Maximising brand engagement in a new era of in-depth insights

Innovative Object and Scene Detection technology enables the gathering of visual data which can be used to enhance brand engagement 

All marketers have a single, defining goal: make every dollar count! Wasted marketing spend is an ever-present challenge and one that keeps them up at night. In fact, according to a survey conducted in 2018 by Rakuten Advertising, marketers reported that they waste an estimated 26% of their marketing budget on the wrong channels or strategies. More worrying is that 50% of respondents noted that the most pressing issue was shifting customer expectations. Meanwhile, a survey by Hubspot showed that ‘proving the ROI of marketing activities’ was one of their biggest challenges, (43%), second only to generating leads (63%).

This clearly points to message relevance being a major concern for marketers, especially in a world where consumer sentiment can change in a heartbeat. Gaining a better understanding of how consumers engage with their brand and products is key to plugging that gap and increasing marketing relevance.

Visual-AI based brand detection technologies have provided a partial solution to this problem. It has revolutionised how brands can monitor key data at scale. Analysing billions of images to spot consumer brand engagement, moments of consumption and even emerging influencers are all now possible with a few clicks of a mouse. But brand marketers need more. They want data that allows them to see correlations and patterns that link their brand to objects and scenes in images.

When humans look at images and videos on say, social media, we can easily recognise various objects and setting of an image within seconds. However, to identify statistical significance requires you to check hundreds of thousands or even millions of images. This would take days or weeks for a human to trawl through. As the popular meme says…‘ain’t no one got time for dat!’

VISUA’s Object and Scene Detection can replicate the visual recognition abilities of humans, but in significantly higher volumes, completing those same colossal tasks in a matter of hours. This system also adds vital additional layers which can dramatically improve marketing relevance in future campaigns and activities. Most importantly, it can deliver a very fine level of granularity. In fact, it can detect between 11,000 categories of objects and scenes. They include things as diverse as types of food and animals (not just ‘dog’, but ‘labrador’ or ‘sausage dog’) and scenes such as living room, beach, mountains, etc.

The opportunities this data provides for marketers are vast, but for me, there are three key benefits that really stand out:

Understanding Consumers

There is nothing as valuable to a brand as understanding consumer-brand engagement. As an example, let’s look at Stella Artois beer and the upcoming Champions League final. It could now be discovered, for argument’s sake, that 53% of Stella Artois logo detections are in a home setting and feature bottles, as opposed to cans or glasses. Additionally, if a pizza and a TV showing the football match are found in the frame, it can be contextualised with the existing target audience persona data developed for Stella Artois, presenting the brand with an opportunity to refine it.

Campaign Planning 

Later when planning a winter campaign, Stella Artois marketers might discover that glasses featuring the lager’s logo commonly appear in an outdoor setting, with mountains and snow present. This presents Stella Artois with an opportunity to plan a cross-channel campaign positioning their lager as an apres-ski beer of choice, something which they have not attempted before. Thanks to the Object and Scene Detection data, they can be confident that many consumers will accept this positioning and find it relatable to their own interests. 

Building Brand Partnerships

While implementing Object and Scene Detection, it might also be discovered that in a majority of instances where the logo appears in a home setting, a specific food snack, let’s say nachos, is detected. This data can be used to open conversations with a complimentary brand, such as Doritos, about an effective brand partnership that could encompass anything from cross-promotion to discounts on joint purchases.

From these example use-cases, you can see just how powerful Object and Scene Detection can be when used in the context of brand data insights. It provides an immensely precise and detailed method for truly understanding and managing brand sentiment. This technology is the missing link that marketers have been needing to truly understand brand perception and engagement in this visually dominated world and a highly significant tool that can eliminate marketing spend waste.