could be as simple as opening up a new browser or clearing your cookie cache. Indeed there’s a script online that can be used to hack the leading provider of embeddable online polling software.
So these four areas are enough to undermine the data value of most online polls. But is that the end of the story? I’d argue not – after all web polls have a number of things going for them, namely:
That huge Internet reach we mentioned? If used in an effective way, web polls can attract a very high and relevant participation that just isn’t achievable through conventional market research methods. Web polls have the potential to be more inclusive and more accessible, reaching more people.
- Speed and targeting
Mobile Internet and our ‘always-on’, connected world also presents web polls a unique opportunity to collect results far quicker than by other means. As people choose mobile apps over desktop, web polls can also access geo-locating and push notification technology in order to specifically target people in a specific location, and/or at a specific time.
- They’re unpaid
We’d all like money for nothing,right? Except it’s not for nothing. In conventional market research, it’s standard practice for respondents to receive some form of payment for completing a survey or participating in a poll. This is only fair as the time, effort, tedium and inconvenience commands it. However, the payment tends to be very small – so as a motivating factor it is biased towards those who feel it’s a worthwhile use of their time. Web polls trade in a different currency (the entertainment factor) and consume less time, so they don’t have this problem to contend with.
- Engagement & Feedback
In our busy lives we don’t all have time to fill out surveys, so the quick “one click” format of web polls is viable for more people. Combine that with the instantaneous rendering of the real-time vote count, and the experience can be both fun and gratifying.
So how can we keep the advantages and dispatch with the disadvantages of web polls? This is the challenge we seek to address.
Keep the good…
Pollstation’s polls are evidently web-based and offer the same user experience as conventional web polls – so they’re accessible, inclusive, offer super quick voting – and instant results. Plus we’re adding some neat bells and whistles – some of which are already in place.
And Kick out the bad…
Unlike typical web polls however, we require registration in order to vote in the majority of our polls. Certainly some will moan at this, but since a simple one-time registration process gives you access to all our polls, it should be a fairly minor inconvenience. (Especially when most appear quite happy to use the login with Facebook feature.)
Registration allows for a greater control on participation and establishes context around who is voting. Even a small number of details in a user profile can help us to derive greater meaning from the poll result, and will allow us to add filter functions giving users the chance to examine and explore the data.
Furthermore, in the near term it will allow Pollstation to apply real-time dynamic weighting to the results, in order produce an additional view of the data that’s more in tune with what you’d find from market research quarters.
In terms of our audience, of course people gravitate to websites that tell them what they want to hear. This is both a challenge and an opportunity for us to differentiate – and we choose to do this by maintaining absolute neutrality and by providing an inclusive experience that’s welcoming to all.
We don’t need to “own a niche audience” and play its tune. Because we’re polling across a broad range of subjects, we anticipate continuing to grow our user base with people whose initial interest comes from a variety of diverse and different areas, and goes on to spread in other areas.
Knowing who our users are also presents Pollstation with the opportunity to precisely target polls to specific geographic catchments and demographics. This also allows us to distinguish between voters we select and those that self-select.
Hopefully this offers a bit of insight into where we’d like to take Pollstation and how we intend to develop the user experience to make it even more interesting as we progress. This hasn’t been done before, so please join our journey and see what we can make of it!