Positioned to be the future of Ad Tech, Contextual Targeting is making a strong comeback as global privacy laws become increasingly restrictive and tech giants shift browser technology to maximize user privacy by phasing out third-party cookies entirely.
Digital Advertising Primer
Contextual advertising is a form of advertising which matches digital ads to the most relevant websites within the display network based upon keywords and topics of a web page. In short, it maps ads to the websites and web pages with the most direct relevance to the product, service or whatever else is advertisers are promoting.
GDPR Changed Everything
As success rates for agencies, media companies and publishers across the ad tech industry continued to soar, so did reports of data breaches and privacy scandals. Revelations of how numerous businesses and platforms had been misusing end user data — selling, sharing, or allowing unauthorized 3rd party access — left millions of end users exposed and feeling as though their privacy had been invaded. The massive pushback resulted in significant changes in terms of government regulation as well as modifications to some of the largest tech platforms.
Driven by massive individual pushback and demand for greater digital privacy, the EU was the first government agency to intervene and reform data privacy regulations governing how data can be collected, used, and shared.
In May of 2018, the General Data Protection Regulation (GDPR) went into effect. While the biggest impact from GDPR was somewhat contained to businesses operating within the EU, the impact was also felt by global companies with clients operating from within the EU. GDPR also came with broader implications to the entire ad tech industry in terms of what constitutes Personally Identifiable Information (PII) and drove discussions about what types of data could be harvested and stored, which parties would be held accountable for maintaining and protecting any stored data, and ultimately, how the data could be used to generate profits.
The EU was the first to throw a wrench into the cookie-based targeting and third-party data usage, though it is not alone. Shortly after GDPR, Australia expanded it’s 1988 Privacy Act and enacted the Australian Privacy Principles (APP) to regulate the collection, use, disclosure, storage and disposal of personal information.
Canada implemented the Personal Information Protection and Electronic Documents Act (PIPEDA) which regulates how private-sector companies may collect, use and disclose personal information for commercial or for-profit activities. PIPEDA also applies to any of Canada’s federally regulated businesses such as banks, airlines, or telecommunication companies.
And, in the United States, California passed the California Consumer Privacy Act (CCPA) in 2018 which went into effect as of January 1st 2020. While the CCPA is limited to residents of California, multiple other states are working towards creating their own privacy laws including Nevada, New York, Washington, and Texas as well as increasing momentum for a US Federal Privacy Law.
The Technological Shift
On the tech side, web browsers are engaged in a battle over adoption and market share with privacy and security becoming key differentiators.
The tech industry giants like Google, Microsoft, and Apple are responding to accelerating privacy demands and government regulations by using DNS + HTTPS (DoH) to enable encryption which maximizes user privacy at the browser level. DoH bypasses locally held DNS nameservers and sends encrypted traffic to a central server instead to anonymize web activity.
In 2019, both Google and Mozilla shifted to DoH for their Chrome and Firefox web browsers so users would benefit from the protections of encrypted DNS traffic. Apple is reportedly going even further on behalf of user privacy and baking compliance into the iOS which forces compliance and essentially removes the ability for a consumer to ‘opt-in’ to targeted advertising — breaking the programmatic advertising ecosystem.
The combination of privacy demands, growing regulation and evolving technologies have not yet disabled the cookie-based behavioral targeting, however, a future where the screen and user are completely anonymized is imminent — and could come within the next two years. Just a week ago, Google announced its plans phase out third-party cookies entirely in favor of using its new Privacy Sandbox — a privacy preserving API. Google is proposing that individual user-level information be stored in the Chrome browser, letting outside ad tech companies do an API call to the Privacy Sandbox to receive personalization and measurement data without user-level information.
So, while there is still a substantial amount of development that must take place over the next couple of years before we’ll see third-party cookies become obsolete, these changes will continue to disrupt the advertising industry forcing advertisers, DSPs, SSPs and marketers across the board to adjust to advertising without third-party cookies.
Adjusting to Cookieless Targeting
While some may see reverting back to contextual targeting as a move backwards, innovative technology and artificial intelligence have significantly improved, offering even greater opportunity for the advertising community to reach and engage target audiences and ensure brand safety while still adhering to compliance laws and reducing the risks associated with data misuse or security breaches.
Contextual advertising is making its comeback as it seems to strike the right balance between supporting user privacy while still driving ROI for brands and the entire advertising community. Contextual targeting leverages first-party data, rather than third-party data, allowing users to maintain data privacy for any sort of data that gets collected. And, advertisers are finding that contextual targeting can be more engaging as advertisements closely match the web users current real-time intent (i.e. shopping for a car, booking a flight, etc.), rather than perceived intent based on previous online behavior from as far back as 30 days.
How Contextual Has Changed
Artificial Intelligence (AI) and Machine Learning (ML) have made contextual targeting much smarter. Recent advances in AI/ML can now deliver highly accurate categorization and understanding of web content across a growing number of languages. Sophisticated algorithms enable machines to achieve a deeper level of understanding and better identify content topics, sentiment analysis, and even nuances such as homonyms. The increased level of sophistication work to mitigate the risk of mismatched content and ad placements to protect and maintain brand safety.
Another key difference in today’s contextual targeting is the capability to achieve full-path URL content categorization which classifies highly dynamic content at the page, post or article level. The ability to categorize web content at a granular level is crucial because it encompasses millions of sites with aggregated and/or highly dynamic and diverse content — like news sites, blog platforms, etc. — to ensure ad placements are accurately matched to the most relevant content to drive greater ROI and ensure brand safety.
Preparing for the Next Steps
The evolution of both privacy laws and technology over the several years have been a clear indication that the decline of cookie-based targeting was inevitable. And, as of last week, Google set the official timeline for its demise. If you are part of the ad tech community and you have yet to explore the new side of contextual targeting, the sooner you act the easier your transition will be.
zvelo partners with Ad Tech platforms on the buy and sell side to help enrich data and augment contextual targeting efforts. Delivering the market’s leading AI-based web content categorization and brand safety services, zvelo offers unmatched granularity with nearly 500 unique content categories, full IAB taxonomy support, as well as flexible deployment options — including the new zveloAPI™ which delivers full-path web categorizations. Additionally, thanks to ongoing advancements in natural language processing and lemmatization techniques, we support over 200 languages with over 99.9% classification coverage of the ActiveWeb.