Video ads are incredibly effective but also very expensive. How can advertisers be sure they’re getting in front of the right audience? Åukasz Abgarowicz, VP of agencies at RTB House, explains how deep learning, or advanced AI, holds the key. Â
Over the past few years, we have seen video become a vital part of successful marketing strategies, with video consumption growing rapidly. In fact, in the first 18 months of the pandemic, video consumption increased by 41%Opens a new window .
Having video advertising is one thing, but with 86% of businesses now using video as a marketing toolOpens a new window , advertisers need to think more about differentiating themselves from the crowd. Personalization, context, and targeting are key to doing this while also achieving brand visibility. In fact, research shows that today 56% of consumersOpens a new window think that brands need to establish a deeper understanding of their needs and create more meaningful connections with them.
However, many campaigns end up falling short of this, meaning budgets are wasted. So, in a world where many identifiers are disappearing at a fast rate, how can marketers deliver personalized experiences? That’s where deep learning comes in.Â
What Is Deep Learning?
Deep learning, or advanced artificial intelligence (AI), is the most advanced AI set of algorithms capable of understanding human decisions with unprecedented precision and predicting future actions based on online behavior. For example, it can make language translators translate like a human native.Â
While traditional machine learning cannot make decisions without being guided by human input and has difficulty handling unstructured and granular datasets, Deep learning uses its knowledge of previous events to respond in real-time.Â
Further, it can analyze huge amounts of data in fractions of a second and use this information to decide which ad creative will be the most effective for the desired audience. In addition, Deep learning adapts autonomously without needing further input from humans and is up to 50% more efficient than standard machine learning AI-based approaches.Â
As a result, the technology can enable advertisers to maximize the efficiency of their targeted ad campaigns in ways beyond the capabilities of even the world’s best media planners.Â
It can also support multiple targeting strategies – whether the advertiser uses contextual, behavioral, or demographic data – enabling content to be adjusted to the target group’s preferences.
What Does this Mean for Video Campaigns?
Deep learning can make a significant difference in efficiently gaining attention, which is crucial given the current economic environment, where marketers will reconsider how they spend their budget.
We know video consumption has been growing rapidly over the past few years, but this also means it costs brands a lot more of their marketing budget. With the amount of money spent on producing video ads, advertisers need their ads to be seen by the right people, in the right places, at the right times.
Deep learning’s ability to pick the right time, place, message and right audience through its real-time predictions enables advertisers to improve the customer journey and identify specific touchpoints in this journey to create the most effective video advertising strategy. In turn, this increases the brand’s visibility and maximizes the cost per completed view (CPCV).Â
Deep learning-powered technology for video ads can also increase viewability up to 33% and viewer completion rates (VCR) up to 47%. This increase in effectiveness is vital to brands and marketers because the average viewer retains 95% of the message while watching it, as opposed to just 10% of the retention rate when reading it.Â
See More: 10 Tips for Advertising Success in Video & Mobile Games
The Evidence
Steve Madden, one of the most iconic footwear brands, wanted to run an effective video campaign to expand its reach with new users ahead of the festive season.Â
The challenge was that it was a crowded marketplace where Cost Per Impression (CPM) was steadily increasing. As a result, Steve Madden wanted to launch an online and offline video campaign to stand out from the competition. Further, the media budget needed to be used to its full potential to maximize ROI.Â
In order to meet the campaign objectives, a Deep learning solution was utilized to allow the footwear brand to reach the most engaged and quality users. A 15-second video campaign was launched that enabled quality complete views in prime placements with the world’s best publishers. The digital creatives used in this were also part of a TV campaign featuring model Reef Neeman as Steve Madden’s presenter.Â
Ultimately, the purpose of the campaign was to bolster Steve Madden’s brand image among young female audiences, and with the power of deep learning, Steve Madden was able to target this audience based on chosen interests and publishers.
The results spoke for themselves. Steve Madden was able to extend the reach of users and achieved over 70% viewability as well as a best-in-class cost per completed view (CPVC) that was 25% lower than local benchmarks. In addition to this, the fashion brand was able to reach younger female audiences with ads running on key publishers such as Ynet, Walla and Sport5, which in turn ensured the CPCV was lowered without compromising the quality of websites used. To that end, the campaign proved the unmatched power of deep learning in reaching users through video.Â
As the world moves closer towards the disappearance of many third-party identifiers, now is the perfect time for marketers to begin to harness the power of deep learning and its extraordinary abilities based on privacy-first, data-led predictions.Â
Deep learning brings a plethora of advantages to marketers, in particular, more efficient media buying alongside real-time optimization of both your buying strategy and your creative output, delivering increased brand visibility, more impactful video campaigns, and better value for money.
How are you harnessing the power of AI in video marketing? Share with us on FacebookOpens a new window , TwitterOpens a new window , and LinkedInOpens a new window . We’d love to know!