How computer vision and deep learning boost sales and marketing

In this ‘customer is king’ era, companies are constantly searching for new ways to attract clients and get closer to them. Artificial intelligence can help them understand customer preferences and wants, personalize their offerings and target clients better. The AI technology that enables the most exciting customer engagement applications? That may well be ‘computer vision’!

Computer vision, which is basically synonymous to image or video analysis, has been around for a while. Yet by relying on machine learning, specifically deep learning-algorithms – which rely on artificial neural networks to learn – it now outperforms the human brain when it comes to extracting and classifying millions of images and recognizing patterns in images.

Popular applications abound. Computer vision, for example, enables Facebook to recognize your face in pictures and tag you and allows you to add a puppy face to your picture on Instagram or Snapchat. It’s also the technology behind Google Images: if you type ‘crazy marketing guy’ in the Google Images search bar, you’ll find relevant images in no time. Biometrics, self-driving cars and smart security cams also deploy computer vision.

Unlocking new capabilities

Now, how can computer vision technology help your sales and marketing teams? Here, too, possibilities abound:

  • Monitor social media: search social media for mentions of your products and services not only in text but also in images and videos, for more insights.
  • Simplify the online shopping experience: with visual search, shoppers will find your product more easily and save time. Consequently, they are less likely to abandon their baskets.
  • Boost sales: Visual search features like the Pinterest Lens (aka the Shazam for images) make finding and shopping based on a single image easier for your (potential) buyers.
  • Optimize the shopping experience and boost cross-selling opportunities: allow customers to browse, compare, and narrow their choices through image-generated similarities vs. manually attributed classifications. Customers can use shoes they like, for example, to find more options in similar styles. Or, if you suggest other clothing or accessories in the same style, they might be tempted to buy these, too.
  • Offer personalized experiences: gathering real-time visual customer data makes it easier for you to tailor experiences to individual customers.

Which image will be the most popular?

Just recently, one of our colleagues from delaware the Netherlands finished a master’s thesis on the topic of computer vision. His objective: to research and try out in practice whether, and how, machine learning techniques can help predict which images are more likely to be popular – and therefore more effective for marketing purposes.

Using a collection of about 150,000 Instagram posts and corresponding images, predictive models were trained to forecast the number of likes that a particular image will attract. The results are promising! Based on the findings, we have plans to develop an application to which customers can upload their own data sets – to gain insights into the most effective imaging for their online marketing activities.

It’s a visual world

The examples above are just the tip of the iceberg. With visual content poised to overtake text in importance, computer vision holds promise and is sure to unlock many capabilities for sales, marketing and all other company departments.

Interested? Get in touch with our experienced team to see how computer vision and deep learning can boost your marketing and sales initiatives.

Need some help identifying the best applications of Industry 4.0 technologies in your company? Get in touch with one of our experts.

Our expert

Wouter Labeeuw

Wouter Labeeuw

Wouter Labeeuw works as data science and machine learning consultant, adding intelligence to applications. He is a computer scientist with a PhD in engineering, where the research was focused on applying machine learning in the context of electrical demand response. In January 2016, he started at Delaware, still focusing on machine learning but in a broader context. Within his current role, he is responsible for the data science team within delaware.

Contact us