A new website that promises to turn online recipes into supermarket shopping lists and online food orders is set to launch this autumn.
Nick Holzherr, a runner-up in BBC television series The Apprentice, is behind Whisk.co.uk. He aims to have all the UK supermarkets that trade online – Asda, Tesco, Sainsbury’s, Ocado and Waitrose – operational on the site by the end of the year.
At soft launch, set to be in about two months time, the site will feature two or three supermarkets.
Users of the site, which has already won £170,000 in investment from two institutional investors and four angels, can add recipes from any website to a single virtual basket. Using semantic technology, Whisk will work out what store items they need to buy to cook the recipe and give them a suggested list of shopping items to buy, which the user can amend before completing the purchase on the supermarket’s website.
“Because we use sematic technology we can do it on the fly, analyse lists of ingredients and make it into purchasers’ lists,” Holzherr told Internet Retailing.
“Our software analyses static, unstructured recipe content online, pulls it apart and breaks it into structured data. We then link every single recipe ingredient to all the items there are in the store. We update every day against availability at the supermarkets. For example, we would break down a line of text that read: 150g of lean steak, chopped and grilled to: this is steak. It’s lean. It will be chopped and grilled so you need certain type of steak and you’re going to need 150g.
“Then we look at the user, and based on their age, gender and past purchases, we give them a price score that reflects the type of items they are looking for, and we select an item from the supermarket for them.”
The shopping list may include a limited number of sponsored products placed by brands who have a relationship with the site. Whisk’s earnings will come both from these relationships and from a commission on supermarket grocery orders placed through the site, which will be free to users.
The site is aimed primarily with women with families aged between 25 and 44 and will learn more about users as they continue to shop through it. “Our system then has quite a lot of artificial intelligence built into it,” said Holzherr. “It machine learns how user will interact with the ingredients we recommend and what the user wants. As you use our tool it becomes more and more personalised and more suitable for you. We would hope to have a good match already through information we can pull down through Facebook and Open Graph, Twitter, and use to tailor store items to the person.”
Nick Holzherr will be explaining the semantic technology in more detail at SemTechBiz, the semantic technology and business conference, which is being held in London on September 19.