From the avocado boom to a mass craving for kale, culinary trends seem to appear overnight. But how do retailers and producers know what people want?
Can you think back far enough to when an avocado was a sort of exotic rarity, rather than something everyone had with their toast for breakfast? Can you remember a world before kombucha?
To the average consumer, the latest food trends seemingly arrive from nowhere. And when they do, those that stick around are everywhere, from corner cafes to supermarket shelves. Which all begs the question, how does this happen?
Food and beverage is a trillion-pound industry. So, being able to predict these changes and understand consumers’ developing needs and desires can lead to big business.
Increasingly, companies in this sector are investing in technology, data science and analytics, hoping to gain that all-important glimpse of the future and predict the next big thing.
Here are four ways data is being used to find (and provide) tomorrow’s avocado on toast.
Sentiment analysis, natural language processing, and computer vision
It’s perhaps human nature to always look for the next big thing. Whether it’s driven by a new health fad, or ethical considerations, new trends emerge in our eating and drinking habits with increasing regularity. This means keeping up with them can be hard, let alone predicting them.
To this end, retailers and food producers have been using data analysis and AI to predict future trends for a while now.
Companies like Tastewise specialise in this area, providing almost real-time industry insights and predictions. It uses the latest techniques, like machine learning, predictive analytics, computer vision, and natural language processing to automatically mine data from social media posts, restaurant menus, and online photographs and recipes. That data is then used to track trends before they break.
Similarly, companies like Singapore’s Ai Palette are using AI and machine learning to determine the popularity of certain flavours in different markets. And, importantly, to predict how long these trends may last so its clients can make smarter investments.
The company’s AI also uses named-entity recognition (NER) to identify new terms or words that are being used with increasing frequency within its data pool – meaning it can spot the trends that are so close to emerging you can almost taste them.
Taking tips from the pros
Long before they arrive on our supermarket shelves, most new food trends begin in restaurants. And the majority take on a journey that’s relatively easy to predict. At least, this is the theory behind the work at Datassential, a company that has coined the term The Menu Adoption Cycle.
Datassential claims there are four stages of the Menu Adoption Cycle, ranging from the initial inception of a new ingredient or trend, through to adoption, proliferation and, finally, complete ubiquity. The company uses penetration analysis to map these trends as they emerge and grow in restaurants.
For example, it was able to track the growing popularity of kale from 2009 to its mainstream appeal by 2013.
Monitoring health trends
Perhaps the best way to predict future trends is to look at societal drivers. It’s not just restaurants that decide what we eat. As we understand more and more about how the food we consume impacts our overall health, the choices we make as a society have begun to change.
In fact, according to Forbes, 54% of all consumers, and 63% of those who are 50+, care more about the healthfulness of their food and beverage choices in 2020 than they did in 2010.
Actively monitoring these trends and concerns can help us predict where the industry will move. Whether it’s a new, celebrity-endorsed diet, an increased emphasis on gut health or a reduction in sugar intake, these societal shifts can help us ascertain what products are likely to fly off the shelves in the near future.
The tricky part is knowing how long these trends will last. Seemingly unrelated global events, like the COVID-19 pandemic, can have an impact on how people view the importance of health. But will that impact last the test of time, or will it rescind when the status quo resumes?
The only way to know for sure is to continue to collect, combine and analyse both past and present data, and use tools like AI and machine learning to help us identify behavioural patterns and how they impact demand.
The ultimate in personalisation
One interesting aspect of health and nutrition is that it isn’t a one-size-fits-all affair. As data and analytics continue to unravel the mysteries of the human body, we learn more about what’s best for us not just as a species, but as individuals, too. For this reason, one trend that has been emerging over recent years is tailored dining. And in this instance, data isn’t so much predicting what people eat as it is telling them what they should.
Research from SevenRooms tells us that more than half of all Brits would like the option of a fully personalised menu tailored to their specific needs, likes and dietary requirements. And 48% would be willing to hand over their data to restaurants to see that happen.
One example of this in action comes from high-street mainstay YO! Sushi, who a few years ago launched the UK’s first DNA-based dining service. After receiving a customer DNA sample, Yo! crunches the data, carrying our gene analysis to ascertain which foods people should be eating more or less of – and recommends dishes accordingly. It’s hard to imagine anything much more personalised than that.
Want a taste of tomorrow?
At The Smart Cube, we work with some of the world’s biggest brands, using a combination of human and artificial intelligence to predict market developments and drive better decision-making. If you are interested in how our approach to consumer and market insights, using the latest data-driven techniques, can help you succeed, get in touch.