Data is your best friend, unless you make it your worst enemy.
Kira Clifton - VP, Media Strategy & Analytics
Performance Marketing: Rethinking your approach to driving sales
As digital commerce rises, it’s more crucial than ever that marketers start to rethink their approach to driving sales. While there has been a rise in spend on performance-based tactics close to the point of purchase, the need to balance brand-building and performance marketing for digital sales is absolutely essential.
To successfully reframe the marketing discipline for a world of digital commerce, we must utilize data in a way that works – through a filter of human curiosity.
The dangers of simply focusing on big data while neglecting thick data.
As the head of media strategy and analytics at Hart, I'm clearly a huge fan of data. It’s accessible in troves, measurable and frequently offers us insights into what’s going on below the surface. It’s a powerful tool in bridging that connection between brands and their changing customer. However, if we’re only looking at the numbers, and failing to pay attention to the questions the numbers present, we are missing out on a huge piece of the story.
When I first heard the term “quantification bias,” it came from Tricia Wang, Harvard fellow, co-founder of Last Mile and all-around skilled tech ethnographer who’s an expert in making sense out of data. During her 2017 TEDx presentation titled, “The Cost of Missing Something,” Wang explained the dangers of relying only on big data, which is the quantifiable, massive amounts of data, and ignoring thick data, the qualitative, emotion and context-driven data that’s unearthed by a smaller sample size of human experiences, stories, patterns and depth.
Wang warns that many times we’re simply focusing on big data while neglecting thick data, just because thick data isn’t quantifiable. This hit me hard, and articulated my caution of data all too well.
Following data can lead you right off a metaphorical cliff.
Let me give you an analogy. When GPS systems first emerged commercially in vehicles in the late 1990s to early 2000s, there were a striking number of stories in which drivers would literally trust their GPS off a cliff. Meaning, they would blindly follow the technology’s directions, even if it meant ignoring their own common sense, like swerving into oncoming traffic or even off the road. While perhaps not as dangerous, when we only focus on the patterns big data offers, rather than the questions within its gaps, we’re following data right off a metaphorical cliff.
As a marketer, you may be familiar with the rise and then demise of 3rd-party cookies. This is a prime example of how the industry followed data off a cliff. Cookies were originally developed to improve customer experience (CX) in the digital space, like remembering login information and preferences (1st-party data). Then, we quickly realized the data collected across sites (shifting from 1st-party data to 3rd-party cookies) could be used to identify and develop scalable customer target segments, as well as increase the ability to measure attribution (the impact of each touch point - channel and tactic). For this reason, the technology quickly gained popularity. So quickly in fact, we failed to use our common sense to question, 1) the quality of the data being collected and, 2) the consumer privacy concerns collecting this data presented, until it was too late.
I have another example for you, from my own work. I was leading a project for a client who had nearly an 80 percent category share across the organization’s portfolio of brands. At a crucial point, message attribution results came back low, and the immediate reaction from the client in question was to halt the project. While the results were certainly concerning, the focus was solely on those low results and not on the larger questions the low results presented. So I simply asked, “Is this misattribution a negative, or does it signal the need to consider the halo effect of each of your brands over each other?”
By asking this question, rather than focusing only on the messaging attribution results, we were able to shift our focus to better understand the inter-relationship of all of the brands in that category. From there, we developed a category strategy from what had previously been brand-specific campaigns, something we may not have realized was needed if we had halted the project and failed to ask, why? It’s the deeper causation where most marketing data stops at correlation.
Data can be our best friend in advertising and marketing. It can increase effectiveness, drive business results and so much more. But, it WILL be our worst enemy if we follow it off a cliff. To avoid that, we have to look at it through a filter of human curiosity:
- Why is the data showing me this?
- What is the cause of this?
- What could this mean?
- What doesn’t this show me?
- What’s an alternative scenario we should consider?
If you’d like to implement human curiosity in your own marketing strategy, I’ve put together some specific opportunities and key milestones to help keep you off of the data “cliff.”