Data Science & the food industry
If one would search images on Instagram by content, pictures of food would probably come up in one of the first spots. Food pictures are a perfect example of how artistic and functional aspects intersect and their frequency and popularity is directly fed by its relation to everyday life. As it is the nature of the human body, we all need nutrition – regardless of socioeconomic standard. Phrased from an investor’s point of view: the market is huge.
In the words of famous investor and entrepreneur Peter Thiel, however, it needs to be reiterated that just because there is a lot of restaurants and the market is obviously huge, doesn’t mean that it is an attractive market: competition is for losers. But, technology is a game changer. New and innovative, food-centric business models have been popping up with high regularity, heavily venture-backed. Time to look at some of the ways how data science helps to transform and inspiring the food industry, enabling entrepreneurs to escape competition.
Discovering new recipes
We all have our favorite restaurants, our favorite recipes for preparing a delicious meal at home. The variety of possibilities to combine ingredients is sheer endless. Add to that the multiple ways of how to prepare them. Within recent years, online recipe databases have been set-up, allowing for extensive analysis of the predominant ingredient combinations in various geographics. Combining ingredients with their associated flavor, researchers were actually able to determine in a statistically significant way, how certain cuisines share similarities. North American and Western European dishes have a tendency to be based on ingredients, which share various flavor compounds. This is opposed to East Asian and Southern European recipes, whose meals tend to avoid a lot of ingredients with similar flavor. Data crunching (or rather: recipe crunching) is finally allowing foodists and scientists to explore, what components actually drive taste and popularity. Based on an enhanced understanding, intelligent algorithms can help restaurants, chefs and food enthusiasts to explore new ingredient combinations. Smart support systems for the creative processes are an exciting result, which could increase variety in food offerings.
Innovation in Food Delivery
While supporting smart experimentation and diversification in recipes, data science is also driving a consolidation play, which is currently enacted by various players in the industry. Foodora, Deliveroo, Lieferando.de and others: there is an intensive wrangling, supported by heavy venture capital, for the dominance in the food delivery space. Establishing a central point of contact for food delivery requests, bundling the offerings of a large amount of restaurants while providing them with delivery services, thus enabling more revenue than through regular sit-in orders. How are those services possible without a large mark-up for end-consumers? Well, currently they are possible as VCs are primarily looking for growth, and profitability will only surge on the agenda at later stages.
But generally, online food ordering platforms gather a lot of information about ordering patterns and preferences. Using machine learning algorithms, this allows them to dispatch their drivers (on bicycle, most often) efficiently and provide them with maximum utilization. Outperforming others in convenience and time-to-delivery, while operating highly cost-efficient in the backend is their ‘recipe’ to bring innovation to the food sector. While their core hypothesis still have to be validated over time, the fact that data science enables their pitch for market dominance is intriguingly fascinating. As it gives us the convenience to watch that movie on a rainy Friday evening with a delicious beef ragu from our favorite restaurant.
Connecting taste and business decisions
The modern food company is a tech company,said Ooshma Garg, Founder at Gobble at the a16z podcast (podcast of infamous Andreessen Horowitz). While that is a rather bold statement, it is certainly true in such regard as tech and data science are the key levers to drive optimization in various aspects of the business. Gobble is a service, which provides primarily young families with fresh 10-minute dinner kits. Supplying more than 1000 customers on a regular basis with high volatility on meal choices, the right prediction regarding demand is essential to manage the procurement of supplies. In a way, the company is able to utilize both insights from ingredients’ flavor composure and also customer behavior, purchasing history and individual preferences to improve and customize their product offering towards the customer and also design production (preparation of meals) and procurement more streamlined and demand-driven. Connecting taste and flavour of food all the way to enterprise functions is an exciting application for data science and could pave the way for continuous innovation in food businesses.
So, will those be the cornerstones for entrepreneurs to “escape competition” and establish those monopolies in the food industry that Peter Thiel seeks? We’ll probably have to wait for a few more years to be able to assess real results, but as of for now entrepreneurs are uniquely positioned with the right “ingredient” to make it happen: data science. Defining the right areas to explore the potential of data science in order to drive innovation and transformation in the food sector will be the main challenge of the involved data scientists. And exciting to watch for all others, which can draw inspiration from this fast-moving sector.