News

Industries

Companies

Jobs

Events

People

Video

Audio

Galleries

My Biz

Submit content

My Account

Advertise with us

Let your local research fly online

Online research isn't just the next thing, it's the only thing. And if you do it right, it's more than just fast and cheap - it gives you accurate, reliable data, even in Africa...

All your consumers are online - so why isn't your research?

Grant Fraser of Panel Services Africa admits there are all sorts of glorified research services in the country, but it's their own partner Fly Research that they most emulate - Fly is the UK's quickest bespoke research agency, with experienced researchers using the latest mobile and online technology - that's right, they only do online and they do it fast and accurately - which is their secret strategy. That's because more and more people are spending more and more of their time online. So why not use this as the best platform to find out more about their interests and experiences with your brand?

Grant Fraser of Panel Services Africa
Grant Fraser of Panel Services Africa

To explain the point, Grant brought Greg Ward, a director of Fly Research cited on his own company website as 'an ideas man who loves a research problem challenge, over to South Africa on his annual trip to our shores, to explain just how important online research is.

Online research is no longer just a "funny thing on the side" - it's proven itself as an effective way to get results - within the hour, at a push, and at a third of the price of traditional face-to-face or telephonic surveys. Online penetration in the UK is at around 90%, so it makes sense to conduct research online.

Some might argue South Africa is still significantly behind the curve in this regard, but Ward says to think again: Based on his annual visits he can see that we're actually catching up quickly, especially with the rise of the R500 smartphone handset. In fact, Ward expects us to be fully on-board the online research trend by the time he visits again next year.

The reasons are multiple - in addition to the increased speed and reduced cost, there's the added benefit of accuracy and validation of results, as the survey respondents are self-completing, so there's no chance of a field reporter misunderstanding the respondent or making an error during data capture.

So the question has changed, essentially from "Why would we do research online?" to asking for more information about it and being more open to it, and a shift away from thinking of what the internet couldn't do to what it could do - the future of market research is online.

Greg Ward of Fly Research
Greg Ward of Fly Research

The South African online market research model

Ward says one of the questions he's asked most frequently when he comes to our shores is just how different South Africa is. He says it has its unique qualities, as does every other country, but as someone who loves coming here, he sees much more that's the same across the globe than the average South African does. We drive the same cars and have the same aspirations, there's more that's similar than different. That's why we should be taking online more seriously, the way Europe and the US is doing.

One aspect of this is to look at a very sophisticated conjoint or trade off. For example, a typical survey questionnaire will ask respondents whether price is important to them, as well as whether quality is an important factor. These are not hugely helpful results when taken in isolation, so we need to understand the personal trade-off between them, which will differ from person to person. In doing thousands of interviews and trying to understand the dynamics of how people make choices, we can help brand owners understand the potential. Fly Research focuses on this through its ACBC or adaptive choice-based conjoint, which looks at a complex market topic like music streaming and the things that might matter - whether these are audio quality, the size of the music library, the social media connections and more, focusing on how many of those actually matter to each specific respondent. The choices are adapted as you go along based on how you respond - for example, you can see a specific respondent turns down the option of 'standard quality audio' each time it's offered and move on to ask whether she wants high quality audio, or even drop the question entirely. "Imagine doing that over face-to-face or telephonic," Ward asks - certainly not likely in a 20-minute questionnaire.

Ward's prediction for the next time he comes out is that South Africa will definitely be using more online research because it's smart to do so, not just cheaper and easier. Think for example of how the UK general election pollsters got it wrong (here's a reminder of how Pondering Panda linked this to non-representative samples and invalid data). One way around the problem is to use a continuous panel in your polling. While many only have a 4% response rate, Fly Research can get 95% of first-time respondents to complete future surveys. It's all about compliancy. If you're using a continuous panel, it's the same people sampled again and again, so it's their individual preferences that change over time and you can eliminate any 'between sample' error. The usual problem is figuring out how much it's changed, but how do you know the other variables haven't changed too - what's the churn underneath the surface, asks Ward. If you're using a quality continuous panel, where you build a relationship with respondents who are treated with respect, you can always go back and ask them why they've changed their thinking since the last time around.

