This article on advertising and AI is part of our Vogue Business membership package. To enjoy unlimited access to our weekly Technology Edit, which contains Member-only reporting and analysis and our NFT Tracker, sign up for membership here.
Advertising is poised for an upheaval this year. Tracking cookies are disappearing from websites, generative artificial intelligence is coming for ad copy and new devices are adding a digital layer to physical ad space.
In the background, a number of factors are keeping executives up at night: economic and inflationary pressures, the need to keep pace with changing consumer preferences and profitability and margin pressures, according to a new Ernst and Young survey of more than 250 US consumer packaged goods and retail executives. Almost half say the current economic environment has changed their business strategy.
That’s why advertisers now favour practical investments that save money and enable current teams to become more efficient — with a dose of experimentation. “The conversation now is understanding the big picture to make decisions that set yourself up for success for this broader AI and data generation, while being able to pull that down to specific use cases that you want to deliver in the next three to six months,” says Jay Wilder, VP of product marketing, Marketing Cloud at Salesforce. “It’s a balance between the horizon and the foreground.”
These new focuses build on each other, ranging from the immediate need to upgrade a data strategy to the emerging need to use that data and leverage generative AI. Eventually, the new norms of digital advertising could expand to smart glasses, which bring digital advertising outside of existing screens. The link is AI.
While CEOs are exercising spending caution in many areas of their business, the targeted application of AI is not one of those areas, says Purva Gupta, co-founder and CEO of Lily AI. Three-quarters of the 274 retail decision makers surveyed by Google Cloud believe generative AI can be practically deployed in their business. “Practical prudence is in play, of course, yet practical prioritisation is in play as well.”
Changes to cookies and data
For decades, brands have worked with ad tech platforms to place targeted ads across websites through tracking cookies; that’s why someone looking at a pair of shoes on one site might subsequently see ads for that same shoe across other sites. That type of tracking is being phased out, amid governmental regulations such as the UK’s General Data Protection Regulation and The California Privacy Rights Act, which require sites to ask permission.
Google has started blocking third-party cookies from its Chrome browser in a bid to “make the web more private”. It plans to automatically block all of these cookies by the end of this year. Competing browsers have already made similar changes, but Chrome is by far the most widely used. Brands will be compelled to pivot from “third-party data” – meaning information automatically collected from other sites — to first-party and zero-party data, meaning data that is collected directly and explicitly from their own customers.
While brands have known this was coming, they aren’t fully prepared, says Derek Slager, CTO at Amperity, which works with companies like Tapestry Group to build customer data platforms and analytics. “We haven’t really experienced a world without the cookie just yet,” he says. “It is a really big deal. When talking about the ‘cookie-pocalypse’, I’ve been surprised how many people shrug their shoulders.” Now, if brands want to track people, they have to rely on people giving permission for the site to use cookies, and as much as 85 per cent of people typically say “no”, he says. Additionally, with Chrome’s new policy, even if a user agrees to third-party cookies, Chrome intends to phase out support for them, regardless of consent, he adds. Meanwhile, brands are likely to rely more on advertising via big tech companies such as Google and Meta to acquire customers. As the competition increases for ads on those platforms, it will become harder for brands, he says. “There are only so many slots, so it just gets more expensive.”
As an alternative, brands are likely to focus on their own sites and their own customer-supplied data, gathered via loyalty programmes, email logins and by proactively asking for feedback or preferences. Advertising to existing customers, via upselling or cross-selling on email or SMS, might become more of a focus, says Salesforce’s Wilder; the concept of advertising is expanding to audiences across the customer journey. “I hear a lot of, ‘How do we grow the depth of our current customer relationships?’ Brands are trying to move away from acquiring customers continually.”
Brands might collaborate with other brand loyalty programmes, Slager says (which would still require customer opt in). A multi-brand conglomerate, such as Tapestry or Gap, could link up its loyalty programmes to cross-sell to customers, or consider new ways of targeting across brands. Companies with similar customers could also join forces. Nike and Dick’s Sporting Goods linked loyalty programmes, and more than one million of Dick’s loyalty members linked accounts with Nike, according to a recent earnings call. In September, Lululemon and Peloton signed a five-year partnership.
There’s also an emerging market for retail media networks, which is when a multi-brand retailer, such as Amazon, Macy’s or Target, sells advertising on its own properties to brands that it carries. This offers multi-brand retailers an advantage, Slager says. A recent Cowen report, by managing director and senior equity research analyst Oliver Chen, finds that RMNs (retail media networks) are valuable because they can match ad impressions with sales, looking at insights such as conversion stats, acquisition channel performance and campaign engagement data. Prada, advertising via Macy’s, doubled its return on advertising spend and saw an 85 per cent new customer acquisition rate, Cowen reports. Cowen estimates that US retail media spending will grow to $82 billion by 2027, from $38 billion in 2022. Search on retail media will account for more than half of all new search ad dollars this year, according to Insider Intelligence, with just one out of five going to Google. “Retailers can target their own customers using their proprietary ad stacks, which becomes a scalable alternative to brands’ digital marketing campaigns,” the report states.
Wilder feels that brands might come to see first-party data as not merely a replacement, but rather an improvement, as it’s unique, proprietary data that allows them to be more “nimble”, he says. “It provides the value they are looking for, versus being all things to all people at all times.”
