Leveraging Big Data from Florida and Ontario

Ongoing research has taught us that improved taste is a key driver in getting consumers to purchase more fresh fruit and vegetables. Unlike apples or potatoes, sweet corn is an item that really benefits from a reduction in travel time from field to consumer. Optimal eating quality lasts up to fourteen days from harvest, provided it has been stored correctly.

A key grower challenge has always been matching production with demand. Predicting demand is very complex, driven by a number of factors including weather, price point, ad/display placement, and more. Optimizing production can be equally complex. Given all of this, how do you build the optimal supply chain to satisfy consumer desire for better tasting sweet corn?

Let’s start with a Florida grower trying to meet demand for sweet corn in Toronto (for more information about suppliers and retailers in Toronto and Ontario, see our supplement). His goal is to sync production with demand. In Toronto, the retailers are making merchandising decisions related to sweet corn while the grower is planting his crop. Both sides are working independently and while both are intent on improving sales, neither is optimizing the use of data to improve results. Enter Big Data!

Big Data, discussed in our feature article, starts with the consumer. Each retailer will craft a strategy to grow the corn category. With Big Data, they can craft micro-strategies by consumer segment with targeted promotions in the extremely diverse Toronto market. A more varied consumer segment makes predicting demand more difficult as each segment acts and reacts to price and promotion in a different way.

Retailers can couple this information with online data measuring the frequency of sweet corn recipes in social media, and even track Toronto’s weather, as experienced merchants know that outside temperatures can impact consumption. Retailers can also track competitive activity on sweet corn as it is often a featured item during key summer selling periods. Finally, by blending or synthesizing all this information with their basic merchandising decisions on price, placement, and promotion, retailers can predict demand better than ever.

Down in Florida, the grower can use this forecast data to build a production plan that creates supply just in time for demand. The grower can integrate data by field, crop input, and predicted weather patterns. For instance, he/she may find that one field can produce sweet corn in 78 days while another may need 81 days. All the while, the grower can implement sustainable measures by micro-targeting inputs like water, pesticide, and fertilizer to avoid over use. In the event of an unexpected weather pattern, the grower can analyze the data to determine its effect on harvest date. Imagine the impact of sharing this information with the retailer and the trucking company. “Sweet corn will be three days early—adjust your marketing strategies to assure your shoppers get great flavor, and shift your trucks to load earlier.”