The advantage of accuracy

That's one of online research - you can understand the churn of gains and losses under the surface, plus you can do more with the data. Within market research codes, the resulting data is anonymised. But that doesn't mean you can't get deeper data if you'd like. In fact, it's all the easier to do so, as you've already got the respondents' demographics level information. So if you simply want to ask a question on lifestyle, you can do that and only pay for the handful of questions asked, as opposed to starting from scratch with all the basics. Ward calls this a laser-targeted smart service as the client may only need to pay for the targeted questions, and still receive a wealth of helpful data.

In using a continuous panel you're getting more than just a window of time. As you're questioning the same people, you can start looking forward, not just in rear view mirror, and actually predict future behaviour.

Let your local research fly online
© Scanrail – 123RF.com

Fraser added that good research should be about raising more questions, not just answering them. After all, if you preach to know all the answers already, why did you ask questions in the first place? We don't often ask why people become respondents. Ward says of course payment is a factor, but just as important is the belief that your opinion is being heard. Take for example, the fact that things like age and gender don't change at will. Most access panels keep asking the same basic questions, which just sends the message that they are not listening. With online research, the respondents' age and gender is stored on the database, so it's not necessary to ask the same questions each time. It boils down to being more responsible and actually gaining better quality information as you're only asking the things you didn't know already, which helps in keeping it short and to the point.

Flash research and 'customer in the room' simplify and speed up online research

Ward explained two of the offerings that set Fly Research apart from its competitors. The first of these, called Flash research, is "as next generation as it comes", he said, explaining that it's a tool based on computer Flash hardware, which allows you to drag and drop clips onto certain zones based on your affinity for them. This is fun for respondents as there's a gamification aspect that alleviates the 'death by grid' problem and offers better context. This, in turn, means you get real rich data that you can rationalise. One example of this method of research, based on ascribing descriptive phrases to well-known TV stars in the UK, was so effective that Ward reports they got a 93% approval rate for it. They were also commended for 'making it shorter' - odd as the survey had an average dwell time of 33 minutes, up to double the average 15 to 20 minute completion time. This was probably because TV is a nice topic, Ward says, but it's the nature of the Flash-based 'next gen' survey, too. It isn't just a quick, cheap and simple research option, it's also quite clever stuff.
Ward spoke of using this as a basis for the "have internet, will travel," model, where you can create an online survey, download it onto a tablet and take it out into the field, even if it's a rural area without internet access. The device then pushes the data back when it has internet connectivity again or when you're ready to download it. This is a great way to bring complicated online surveys to people who aren't technically online.

Next is Customer in the room, which plays to the speed of things. This is helpful if the client "rings at 4, puts you in the field at 6 and expects results back at lunchtime the next day - we can even get them back in half-an-hour if we push it," Ward explained. Think of a big strategic meeting where the entire company's involved - everyone is there except for the customer. With 'customer in the room' the client is actually there, online in real-time as you discuss different ideas. You can literally watch the data come in from anywhere online. Even better? Fraser interjects that while most research companies say you need a thousand people to be representative, you can simply look at the first 'fifty or so completes' to get an idea as the die is cast quickly. The best part is that all of this takes place as the meeting is taking place, so you can refine the good ideas and drop the poor ones or defend or rephrase them if they're misunderstood. This is 'real-time research in a real sense', as you're getting the results back right there and then. This is a huge step as often strategic meetings reach a crescendo but go downhill fast and become time wasters if no real conclusions are reached. Fraser says to get any value out of the day you need to reach some concepts to test, as soon as possible.

Seems online isn't just for the future - the future is now. For more on getting started with online research locally, contact Panel Services Africa on http://www.panelservicesafrica.co.za/contact.

About Leigh Andrews

Leigh Andrews AKA the #MilkshakeQueen, is former Editor-in-Chief: Marketing & Media at Bizcommunity.com, with a passion for issues of diversity, inclusion and equality, and of course, gourmet food and drinks! She can be reached on Twitter at @Leigh_Andrews.
Let's do Biz