Artificial intelligence for augmentation and automation
Nearly all executives surveyed by Ernst and Young are experimenting with generative AI in some capacity. The AI renaissance sees new opportunities for creativity and efficiency, but most new applications require brands to start with organised data. “It’s more important than ever to ensure one’s product data is incredibly clean, robust and accurate or else it is not ready to train on, thereby creating irrelevant results and undesirable consumer experiences,” Gupta says. Lily AI, for example, just released new tools that enable brands to automate SEO-optimised marketing copy that makes use of a brand’s first-party data.
Google, Salesforce and Microsoft all just announced new AI tools for retailers. Microsoft is introducing a creative studio whereby brands advertising on a retailer’s site can auto-generate and edit optimised banner content using generative AI. Separately, a “copilot” tool enables marketers to type in a desired outcome or upload an existing brief to receive a generated project board, including AI-recommended content. They can also produce brand-authentic images that are aligned with a company’s themes, fonts and product images.
Salesforce, meanwhile, is giving retailers predictive insights into promotions. A new tool combines customer marketing and promotions data for retailers to forecast promotion revenue and coordinate pricing and rewards initiatives. Salesforce is also letting retailers target and personalise content towards specific customer segments, based on their customer data, using generative AI prompts. And retailers who use Google Cloud can “custom-tune” a large language model to their product catalogue and shopper search patterns to improve results on customer queries for retailer sites. These are in addition to tools from Google, Amazon and Meta that enable advertisers to change product image backgrounds or make other ad changes using generative AI.
Cowen analyst John Blackledge expects that these big tech platforms will continue to launch additional Al-powered marketing tools, adding that the productivity gains could help offset the cost of the new tools. Retailers see an average return of $3.45 for every $1 they invest in AI, according to an IDC study commissioned by Microsoft.
Ad networks were one of the first to leverage AI, creating dynamic ads that are rule-based to maximise engagement and spend, says Dave Anderson, VP of product marketing and strategy at digital analytics firm Contentsquare. Now, generative AI can help resolve “traditional bottlenecks” in creating advertising content, Salesforce’s Wilder says, as creating manual content limits abilities to personalise content on a large scale. Now, he says, the conversation is focused on getting the data organised to enable more experimental, targeted and personalised ads. An advertiser might amend email subject lines depending if someone has a service ticket open, or if they just bought a specific product. A high-value audience might be a good candidate for a VIP loyalty programme offer, while a new customer might benefit from influencer messaging that conveys the brand’s essence. Brands might amend visual assets with specific geographies or seasons.
By next year, about 30 per cent of marketing messages from large organisations will be “synthetically generated”, Gartner predicts; in 2022, that figure was less than 2 per cent. Lily AI’s Gupta calls it “the great augmenter”.
Virtual billboards bring digital advertising to the outside world
Big tech companies have begun releasing a new generation of eyewear that can “see” and interpret the world around the wearer; some mixed reality headsets can add digital content to the field of vision, while smart glasses can include a generative AI assistant that can respond to visual cues. Digital advertising is expected to expand to the outside world through these devices.
This isn’t entirely far-fetched. Recently, Meta CEO Mark Zuckerberg demonstrated how the Meta and Ray-Ban smart glasses see a shirt and provide styling advice, via Meta AI. Louis Vuitton, through a Snapchat Lens, covered historic landmarks in digital artwork. Coach added an AR mirror to its store window to enable people to try on digital bags without entering the store. And Maybelline cheekily added digital mascara to a video of a London Underground train, using CGI. The Institute of Digital Fashion plastered QR codes around London to promote its digital fashion. A Tommy Hilfiger ad in London invited people to scan a QR code to access an immersive experience and a guided path to the brand’s store, where a second AR experience was accessible via a store window.
These types of experiences could be combined with digital advertising technology and smart glasses to enable outdoor ads to offer personalised content to passers-by. The release of the Apple Vision Pro next month only increases the hype for spatial computing.
“Our premise is, as we start shifting from mobile to wearable and from flat, 2D web into immersive and spatial web, advertising will need to shift too,” says Dominic Collins of Darabase, who worked with Tommy Hilfiger on the London project. Darabase is a combination of “database” and “AR”, as the company is compiling a database of buildings upon which future advertisers can place digital ads. Combining outdoor media with digital ads that can be targeted and personalised offers a huge opportunity for building owners, he says. Now, the focus is on securing digital property rights.
Similarly to current web ad pop-ups, customers can opt in or out, and there are likely to be limits to how many can be in one place. Collins notes that with AR experiences, especially, it’s less “interruptive” because consumers need to take some action to access it. Plus, consumers wouldn’t like automatic pop-ups. “Unilever aren’t going to spend money if there are a gazillion other ads around them, so we will be forced to find ways of creating engaging immersive experiences that users don’t find annoying,” he says.
Additionally, people wearing glasses might be able to set preferences related to their interests — more than they are able to when browsing the web, says Olly Bray, a partner at law firm RPC who advises luxury brands and specialises in future formats like these. “Fashion brands who want to be eye-catching can have a live catwalk come to life,” with the consumer’s consent to receive fashion images as they walk around a city.
To begin, they advise that property owners start thinking about how they might sell digital advertising on their own buildings, as those who create commercial, location-specific AR experiences need permission. (Louis Vuitton, Collins notes, was smart in that the AR content on landmarks was art, rather than branded material.) Additionally, brands should explore how AR content might add value to the outside world, he advises. A project to promote the TV show Bridgerton, for example, enabled people to hold up their phones in London’s Mayfair neighbourhood to be taken back in time. A retailer could use a similar approach at an historic location. “There are lots of ways on mobile that you can start to engage people.”
To receive the Vogue Business newsletter, sign up here.
Comments, questions or feedback? Email us at feedback@voguebusiness.com.
