Luke Brown, chief executive officer of data-driven agency Affinity, acknowledged that data can be difficult to work with, especially with sizeable data sets, and that one size doesn’t fit all when it comes to getting actionable data for different businesses.
“We subscribe to the Goldilocks theory of data: not too big, not too small, but just right. That’s because working with big volumes of data is cumbersome, it takes a long time to compute. There’s huge storage issues. There’s privacy and sometimes there’s security issues around the data itself,” he said at the recent Mumbrella360 conference in Sydney.
Brown advised marketers against spending huge amount of money buying data and tools before the foundations of a plan are in place.
“It’s not about the resources that you have in this space. It’s about being resourceful, planning better upfront, and making sure that you don’t go fishing with dynamite. To do that you need an effective data strategy plan,” he said. (For more on how to use data more effectively in marketing, read WARC’s report: Five steps to build a predictive data capacity.)
“We need to move beyond clicks, and shares, and impressions, and views, and get to something that’s meaningfully moving the business goal. Then you have behavioural goals, all of those things will deliver metrics, and out of those metrics will be things that are really important – we make those KPIs,” Brown explained.
“Once you understand how you’re going to measure all the actions that are important for your business goals, you can then go and get tools… In this space, we see a lot of interesting facts but very little that you can tangibly action as a result. What you need to get to is actionable insights and what we call ‘operationalised predictions’.”
Understanding the role of data in achieving a business-specific goal is a useful approach, he added, and isolating a consumer pain point and using data to help solve it has been one path to success. Marketers should constantly be asking questions, he said.
“Asking all of those questions – asking why, why, why like a toddler – until you can’t ask that anymore. Then you get to a really great place where you’re prepared to work with data at scale, and you know what you need to do from there,” Brown advised.
Sourced from WARC