How AI and image recognition are transforming social media marketing
Contributor Rob Begg discusses the explosion of visual content and how artificial intelligence and image recognition are making it easier for marketers to identify visuals in social media for better metrics and customer service.
If a picture is worth a thousand words, social media users are speaking volumes. People now share more than 3.25 billion photos a day on the world’s biggest social platforms, including Facebook, Instagram and Snapchat. In 2012, that number was less than 500 million.
Data also shows that social media users gravitate toward visual content: Facebook users are 2.3 times more likely to engage with posts that have images than with those that don’t, and tweets with images receive 150 percent more retweets than those that are pure text.
It’s safe to say that social media is now primarily a visual medium, and marketers can’t afford to look the other way.
Tapping into the photo explosion
Driving this visual explosion are two main factors. First, starting about a decade ago, the rise of mobile devices put a camera in everyone’s pocket. Suddenly, consumers no longer needed a desktop device to log into their social media accounts, and very quickly they began to engage on social media platforms on the go and on the same phones or tablets they used to snap photos.
Second, image-based social networks such as Instagram and Snapchat emerged to cater to this new reality. We’ve also seen text-based social platforms, such as Twitter, respond to user behavior by offering more opportunities to put images front and center.
No doubt, marketers know the importance of social media marketing. Nearly 80 percent of social media managers (registration required) see a strong ROI from social, and companies with the most mature social media programs are more profitable and better at retaining customers than those who fall behind. Forward-thinking companies also realize how important images are within marketing content, with 74 percent of them using visuals (PDF) in social media marketing.
However, while brands are excelling at sharing compelling videos and photos that evoke an appealing lifestyle or backstory for their products, they haven’t figured out how to discover and respond to images that the public is sharing.
The visual web and the future of marketing
The photos people share on social media represent the consumer behaviors, wants and needs that often are undetected by marketers.
If a person posts a photo of a new product, but doesn’t include any text saying the product’s name, social media monitoring probably won’t capture it. As a result, companies miss out on big opportunities to learn about and communicate with their customers.
For example, someone might tweet a picture of an airplane with “#fail” or share a photo of sneakers on Facebook with the caption, “Where do I buy these?” Marketers who are only doing keyword searches in text would never be able to uncover these posts.
Until recently, the process of searching for images of a certain product or brand has been time-consuming and incomplete. Many social media marketing tools allow marketers to capture posts that mention a brand or product — some of which may include a photo. But that process is cumbersome. A marketer still has to go through each image manually to determine relevancy and draw insights.
Now, artificial intelligence and image recognition are making it easier for marketers to find visuals within social media, even when they’re not accompanied by an explicit text mention.
Today, we’re seeing the evolution of social media marketing, where companies can automate the discovery and identification of images across social media and respond appropriately.
Image recognition for better metrics and service
AI-based image recognition tools scour social media sites for photos and compare them to an extensive library of images, scenes and characteristics. Compared to a person manually reviewing photos, technology can uncover the right images at an unparalleled speed and scale.
For example, a company can use image recognition to quickly identify images that include the brand’s logo and determine whether an investment in product placement at a certain event paid off.
In another hypothetical use case, it would be a large investment for a beverage manufacturer to answer the question, “What is the most common place where people drink my iced tea?” Now, the company can easily comb social media posts for photos of people drinking that particular tea and figure out whether they were more likely to be at the beach or at a picnic.
Image recognition across social media also offers big benefits to companies when it comes to customer service. If an automaker has a recall for a defective, outwardly visible mechanism, image recognition technology has the power to instantaneously uncover cases through images on social. Once the photos of the defective mechanism are gathered, the marketing team has a range of options to best react to each user — one of which is to alert customer service.
Marketers have always invested where consumers spend their time. Being able to interpret what people are doing and saying on social media through photos is no exception. Companies have responded to the surge in images on social networks by creating their own visual content.
With the advent of AI and image recognition, they can go even further by listening to what people are saying through images and using those insights to deliver a better, more seamless and delightful consumer experience across all channels and touch points.
Opinions expressed in this article are those of the guest author and not necessarily MarTech. Staff authors are listed here.