By Dan Ward, Group Business Director, Seedtag
You’ve spent months planning and concepting a campaign, but come launch, vital questions remain. When is a consumer engaging with your brand? When are they primed to make a purchase? How does the digital advertising industry answer these questions in a privacy-first era?
Thanks to AI, advertisers can fill in these final pieces of the advertising puzzle with contextual purchase intent signals and attention metrics. Using predictive modelling, AI engines can show — with greater detail than ever before — exactly when a consumer is locked-in to your brand’s messaging and where they are in the sales funnel. Better yet, these intent insights come entirely third-party cookie and PII-free.
Why is intent so crucial to advertising effectiveness?
Without understanding consumer intent, advertisers have no way of knowing where the person they have reached currently sits in the marketing funnel. Intent drives the decisions on when, where, and how to engage with potential customers. Are they in the consideration phase, and therefore more responsive to awareness-raising campaigns? Or is their mouse hovering over the purchase button, awaiting a few final nudges to make the leap?
In the Wild West days of the internet — pre-GDPR — intent could be gauged through invasive and unregulated data collection, where individual consumers could be tracked from site to site to piece together an intent profile with unsettling specificity.
Today, such behavioural tracking is far more difficult, requiring explicit consent from the consumer while abiding by the strict data collection and processing regulations that are becoming the global norm. Only a limited sample of consumers will be viable for individualised tracking, as many will not consent or will be using platforms or devices where tracking and cookies are stripped out, such as the entirety of Apple’s ecosystem. As for the rest, probabilistic modelling is required to fill in the blanks.
But what if instead of attempting to spy on consumers to understand their intent, we simply let them show us?
How can context and attention reveal intent?
When an internet user browses the web, they leave behind a trail of non-identifiable signals that reveals a great deal about their interests, behaviours, and — crucially — intent. These signals form the basis of contextual advertising, where the content of a web page, the user’s interactions on the page, and movements from one page to the next are used as the basis for assembling and targeting audiences.
Each data point doesn’t tell us much in isolation, but when trillions of these signals combine and are fed into deep learning AI algorithms, complex patterns of cause and effect can reliably indicate audience traits. This modern iteration of contextual advertising goes far beyond simply matching keywords on a page. Contextual AI engines can analyse the overall theme and sentiment of the content, user interaction patterns, and even environmental factors such as time of day and weather conditions.
Contextual signals provide a nuanced understanding of where a consumer is in their purchase journey. For instance, a consumer reading an article about summer fashion trends on a sunny day is likely in a different stage of the buying process from someone reading the same article on a rainy day. The former might be in the consideration phase, actively looking for new outfits, while the latter might be merely browsing. By interpreting these signals, advertisers can serve ads that are relevant to the consumer’s current mindset and intent.
Of course, none of this matters if the consumer doesn’t notice the ad, which is where attention comes in. Unlike traditional metrics, which rely on black and white clicks or impressions, attention metrics focus on the quality of the engagement. These metrics assess how long a consumer views an ad and their interaction with the content, measured through opt-in panels that — like modern contextual advertising — have their data fed into predictive AI models to be extrapolated across all web users.
Attention and intention are closely aligned. If a user spends a significant amount of time viewing an ad for a new smartphone and engages with interactive elements within the ad, it indicates a high level of attention and potential purchase intent. This more granular engagement data is invaluable for identifying high-attention placements that will be effective in swaying consumers who are genuinely interested in the product.
A predictive, privacy-first future for intent measurement
Context and attention are even more powerful in combination. Together, they provide a comprehensive view of consumer behaviour and intent without intruding on privacy. Both deploy AI-powered tools can analyse vast amounts of entirely anonymised data in real-time, identifying patterns to make accurate predictions about consumer actions without any chance of an individual being identifiable while also being applicable across any digital advertising environment.
For example, a user who lingers on a page reviewing different types of coffee machines, scrolling slowly and engaging with interactive elements, signals strong interest. If this behaviour occurs on a weekday morning, the contextual signal suggests they might be a working professional looking to purchase a machine for home use. An ad tailored to highlight the convenience and efficiency of a particular coffee machine would likely resonate well with this consumer.
As privacy regulations tighten and consumers become more aware of how their data is used and misused, adopting privacy-first advertising strategies is not only compliant, it also builds consumer trust. By focusing on non-invasive data such as contextual signals and attention metrics, advertisers can respect user privacy while still gaining valuable insights on consumer intent.
If behavioural, contextual, and attention-based intent measurement all rely on probabilistic, predictive modelling now, why wade into the murky and strictly regulated waters of individual-level data collection? With AI, we no longer need to hoover up all the data we can get, so let’s have consumers show us their intent and meet them there.
