YouTube has unveiled new tools to help companies capture consumers’ attention on its platform. These are “Video Ad Sequencing”, “Custom Affinity Audiences”, “Director Mix”. It is also working with Nielsen Matched Panel Analysis (MPA) to roll out a new tool.
The Video Ad Sequencing, said the company in a blog post, will allow advertisers to take consumers down different paths, depending on which ads they are engaging with. Meanwhile, the Custom Affinity Audiences is enabling companies to further fine tune their audience targeting by expanding on the ways they can leverage Google’s ecosystem, which allows advertisers to reach audiences more tailored to their brands.
Through this new enhancement, companies can target consumers based on the type of searches they make as well as the place and apps they like. Discussing the new tool with Marketing, Dominic Benton, regional analytics and innovation director of MEC, said targeting of intent rather than by mass audience is definitely an attractive proposition for advertisers – provided they are comfortable with being on the platform.
He also weighed in on the new tool that YouTube and Nielsen Matched Panel Analysis (MPA) are working on, saying that using geolocation data to attribute offline sales seems promising, “but the limited accuracy of mobile location makes it hard for the attribution to be accurate”. Benton said that the only stores that can attribute using location are large hyper markets (stores such as IKEA) which have huge surface areas and limited other stores nearby for misattribution.
Meanwhile, YouTube also explained that the new partnership with Nielsen will lead to a “fast, media-agnostic way” to ascertain which online ads drive offline sales. The blog post stated that the implementation of Nielsen MPA allows clients to measure video alongside other Google media.
Brendon Chase, group strategy director, McCann Worldgroup Singapore & MRM//McCann strategy lead, APAC said when it comes to geolocation targeting, if “the methodology is right for the product purchase window, then companies should embrace attribution”.
However, the trouble may lie in some markets where measurement and analysis are not as conclusive and cheap as companies would like.
He added, “For example, how would you attribute a YouTube ad for a B2B purchase which takes a year to close the sale and includes up to half a dozen people in the decision-making process? And was it the exposure to the ad, or was it the five to 20 other pieces of content on YouTube which made a customer buy the product?”
How effective is YouTube’s Director Mix?
YouTube’s new “Director Mix” simplifies the process of producing different versions of the same creative tailored for different target audience, as the system is able to generate “thousands of versions” of the creative to match the company’s various audience segments. Companies are only required to provide the basics of a video ad such as different voice overs, background and video copy.
Chase said companies should embrace technology where it makes sense, especially for performance-based campaigns or messages that can be targeted at different segments. He cited the agency’s work for Clorets Mint Tab in Japan which saw two different videos produced – one by creative director Mitsuru Kuramoto and the other by its AI creative director AI-CD β. The videos were made available for a nationwide polling and the human ad ultimately won, albeit by a narrow margin.
On the other hand, Benton said artificial intelligence (AI) is only as good as the human calibration that sets it up.
If poor creative assets are uploaded onto the Director Mix, the combination that comes out will always be poor, no matter how sophisticated the model is.
“It will be extremely difficult to make assets that work with multiple different combinations of other assets unless they are hugely generic, but by making generic assets it loses the personalisation element that it intended to create,” he added. Also, personalised components might pose a challenge for advertisers.
Benton cited the Campbell Soup example that Google used to showcase its Director Mix, which created copies based on the content people were about to watch. For instance, if consumers are watching clips from Orange is the New Black, they would see a bumper ad asking them “Does your cooking make prison food seem good? We’ve got soup for that.”
Such niche personalised elements might not work as well with generic elements in the rest of the creative piece, if the creative had been created from scratch with an intended purpose. He added that creative creation, especially video creatives, is expensive.
“[While] using AI to create “thousands of versions” might not be the best starting point, but using some form of dynamic video creative is definitely something I can see catching on to minimise creative spends and giving advertisers more flexibility,” Benton said.
Meanwhile, Paddy Crawshaw , OMD APAC’s head of strategy, said the capability is “fantastic” in principle as it democratises an advertiser’s capability to deliver against the promise of addressability – to begin serving relevant, interesting, intriguing and persuasive messaging to audiences within and beyond the immediate demand opportunity.
However, Crawshaw said just because the message is appropriate to the audience it does not necessarily make it relevant or compelling. That requires applied human insight which would have to come from brands and agencies being able to explore the data that defines the segments and be able to create their own.
He added that YouTube’s Director Mix does not absolve the advertiser of the need to build a greater “Why” for the brand as the bedrock for a broad base of universal appeal. He said, “In other words, the ‘thousands of versions of content’ are only valuable if there is one, salient version of the brand story out there delivered across channels that is resonant with consumers.”
Leaving the stains of the past
While this announcement is another step towards effective targeting and audience engagement by YouTube, it does not take away from the fact that its image has been impacted by the issue of placing advertisements over questionable content on its site.
According to MEC’s Benton:
In terms of repairing YouTube’s image, this announcement will do nothing to ease advertiser’s fears that their ads could be displayed against content they wouldn’t want to be associated with.
“YouTube have put some little fixes in place to try to meet advertiser’s expectations but it is still not enough,” Benton said. However, he did add that the new features were a step in the right direction and it is good that YouTube is constantly innovating.
Also weighing in on the topic is MRM//McCann’s Chase who said while YouTube is trying hard to ensure ads are not seen around questionable content, it is still after all a social media platform that is not as controlled as traditional TV or linear broadcasts.
Although this is part of YouTube’s allure, it can also be a concern for some advertisers. Unless more human checks and balances are in place, which Chase said would be extremely hard for any company, even the size of Google, some slip ups are bound to happen.