New Old Solutions: Exploring Contextual Targeting in Programmatic Advertising
Global programmatic spend is nearing $550 billion in 2023, and that number will only grow in the future. Since the late 1990s, programmatic advertising has been a driving force in modern marketing, and it is used by virtually every advertiser.
However, lately, it has been going through a difficult period. The ongoing privacy talks have already put an end to third-party cookies, a very useful tool for collecting customer data. Being compliant with GDPR takes its toll, too; marketers have to use transparent and customer-friendly solutions for data collection, which makes it harder to utilize the full potential of programmatic advertising.
Fortunately, there is a solution unchanged by these latest developments. It does not require any data to be effective while still personalizing the content for the customers. Additionally, it reaps benefits from machine learning and AI, both of which have undergone very rapid development in the past few years.
In today’s blog post, we’ll talk about the ad tech uses for contextual targeting and how it helps offset the drawbacks of privacy regulations.
A Perfect Fit: What is Contextual Targeting?
A customer is interested in cooking and decides to visit a website with a broad selection of recipes. Reading through the recipes, they notice display ads, specifically — ads promoting cooking utensils that are especially useful in the exact recipe they are reading about.
This is a prime example of contextual targeting. Advertisers looking to promote a specific type of product look for ad spaces that are correlated with that product. It does not necessarily have to be a website; for example, marketers may strike a deal with TV channels to show car ads during racing shows. The selection of formats is rather broad; contextual targeting can work on DOOH displays or even via audio ads. However, contextual targeting for display advertising is the most popular choice, and for a good reason. The Internet is filled to the brim with articles about virtually anything, and it is easy to find an appropriate ad space for any product.
Contextual targeting can be divided into several types. In contrast with other marketing strategies, the choice of a specific type does not depend on the marketer’s needs but rather on the choice of medium and preferred budget.
Categorized Contextual Targeting
Advertisers can decide on a broad category they wish to display their ads in. For example, they may choose “Sports” or “Food & Drink” categories, and the publisher will show their ad on all websites corresponding to that category.
With this type of CT, marketers leave the majority of choices to the publishers they cooperate with. Because of that, accuracy may suffer since categories are defined broadly, and products may not fit perfectly on every chosen website. This type is the most popular with big DSPs and giants like Google, which have huge catalogs of available ad spaces. Because of its average accuracy, it is usually also the most affordable option.
Keyword Contextual Targeting
To narrow down on desirable ad spaces, advertisers may move on from Categorized CT to its evolved form, Keyword CT. Here, they choose preferable keywords that websites’ content will have to match. These can be anything; for example, marketers can opt for simple ones like “bikes” or pin it down even further to “middleweight sport bikes 2023”. As soon as they are done, publishers’ crawlers will track down websites matching these keywords and place the ads there.
This method is far more accurate since carefully chosen keywords can eliminate the possibility of misplacing ads almost entirely. However, it is a little bit more costly, and the reach may be lower if advertisers opt in for very specific keywords. Still, Keyword CT is superior to Categorized CT in almost every sense.
Semantic Contextual Targeting
The most recently developed and advanced of the three, Semantic CT, is also a good choice. Here, contextual targeting systems employ machine learning and AI to use natural language processing to analyze the contents of a website’s page and decide on ads that will be displayed to a user. In simpler terms, algorithms “read through” the page, deduce what the content is about, and serve corresponding ads.
In this case, advertisers also need to choose keywords, but they may go as broad or specific as they desire; NLPs and AI will do the rest without lowering the accuracy. It is the perfect option for contextual targeting, but it is usually also the most expensive one in terms of budget. Semantic contextual targeting is also the most rapidly developing field since its evolution is closely tied to the advancement of NLP technology.
Benefits of Contextual Targeting
Contextual targeting is often opposed to behavioral targeting since they have the opposite fundamental principle. The latter is based on acquiring customer data, while the former does not require any. That very distinction is where most contextual targeting benefits from, and is the reason for the current contextual targeting strategy popularity.
Data-free Means Simple
Since all types of contextual targeting require no data to process, it becomes a very enticing choice for advertisers. They do not need to work around privacy regulations to acquire customer information, spend their budget on purchasing data sets, or find complex data management platforms to process that information. All marketers need to do to utilize contextual targeting is find the right Data Supply Platform, set up their campaign, and they are good to go.
Personalization Is Still Here
One thing behavioral and contextual targeting have in common is content personalization. While the former relies on the accuracy of the inputted customer data to show personalized ads, the latter uses only the website contents. With semantic targeting, ad servers have little room for mismatched ad displays.
Additionally, advertisers have control over their contextual ad targeting campaigns. They may choose to ignore specific categories or pages, thus increasing the accuracy and relevancy of their ads, which reinforces personalization even more.
Non-Intrusive Option
One of the recent privacy concerns is the intrusiveness of modern ad campaigns. Customers may feel uneasy when followed by several ads for the same product on different occasions — a giveaway sign of a behavioral targeting strategy. Contextual ads are non-intrusive; users only see ads for specific products when visiting websites dedicated to the corresponding topic. These ads do not follow their customers while still generating interest because of their contextual nature.
Is Contextual Targeting The Future?
While price disparity within contextual targeting types exists, it is still a far more affordable option than behavioral targeting. Its benefits and availability make it a rather strong contender among all marketing strategies. Its characteristics also allow it to avoid data privacy concerns, reinforcing its position in digital marketing even further.
While it is true that at present many advertisers turn to contextual advertising, it is largely a sign of unstable times. Behavioral advertising and other targeting strategies are not going anywhere; the advertising world is already coming up with new ways to ethically acquire customer information, and it is only a matter of time before the behavioral approach will return to full power.
As this happens, contextual targeting is also gaining strength. These targeting strategies should not be seen as opposing each other but rather as forces that can be combined and used at different times. Each has its own set of benefits and perfect use conditions. Though, in these turbulent times, contextual targeting may be leading a little bit.
