Retail in Asia

In Shops

Expert Opinion: Why image recognition is a game changer in retail

Expert Opinion: Why image recognition is a game changer in retail
Joel Bar-El, CEO of Trax Image Recognition

English philosopher, statesman and scientist Sir Francis Bacon first said, “knowledge itself is power”. Even though it was said hundreds of years ago, time has only strengthened the claim. According to analyst firm IDC, from 2013 to 2020, the digital universe will grow by a factor of 10 — from 4.4 trillion gigabytes to 44 trillion. It will double in size every two years.

Much of this growth can be attributed to the Internet of Things, whereby objects are brought online as digital assets. Let me give you an example. If we take a can of Coca-Cola, it is a drink on the face of it; there is no technological wizardry. Scratch below the surface however, and you see that this consumer good has a large digital fingerprint. Business Insider says of the 55 billion servings of all kinds of beverages consumed each day (other than water), 1.7 billion are Coca-Cola trademarked/licensed drinks.

When you start to ask questions about size, flavor, geographies, demographics, temperature, and so on, you start to understand how much intelligence resides in the supply chain of this beverage. What makes a brand more or less successful comes from its ability to turn all that information into actionable knowledge. So how do we do this?

Information means money

It stands to reason that if you desire more knowledge, one would follow the larger sources of market information. These enormous pools are known as Big Data and they a can be obtained in various methods.

One of the most significant methods is tapping into people and their path to purchase decisions. It is believed that 80 percent of the consumer’s purchase decisions are made while in front of the shelf. Measuring the shelf execution standards, and linking it to people’s purchase patterns, can generate a wealth of Big Data information.

While purchase patterns can be obtained from the cash registers in relative ease, measuring the shelving execution standards is a much more complicated task. While the retail market is huge, the industry’s ability to grow and compete has been challenged by infrastructure, increased competition, and most importantly, the absence of effective tracking and analysis tools of retail execution in the stores.

In today’s highly competitive retail landscape, being able to control and optimize retail execution at the point of sale has never been more critical. However, only a handful of the traditionally conservative FMCG players have embraced progressive technology to achieve this.

Industry audit solutions today are manual and time-consuming. Auditing takes approximately 15 minutes per product category, involving physical measurements that are prone to human errors and inaccuracies. It is an expensive solution — companies can spend as much as USD12 million annually in a single market to employ a sizeable salesforce to undertake these audits.

Despite the massive efforts and costs, manufacturers can only be content with basic statistics and KPIs, and aggregated reports that take weeks or months to produce. Such limited reporting not only hampers timely and effective response to consumers’ demand and preferences, it prevents manufacturers from making more intelligent, accurate and profitable business decisions in a highly competitive retail landscape.

Image recognition: A game changer for manufacturers

With image recognition technology, manufacturers and retailers can now understand the marketplace and react in real-time. Using images that are taken in store by the manufacturer’s own sales reps, or by consumers (in a crowed sourcing model) are turned into actionable intelligence in real time. It turns sales execution into science. Using image recognition can result in up to 60 percent of audit time saved in stores and can present a 98 percent auditing accuracy — this more often than not a 20 percent accuracy improvement from manual auditing accuracy level.

This means you will have accurate and reliable data on your distribution, know which items are out of stock and the share of your products within the category as well as a wealth of other actionable insights at your fingertips.

Images can tell a story that no manual process can obtain, like shelving compliance and detailed calculation of your product performance vs. your competition. This innovation matters because it takes human error and processing scalability out of the equation. When you are selling billions of products around the world, even a small percentage of variance has a significant impact.

However, with image recognition for retail, it is not only about auditing accuracy or saving time, it is the powerful insights that can be gained in-store. Today, a sales rep only considers four or five elements when doing in-store reviews; but by leveraging image recognition, a sales rep can get 50 different measurements like share of shelf (market share), planogram compliance, pricing, competitive insights and more, just from a few images of the shelf.

In the consumer goods world, most leading brands have limited opportunities for horizontal growth in most markets, thus the emphasis has turned to vertical growth. Leveraging such data, a sales rep can identify all the category opportunities available for the store and due to the real time processing, can do it immediately with automated guided assistance (example of category opportunity is where a competitor product in a particular category is available in an outlet, where the equivalent product for a brand is not).

These in turn, become extremely valuable opportunities for sales reps to upsell, cross-sell and provide range extensions in-store. In some markets, companies like Coca Cola have successfully leveraged the data they have gained through the use of image recognition, they have seen 3 percent gains in market share, better performance of their brands in-store and a positive impact on their bottom line.

Beyond basic recognition technology

Today, image recognition technology is the output from machine learning and predictive analytics. In the retail world, the fine-grained image recognition engine, analyses the image or video capture of any product on the shelf and is able to differentiate between the minute design changes in brands, sub-brands and stock-keeping units (SKUs). For example, think of all the different Coca-Cola bottles that were created for the World Cup. If there is a new product introduced into the market, the technology updates its SKU database automatically, requiring no manual updates from the sales reps.

The innovation does not end there — now manufacturers can look forward to an accurate and reliable audit solution without requiring any data connectivity in the store. Offline capabilities of image recognition on all smartphones, tablets and wearable electronics are fast becoming a reality. For emerging markets and traditional trade channels (mum-and-pop stores) where Internet connectivity is not always available, offline image recognition provides a powerful competitive advantage for any brand.

For consumers, wearable electronics allow them to engage directly with manufacturers at the shelf, make informed purchase decisions and receive targeted sales promotions. The technology is also leveraging video capabilities, which is better suited for modern trade channels (i.e. big supermarkets with long aisles) as it does not require sales reps to take as many images and they can simply scan the shelf.

Analyst firm Gartner sums it up well, “Having visibility into conditions at the retail shelf is not new, but using digital image recognition and advanced analytics to spot opportunities and quickly address them is what makes the evolution from taking notes and measurements by hand at the shelf to capturing images digitally a game changer.” (Gartner, “Image Recognition: The Intersection of Digital Business and Analytics at the Store Shelf” Dale Hagermayer, 12 Aug 2014).

Despite a slow adoption, many leading manufacturers have begun to recognize the benefits of image recognition and are aggressively pursuing the technology to allow for greater efficiencies in their audit and execution processes and drive more intelligent and profitable business decisions as a brand.