Thanks to technology it’s possible to understand media in new ways. And if you can pull the different strands together, says Frances Ralston-Good of Hearts & Science, you can develop new consumer insights.

Technology is increasingly making it possible to look in granular detail at every aspect of the media spectrum, from media buying schedules at one end to people viewing at the other and any subsequent response.

“The real sweet spot sits in the middle,” Ralston-Good told the recent IAB Engage conference. “When you bring those two things together and you’re looking at everything holistically, our belief is that media schedules and media placements actually become a way to understand consumers in a new way, to give you insight you may not have been able to benefit from before, insight you can use and deploy in other areas of your business, in other areas of your marketing.”

“In some ways,” she added, “media is potentially an insight playground.” But that’s only the case if, in addition to organising technology, the “behaviour around behaviours” of three groups are understood.


The behaviour challenge around new technologies is not technological, Ralston-Good stated. “It’s not about teaching people how to use technologies to extract insight from big data. It’s about the way we view data in the first instance.”

The Hans Rosling Perception Test, she explained, shows how people’s perceptions of data are shaped by context, including the political, the media, friends and family; and people tend to accept explanations that reinforce their own beliefs

Getting new recruits to the Hearts & Science agency to do the test is one way “to get our people open to challenging the norm”, she said. There’s nothing necessarily wrong with norms, she added. They’re a useful rule of thumb when considering things like which channel does what job best, “but the problem comes when they become rigid assumptions that undermine the ability to find a unique KPI”.

And it is possible, she maintained, to establish a unique KPI for every single advertiser and brand in the market.


Given those prejudices, it becomes necessary to approach data around consumer behaviour with a degree of caution if marketers are to truly understand what consumers are talking about. There are techniques that can be deployed “to uncover what consumers are potentially trying to tell us through their behaviour”, Ralston-Good suggested.

Example: a credit card company found people were taking up their product during a major lifestage change. But is that really the case, she asked. “Do lifestage segmentations give enough fidelity around how consumers are truly behaving around the product? We compared what they said to the client in research groups with what they did.”

One person spent some time looking at online fashion pages before moving on specifically to swimwear and from there to Yahoo to look for plastic surgery providers and then a credit card. “Getting granular and comparing segmentations to actual behaviours is a critical capability we can bring to bear when we’re planning,” said Ralston-Good.

In another example, 860GB of data across TV and online were analysed to reveal, for a performance client, that high-attention programmes in peak time brought high value response “This is not normal – but this is normal for this client,” she reported. “If you can analyse data at that scale – across multiple channels – you start to see some very interesting behaviours … and start to break down conventions.”

You can also “unpick programmatic”, she added: using a large dataset showed how context in programmatic drives up conversion rates and decreases costs per acquisition. “We’re not any longer talking about either/or, we’re bringing things together and creating a holistic picture in term of how media and advertising works for you as an advertiser.”


“Technology and data brings some tricky conversations onto the table,” she acknowledged, “but they’re ones that open up opportunity in the longer term.” She cited the example of one client organisation where brand marketers dealt with some channels and response marketers another.

“By bringing together all the media metrics into one space to create a complete picture, you start to see brand channels drive response, response channels drive brand.” And so there’s a short-term win from adjusting the channel mix – “we can have brand and response everywhere” – but the implications go much further. “Media is not owned by different departments any more.

“The common currency we can get to here is an understanding of consumer reaction to those different media stimulus.” The norms don’t help you to organise yourselves from the client side, she pointed out.

Finally, reporting and optimisation – “the most tedious repetitive issue in relationships between all parties in the digital space” - is part of a much bigger picture that includes “moving up the maturity ladder” from manual data manipulation to machine learning.

“Once you start introducing machines ... you’re not just counting how media is being delivered, you’re starting to see patterns ... we can accelerate up through this ladder to build a holistic picture.”

“If we can focus on the new tools and techniques, the bit that we really need to get to the next level is a new attitude.

“A media schedule is no longer just a means to reach an audience; a media schedule is something you can use to understand an audience – understand how an audience can relate to you as a brand and your broader business. It’s not just a distribution part any longer.”