[column] Creative gold digging with data science

It was a year ago that key note speaker Felienne Hermans (Associate Professor Leiden University) urged the audience to learn programming at the Market Insights Event. It would be so handy, have so much to offer. For everyone.
These sounds could be heard from several angles. You’ve been seeing graduates for years wiz kids starting small businesses in the corner of data science.
I left it next to me for a while. Data science felt especially useful for companies that want to make certain processes more efficient, automate boring people work, or present personalized preferences online. For the development of product innovations, service solutions or packaging designs, I mainly saw people and their creativity as the most important ingredient.

It’s a bit of a shock: data science has ended up in the creative field. Look at AI solutions such as DALL·E, which generates various creative design solutions based on complex texts. Bizarre. For example, you give a (crazy) command in which you indicate what you want, for example ‘an illustration of a baby daikon radish in a tutu walking a dog‘. The result? Truly an illustration of just that. And I don’t mean results from Google Search, but from newly generated solutions. Or the text input ‘an arm-chair in the shape of an avocado‘. tada. See there different designs of chairs in the shape of an avocado. Self-invented. By the machine.

Images drawn by DALL E for the caption “an arm-chair in the shape of an avocado” and “A baby daikon radish in a tutu walking a dog”.

But also on the online IIEX Event Europe (April 2021) you will see a strong return to the application of data science. Behaviorally (consultancy for shopper growth) presented a solution where concept packaging designs are evaluated based on AI. No packaging test among consumers, but with the proverbial push of a button, the appeal and advantages and disadvantages of your packaging are made clear, among other things.

Men against machine
Netflix, Amazon, Facebook and Spotify have been showing us what you want to see for years. So the step towards designing appealing creative expressions, such as advertisements, products and packaging, seems small.
For example, will we soon be able to develop award-winning advertisements in seconds that take into account brand guidelines, vision, strategy and target group of a brand? And then we still need creatives – no offense – or will it be the case that creatives are mainly inspired by AI-driven solutions, which are presented to them in a few seconds? I can already see it in my mind: ‘Er, let me think… I’m typing in: new campaign Staatsloterij End-of-year draw, but not with a dog, hedgehog or kitten, but with the same likebality’. And voila… a series of proposals.’ Bring on that Effie!
Still I don’t see AI solutions being used en masse to come until creative solutions. You can see that AI techniques are being used by brands in creative expressions, such as in the JBL Find your fit campaign from GoSpooky from December 2020 to support the launch of the True Wireless earbuds. Through deep fake AI technology you can easily put your own face in one of the main characters in the ad. “Funny!” I thought, along with over 200,000 people who also tried it.

Get ready!
As an insight-driven – or should I say data-driven – proposition designer scrape or crawl me for half a year thick data. Fully automated I collect data from written texts; this can be reviews, blogs, websites, but also other text documents. This multitude of data is then analyzed semi-automatically to arrive at useful insights into target group needs, innovation opportunities and (proposition) points for improvement.
I notice that I don’t want to hand over the reins completely to ‘the machine’, but I want to stay focused on the reliability and quality of the data that is being analysed. I still compare results with consumer survey results: Am I seeing the same thing? Am I learning more? What’s different?
The results are useful in outline, but the deeper insights and translation into (proposition) solutions will remain human work for the time being.
Yet! What a great development it is to have easy access to so much data! And what an opportunity to serve brands with this!
Every week there seem to be new, even more advanced, smarter solutions in this area. And as with any development, usability isn’t always quite there yet, but that too is changing. Collecting insights becomes child’s play, for everyone.

Are you ready for the data infusion?


Related Articles

Back to top button