There’s a data backlash going on.
Sometimes it’s expressed quietly, but in rooms full of influence, it’s often loud. And the complaint is always the same: data imposes rules on advertising and it stifles the imagination.
This is true, but the answer isn’t to throw data away. It’s to evolve the way data is being used.
Data shouldn’t be used to dictate, instruct, or generate ideas. Coming up with good ideas is a job for people.
Instead, data should be used to convert ideas into action. To demonstrate why that idea is widely relevant, to show what it can do, and to get people to act on it.
That’s the superpower of good data, and advertising needs it now more than ever.
Data causes problems if it’s used to produce ideas
The data haters are right when they say that data can’t imagine a better future, it can only give you a description of what’s happened in the past. And they’re right to say that if competitors have the same description of the same past there’s no advantage available.
To see why, consider ads produced by AI. What AIs do is observe ads from the past, analyse data on what worked well, and regurgitate and combine the best bits into new forms. It’s the supercharged version of coming up with ideas using effectiveness data and analysis.
The emerging evidence is that this works in the first instance. AI ads outperform those made by people. For example, the chart below shows brand health scores for car ads generated by data and AI in blue versus people in pink. On average, AI ads deliver a higher brand health score and a higher probability that you won’t air a dud.
The authors of the study are positive about the role of AI in creating advertising, but even they recognise that the algorithm has to be trained on something. So far, that is human-made advertising that data showed was effective.
But can AI continue to be more effective than people?
Well, if everyone’s AI creates ads based on the same past set of good ones, new ads will all start looking the same.
Next, AIs will start learning from their own past regurgitations of the original set of ideas. As the stock of human-made new ideas diminishes, people who see advertising will get very bored, and it will all stop working.
It’s a prediction that John Hegarty made well before generative AI came along. It remains to be seen whether progress in AI can prove him wrong, but it’s quite possible that in 2024, he’s even more right.
Dreams + Data = Action
But to argue data has no place in marketing and advertising is a mistake. Not just because saying things like that in public makes you look dusty, but also because data, when paired with ideas, is powerful.
Data can and should be used to test the strength of ideas and to find out whether discoveries apply widely.
And, with those tests passed, data adds a solid, reliable, and secure quality to an idea. It can be deployed to make people feel safe to act.
The chart below is big picture evidence that a good idea paired with good data can change the world. It’s the measure in the IPA databank of short-termism, the % of all entrants who need their advertising to pay back within 6 months.
This chart shows the power of a good idea paired with good data because, in 2013, Les Binet and Peter Field published their data analysis that showed too much short-termism is a problem.
In The Long and The Short of It, they presented their data well, with a killer stat “spend 60% on brand” and a killer chart to boot. And it worked.
At the time of publication, the chart shows a turning point, as the proportion with short-term horizons stopped increasing and, over time, slowly started to fall. So, exactly the thing that this data analysis advised started happening.
This is a big picture story, but similar things are happening in smaller ways within businesses every day. People who have a good idea and know how to support it with a killer chart or stat are getting their good initiatives off the ground.
Like the big financial services business that is using data to help rebalance their media budget away from spending a small fortune on search. The data reveals it’s a good idea, and it also helps their performance marketing team see that they’ll benefit and not be undermined.
Or the NGO that uses data on emissions associated with spending on advertising to help agencies decide which clients to pitch for if they want to be green.
The data makes it easier to do the right thing because the costs and benefits are known.
Just data it
In summary, the data haters are right and they’re wrong.
They’re right that data isn’t the best starting point for coming up with effective advertising messages, copy, images, and films. But they’re wrong to say that data is ruining our industry.
It’s hard for organisations to act. Investing in marketing, just like buying a big new IT system, is fraught. Committee decision making is sluggish, and nearly 77% of people say coming to a decision is very difficult, according to research from Gartner.
Just like a couch potato, it’s really easy for the organisation’s inner athlete to stay on the inside.
And just like a certain famous tagline and a certain brand of new workout gear, data is the thing that helps the organisation to get moving.
Wherever they come from, if you want your ideas to be credible, convincing, and compelling you need data.
Grace Kite is the founder of Magic Numbers which provides practical training and people-friendly analytics for marketers.