Madrid’s startup combining machine learning and crowdsourced market research data Promos has closed a $1 million financing round. The firm blends computer vision AI and crowdsourced shopping receipts to output real-time consumer purchasing behaviour reports.
Typically, market research firms use point of sale (POS) data from a select few retailers to form an overall impression of an entire marketplace. If the last 18 months have taught us anything, “Typical” is out the window. Consumers in the FMCG (Fast Moving Consumer Goods) category haven’t exactly had the option to buy one item from one location, often shopping for different goods from different stores. Offline when possible, but mostly online. This creates a consumer footprint that is all over the shop (literally), and increasingly difficult to measure.
Promos is looking to change this conundrum by leveraging crowdsourced market data and combining it with AI. But how do you gather this crowdsourced data you ask? By employing one of the classic tactics, give the people what they want: cashback offers and discounts. Give retailers and brands the opportunity to run promos within the app, and now you're really on to something.
Using computer vision technology, Promos processes the information provided via a receipt and spits out product-level sales data. Promos’ brand and retail clients then have immediate access to market data.
“Market research for the FMCG industry has failed to keep up with the pace of the consumer. We wanted to develop the type of solution that we longed for in our previous roles in the industry but didn’t exist,” comments co-founder Lina Beinortaitė. “Our mission is to help retailers and brands understand spending patterns and consumer behaviour from a side of the market that nobody else can provide but is imperative in an increasingly multichannel market.”
Promos’ $1 million funding was led by Change Ventures, with angel investors Bolt co-founder Martin Villig, and Andrius Šlimas, who co-founded and sold Oberlo to Shopify participating.
Would you like to write the first comment?
Login to post comments