Leveraging 2D image recognition technology in retail can lead to increased sales potential
Developments in artificial intelligence have led to a greater understanding of the potential of image recognition technology. Industry-wide, businesses are seeing the benefits this type of tech creates from supply chain management to brand growth strategies. While it’s still in its infancy, the potential impact of image recognition on the entire consumer buying process is limitless.
Image recognition shows strong potential to increase sales both in the shopping aisle and online. It offers a way for brands to connect with consumers through their phones’ cameras. It’s also a means of reducing the length of the purchase funnel to increase sales rates. Image recognition technology will become critical to marketers in the years to come as this type of innovation has the ability to impact an entire industry.
The Role of AI in Image Recognition
Image recognition processes run on data. A system must be able to review the various dimensions of an image, and then parse through millions of other photos to find a match. This technology must also understand the context in these photos; or in other words, why the consumer is searching. Artificial intelligence is a critical part of this process.
AI algorithms make it possible to review millions of images in an instant and select the most fitting result. This search process is evolving in the same way that search engines did. Before Google, consumers used web crawlers and other platforms to search based on keywords. However, these platforms worked strictly off flawed keyword systems. Posts with the highest count of a specific keyword were shown first—often with a heavy bias towards that individual crawlers’ sponsors.
Then, Google rolled out an algorithm that was designed not just to look for keywords, but to seek out quality. Using data from millions of existing searches, they found a way to give these keywords context and better understand not just what the consumers were searching for, but why they searched in the first place.
Image search is evolving similarly. In the earliest phase, image recognition was limited to finding only 100% identical pictures based strictly on pixels and size. Now, image searches have grown more sophisticated. For example, an individual can first submit an image taken with their phone. Next, the algorithm will seek out individual items found within that image and return with useful and informative product information on them.
Establishing Image Recognition as an Online Shopping Tool
Limiting purchase steps reduces cart abandonment. Amazon realized this when developing their “1-Click” ordering option, which proved very successful. Through this process, they were able to overcome common issues that lead to cart abandonment before purchase.
Excessive advertisements:
Pop up and banner ads cause 62% of consumers to have a lower opinion of a retailer as they believe this means the customer experience is not a priority. Consumers find ads that interrupt their purchases frustrating and unnecessary, leading to a common cause of cart abandonment.
Excessive form fills:
Having to fill out redundant information—whether it’s a matter of excess forms or system glitches—will cause 63% of consumers to abandon their carts when shopping via mobile. Requiring customers to fill out excessive information without offering an auto-fill option will cause them to search for products elsewhere.
Difficult navigation:
Site navigation remains one of the top causes of cart abandonment. Consumers can’t buy what they can’t find, and will leave a website quickly if they’re unable to easily navigate the site and select items freely.
These three reasons are the primary factors in why 70% of carts are abandoned before purchase, so retailers and brands need to consider ways to work around them; image recognition technology can help.
Snapchat shows us how image recognition can help overcome hurdles in online shopping. The image-based messaging platform recently partnered with Amazon to provide a more seamless shopping experience. Users of the platform can first take photos with their smartphones, and then use these photos to find similar items on Amazon. This process eliminates the conventional barriers in regular online shopping by factoring in three crucial steps.
Optimizing for mobile:
While Amazon makes use of banners and other sidebar advertisements, the mobile app streamlines the page to focus on the item the consumer searched for while minimizing ad space. Mobile optimization ensures that the product the consumer wants to find is front-and-center.
Auto-filling info:
Consumers can link their Snapchat to their Amazon account, meaning they will have all their information pre-filled and can quickly check-out via 1-Click or standard options.
Simplifying navigation:
Navigation is streamlined by putting the search in the consumer’s hands. All an individual must do is upload a single photo, and the system will find several matches that they can quickly browse and purchase.
Creating Better In-Store Experiences Through Camera-Based Mobile Apps
Consumers love taking pictures. Between Instagram and Facebook, they upload an estimated 445 million photos every single day. As consumers have shown their willingness to use their phone’s cameras, brands should seek out ways to leverage this. Image recognition technology can provide a medium for brands and retailers to deliver helpful and engaging experiences while consumers are in the store. The use of image recognition technology allows consumers to gain information about products, in addition to other incentives which may include:
Rewards points:
Shopkick offers kicks, or rewards points, to the consumers that use the app. By using their smartphones to scan and interact with items in the store, Shopkick users can collect rewards points and later redeem them for gift cards. This strategy acts as a way to engage consumers in the store and direct them to specific products.
Product details:
The ability to take an image of a product and gain information about it could be particularly useful for patrons with disabilities, especially those with mobility issues. These consumers could use a mobile app image recognition system to view product information up close, even when they can’t reach it. In this use, the product image would replace the barcode, making the label itself that product’s unique identifier that unlocks information like ingredients, prices, and expiration dates.
Price comparisons:
Using an image recognition program, consumers could take a photo of a product in the store and then gain information on similar products, as well as the costs and availability for that location and others.
Customer reviews:
Consumers trying something new may want to know how other consumers have enjoyed the product. Image recognition software can help them to locate customer reviews online, and to ultimately determine if that product is right for them.
Deals and events:
It’s possible to create deals and events around the image recognition technology consumers use in the store. A brand wanting to introduce a new product, for example, may notify consumers of a coupon or offer on that product when they upload a photo of it. This gamified experience can increase engagement and new product awareness, which will help improve its launch success potential.
The Growing Role of Image Recognition Technology in Retail and CPG
Often, retail marketing technology evolves from technology used to improve production and supply chain management. This was the case with barcodes, which were used for faster inventory management and customer checkout. However, over the years, barcodes expanded into the marketing category, as consumers began using them to participate in rewards programs or gain information in the store.
Image recognition technology is seeing the same evolution now. One of its earliest practical uses in retail occurred before products even hit store shelves. Visual pattern recognition, the precursor to image recognition, made it possible to quickly review massive amounts of freight for both quality and quantity. The system could also make recommendations to correct issues, essentially becoming quality control hubs in the shipping process. This type of technology has also been tested on store shelves to allow retailers to track products and determine if automatic reorders are necessary.
More innovations in image recognition technology are expected in the years to come, as many of these applications become consumer-facing. One unique use involved combining image recognition with other innovative systems to create checkout-free shopping. Amazon rolled out this type of arrangement with the Amazon Go store concept.
These convenience stores allow customers to select items off store shelves, place them in their cart, and simply leave when they’re finished shopping. The consumers pay through their Amazon app, meaning that they never have to wait in line or even take out their wallets. All of their purchases are tracked and managed through image recognition technology, artificial intelligence, and smart sensors. Amazon has had so much success with this concept that it’s reported they plan to open up to 3,000 stores by 2021.
Partnering for the Future of Image Recognition
Image recognition requires a lot of data which is why most brands and retailers use strategic partnerships to leverage it. These partnerships support everything from massive supply chain management programs to smaller-scale mobile app marketing campaigns. Brands will want to take advantage of this new technology as it continues to evolve, so this is the time to start seeking out partnerships. Third-party mobile apps can be an excellent place to begin testing this type of campaign, as they are scalable, easy to implement, and have existing audiences.
Creating proprietary apps around image recognition technology isn’t quite as easy as partnering, but this will change as technology improves. Already, several social media sites have developed tools for use with image recognition.
Image recognition technology is not just a new aspect of the retail industry. It’s poised to impact every segment of it, from supply chain to consumer purchase. The sales potential of this technology is almost infinite, as the consumer shopping experience both online and in the shopping aisle is simplified and improved. Third-party partnerships can offer an opportunity for brands who want to prepare for the future of shopping with image recognition.
