43 ROUND TABLE Audience Analytics
GABRIEL BERGER: Quality every time! It’s better to have one high-quality data source than ten poor ones. BHAVESH VAGHELA: Quality always beats quantity. In the era of big data, too many businesses think amassing all the data you can is the way to go, but they just end up drowning in it, unable to use it to take effective action with customers. If your data is current, accurate and pertinent to your business, you don’t need that much to start gaining valuable insight that will enable meaningful predictions or actions. You can even, in fact, apply too much data, which can slow down and affect accuracy of predictive modelling. It’s about knowing what data is relevant for each stage in the customer journey. ANNA ZAIKINA: There is no one answer. Both factors are equally important. Quantity of data ensures consistency across metrics over time, bringing reliable results that can be used to influence core business decisions. The smaller the data sample, the lesser the chance that it will be representative of the population that is being investigated. Quality control must ensure that the data set has nothing in it – no outliers – that can skew the findings and lead to inaccurate conclusions. FEED: WHICH ISMORE IMPORTANT – QUANTITY OR QUALITY OF DATA? OR DOES QUANTITY EQUAL QUALITY? THE BETTER A BUSINESS UNDERSTANDS THE PURPOSE OF AND THE DESIRED RESPONSE TO ITS VIDEO CONTENT, THE CLEARER IT WILL BE TO FORMULATE THEIR KPIS
FEED: HAVEWE REACHED PEAK DATA? IS THERE ANYMORE USEFUL DATAWE CAN EXTRACT ABOUT VIEWER PREFERENCES AND BEHAVIOUR?
GABRIEL BERGER: I don’t think you can ever say that we will reach peak data. There will always be new and interesting ways to analyse data and to extract and refine insights to further boost the user experience. BHAVESH VAGHELA: You could argue that we’ve reached peak data, but I believe it’s more about using the data we have more efficiently than finding different sources. Machine learning and AI are starting to play a big role in making better use of the data we already have and promise to produce a new type of peak – creating more precise and applicable insights that can be acted on in real time. The new data that can come into play should be drawn from unexpected sources. TV service providers should now think outside the box and look at data not
related to user details and behaviour. For example, other industries look at weather data when tailoring offers to consumers – like a retail promotion on barbecues and garden furniture when sunny days are predicted for a summer long weekend. A pay-TV service could do a similar thing – for instance, pushing out an offer of a free movie package for kids when heavy rain is predicted during a school holiday. ANNA ZAIKINA: It’s not so much a question of reaching the data type limits, it’s more about how the data is used. This is where most of the work is yet to be done. Many organisations still struggle to find the right technological solutions, allocate human resources and build workflows to make sure the existing data is used to its full potential.
feedzinesocial feedmagazine.tv
Powered by FlippingBook