Jon Puleston: Vice president, innovation, Lightspeed
Simon Falconer: Global director, innovation, Kantar Insights
Ansie Lombaard: Global innovation director, Behavioural Data, Kantar TNS
Jeff Levine: Executive vice president and NY market lead, Penn Schoen Berland
Question: Why have response rates been falling in many types of consumer research?
Jon: Increasing competition for our time; we are in an era of information overload and with an ever-increasing number of things competing for our attention and time that are potentially a lot more interesting to do than taking part in market research. How would you like to spend your lunch hour? Doing a long dull online survey or engaging in social media or playing a game on your mobile? There’s a diminishing pool of people: the average person does four online surveys before giving up and never doing one again because they find them boring and frustrating consumer experiences; we have burnt through literally millions of consumers like this.
We are being spammed surveys left, right and centre; part of the problem is that every website you go to, every consumer experience you have, consumers are being asked to complete often very poorly designed dull consumer feedback studies. We can’t differentiate ourselves from other forms of unsolicited marketing messages. And we are being bombarded with all sorts of unsolicited email, phone calls, text messages which results in us shutting ourselves off from third-party solicitations.
Simon: At its simplest, two reasons. First, there are other task/reward offers out there so it’s a more competitive market to persuade people to take surveys, when they may also be involved in crowd-sourced tasks, social media interactions, a brand’s digital customer service and relationship activities etc.
Second, when we do persuade people to take a survey, those surveys can be too un-engaging, long and unfit for the mobile-first world. That’s changing at Kantar, but needs to change faster across the market.
Jeff: Potential participants are harder to contact – a consequence of fewer landlines, busier lives and greater mobility, plus technology makes it easier for people to ignore phone calls from unknown numbers. Privacy and confidentiality concerns also play a part in who is willing to participate.
Question: Is the answer to offer incentives? What problems does that pose?
Jeff: Incentives have been shown to increase response rates. That said, offering incentives may have an impact on response quality, sample composition and response distributions. Incentives are not a good fit for every project, but can be effective if used appropriately.
Jon: There is potential to re-build the relationship we have with our research audiences by better incentives, but right now the market is not prepared to pay for it as there is still a lot of competition from cheap panel suppliers.
The problem will start to self level once supply really starts to dry up – we are reaching that tipping point now in some sectors e.g. surveys aimed at young adults.
Simon: Incentives are just a small piece of the puzzle. The right incentive type, at the right time, can certainly improve response rates vs. the wrong incentive at the wrong time. But a significant proportion of us are motivated by the incentive – it’s more intrinsic – of feeling that we’re ‘making a difference’ and having our voice heard. Better feedback on the role and impact people have had through sharing their opinions is needed.
Question: How about making research more engaging through design – for example through the gamification approach?
Jon: This can have a big impact engaging research audiences with better survey experiences and making the process of taking part in surveys more rewarding can be extremely effective. We have been able to achieve up to 10-fold increases in ongoing participation rates by creating surveys people want to complete using the techniques of ‘surveytainnment’ .
However this is not necessarily a solution for all the bread and butter research we conduct in the research industry. Its only practical for certain types of research.
Jeff: Decades of work has gone into the development of metrics and methodologies. There is a real science behind it. Moving away from traditional methodologies to approaches such as gamification must be done with care. There may be circumstances where gamification is the best approach, but at this time we recommend it be used with certain audiences (e.g. Gen Z) and certain situations (e.g. social media research).
Simon: We’ve seen more engaging designs, with game and competition mechanics, with higher response rates and retention rates for ongoing research programs. UI and UX design were unknown concepts to many survey designs for too long.
Question: What are the most effective ways to conduct surveys (and other research activities) today? What role can social media play?
Simon: Design mobile-first wherever in the world the survey is intended for; understand the audience and their motivations for sharing their opinions, time and ideas. With that, design the task and any rewards accordingly.
Offer redundancy – allow multiple channels and formats for the desired contributions. Some consumers may prefer to share a video, others prefer words. Some like to interact with each other, others are more private. We don’t need as many specialist surveys and research tasks as we may think – there’s an ocean of opinions, behavioural data, emotions and more with social media – already generated by people without any ‘prompt’ by a researcher.
Surveys and other research activities can also tend to be designed as ‘death stars’ – trying to do everything in one. We need to better embrace iterative design – smaller tasks and surveys that allow test and learn in a more agile way. Not only do we see improved insights as a result of this approach, we often also seen improved speed of execution.
Jeff: There is no one-size-fits-all solution. Social media monitoring can play an important role in research. When combined with primary research, it can provide valuable additional insight, particularly in certain situations. For example, appending social media data can provide context to survey results.
Jon: Online research without doubt still remains the single most efficient way to gather market research data – nothing beats it right now for speed and efficiency for global reach and consistency.
Question: Do we ultimately need consumer research, or can we gain the same level of intelligence and insight from analysing behavioural data?
Ansie: Behavioural data offers deeply granular insight into what people do – from the online search you did today to tracking your sleep patterns. Yet it doesn’t capture your attitude or motivations. This is why behavioural data will always be better when combined with consumer research to unpack the ‘why’ behind the ‘what’.
Jon: Behavioural data can be extremely valuable alternative source of research insight but there are so many holes in behavioural data and so many questions that behavioural data cannot effectively answer that there is always going to be a need for more traditional forms of market research.
Jeff: Yes, consumer research is critical to fully understanding your audience, whether it be consumers, government elites or c-suite executives. Behavioural data alone often leaves gaps in knowledge and is unable to assess causality and identify the driving factors of behaviour and perceptions. Consumer research and behavioural data used together provide a complete picture that can be effectively used to drive strategy and decision making.
Source: Lightspeed, Kantar TNS, Kantar