Do you remember when data was considered the new oil? A precious new commodity with great value that was hard to extract and refine. Now data is as common as a pint of milk – and shares many of its characteristics.
How? Well, I’d argue that there is a bewildering array of it widely available, it’s arguably at its best when mixed with other ingredients and, most importantly, it has a tendency to go out of date really quickly.
As a publisher, Time Inc has always relied on data to inform the direction of our content strategies for our editorial properties and advertising partnerships. This isn’t a new thing; now though, we get more of the data, more quickly than we have ever had before.
“Using poor data with great content is just as great a sin as using cutting-edge technology to deliver lazy or clichéd content with boxes labelled ‘Content 1’ and ‘Content 2’ ”
Not everything that counts can be countedThe sheer amount of the stuff around means that you need to make some choices about what data you are going to listen to, and what you are going to leave in the background. As William Bruce Cameron once said, “Not everything that can be counted counts, and not everything that counts can be counted”.
Deciding what counts and what doesn’t is easy: you just need a clearly defined idea of what you want your campaign to do, how you are going to measure it and why that measurement means that the campaign is a success.
That means, however, that we need to rely less on nebulous phrases like “I want people to engage with this video”, and be as specific as “this video will be a success if it is shared by my target audience, as that means that they are using it to define their own identity. And that is a success because it means that they have taken the first step on the road to identifying with my brand”.
While data can tell you a lot of things, it can’t create a tone of voice. For that you need a team of specialists to work alongside your data analysts. At Time Inc our journalists, designers or videographers are every bit as important in the application of data insight to content campaigns as analysts. You still need human beings to create the stories, without that spark of inspiration, we will all just be optimising the same homogenised content in ever-decreasing circles.
Five ways to choose the right data to create the right content
1. Clearly define your objectivesand make sure you choose data sources to help shape your strategy accordingly
2. Keep your eye focusedon actual user behaviour and think how you build on it. Habits are easier to exploit than they are to break
3. Spend as much time thinkingabout the content as you do thinking about the data. If we just optimised according to existing behaviours, we would never create anything new
4. Iterate and don’t be afraidto make mistakes and correct them later
5. Use the freshest data that you can
When data and creative together makes something bigger and better
Putting analytics and creative in the same room can help generate real insight. A while ago we were working on a ‘Cooking with Kids’ proposition for a client. There were an awful lot of differing opinions between the stakeholders about choosing the best time to dial up the activity. Our behavioural data told us that Sunday morning was the time most people were looking at our children’s recipe content, something no one was expecting to see. This meant we could all view it objectively, rather than using data as a drunk uses a lamp post – for support, but not for illumination.
While we continued arguing about timings one of the analysts also pointed out that the most popular times for our audience to fill the internet with pictures of family activities were Saturday and Sunday. Suddenly we had a whole new distribution strategy based around “sharents” that wouldn’t have happened if we hadn’t had analysts, journalists and marketers in the same room.
Where data currently comes to the fore is in content distribution. It is a fantastic tool for optimisation and iteration. But, before we all get beguiled by the science of “programmatically targeted sequential content marketing opportunities”, we should remember that we are really talking about telling stories to people that are as relevant as possible to them. In this context, using poor quality data with great content is probably just as great a sin as using cutting edge technology to deliver lazy or clichéd content. Using shabby data sources could well have the same detrimental effect on your brand as just lifting the words from the programmatic team’s presentation and hitting your audience with boxes labelled “Content 1” “Content 2” and “Content 3” (in the right order of course).
At Time Inc we are lucky in as much as we have vast data sets of consumer behaviours across a wide range of verticals that are constantly refreshing themselves. Some of these are based upon web analytics, tracking and all the stuff that you can pour into a data management platform; others are based upon over a century of editorial experience and expertise. We can bring all of this to bear for our editorial, advertising and content partners, so pop in for a cup of tea if you want to chat more about it. And don’t worry: we have plenty of fresh milk.
“Where Data currently comes to the fore is in content distribution. it is a fantastic tool for optimisation and iteration”