Technology is increasingly providing methodologies to monitor behaviour, without having to ask questions. These typically provide the most accurate measures. Dashboard, a technology driven marketing intelligence business, has built a mobile app that some members of its opted in research community installed on their phones for a three month trial. Here the researchers discuss what they have found after having successfully piloted such a methodology in Zambia.
Market research is about asking questions isn’t it? Yes it is, as this remains the primary way to delve into motivation and understand consumer decision-making. For now, asking people is still the best way to assess brand perceptions, customer satisfaction and dig beneath the drivers of behaviour.
Traditional market research can be thought of as:
|Characteristics of traditional market research||Best used to understand|
|Question / answer – Usually Researcher initiatedProblem focusedSample controlled||Motivations explaining behaviour and attitudesSatisfaction|
Listening in to social media
Asking questions is not the only way to generate insights. Social media and big data are becoming more prominent in the insight function and rightly so. The wealth of information available in these new realms augments our grasp of the market and potentially provides us with the marketing advantage. We can conceive data harvested from social media as being akin to customer eavesdropping – mostly unsolicited.
This makes it harder to analyse, contextualise and often difficult to assess if the commentary is positive or negative. For example, if someone posts that the café they are visiting is busy, this could be both negative and positive. Negative if you have to wait for service, and positive because you are at a popular spot.
|Characteristics of social media posts||Best used to understand|
|Tend to be consumer initiatedLess structured
Bi-polar (only really good and bad posts)
Self-selection – not market representative
|Passion pointsHot buttons
In an effort to save costs, there are marketers who are only using social media, alongside text analytics to inform them of how their brands are performing. This seems like a great way to save on the research budget. In reality however, the optimal approach is to combine traditional “question and answer” research with these listening methodologies or risk losing track of underlying drivers and motivations.
Still any discussion of the insight function in 2016 would be incomplete without including BIG DATA. We believe that this needs to be distinguished from social media data by the nature of the data itself. This is data that exists as a result of behaviour and it is a product of what consumers do, not what they say.
|Characteristics of big data||Best used to understand|
|Systematic – automatically recordedActual customer interactions / behaviour
No motivational / emotional clues
Behavioural and data mining
|Links to business drivers easilyActual behaviour patterns /data mining
Uncovering unknown relationships
Trying to predict future outcomes
Hear no lies
Question and answer type research (survey research) relies on reported behaviour. Giving respondents the benefit of the doubt, these claims are usually more or less honest – unless they are deliberately aiming to deceive. However there is often a gap between what we think we do and what we actually do. Some classic examples include consumption of products like alcohol and tobacco, which are often understated, while usage of premium products is often overstated to boost perceived status or class.
Technology is increasingly providing methodologies to monitor behaviour, without having to ask questions. These typically provide the most accurate measures. No errors of memory, omission or commission. A real researchers dream come true!
Dashboard, a technology driven marketing intelligence business, has successfully piloted such a methodology in Zambia. We built a mobile app that some members of our opted in research community installed on their phones for a three month trial. This app recorded all their behaviour on their phones, and automatically uploaded this data a few times a day to our servers. Personal permission is the only way to get this data ethically. Community/panel based platforms are ideal.
We have been using big data techniques to analyse and report this data, the highlights of which will be discussed in part two tomorrow.
Want to continue this conversation on The Media Online platforms? Comment on Twitter @MediaTMO or on our Facebook page. Send us your suggestions, comments, contributions or tip-offs via e-mail to firstname.lastname@example.org.