Advertising

contextual advertising: exploring the spectrum of options

Contextual advertising is a term that refers to a range of targeting methodologies used in online advertising. Ned Dimitrov, the VP of Data Science at Stacked App, and Yang Han recently discussed the two main types of online targeting: behavioral and contextual.
Behavioral targeting uses stored or historical data, while contextual targeting targets the user based on the particular page they are currently reading. Contextual targeting can use categories, keyword targeting, or newer methods, including AI-based or human-controlled algorithms.

In recent years, the performance of behavioral targeting has declined due to changes in privacy laws and technology. In contrast, contextual targeting has stepped up to make up for the decline in behavioral targeting. While retargeting still performs better overall, contextual advertising has shown to outperform retargeting in some cases.
There are two types of contextual targeting: base contextual targeting and hyper-managed contextual targeting. The speaker explains the advantages and disadvantages of contextual AI for advertising campaigns. Contextual AI can create a context from a few words, find many related words, and disambiguate words with multiple meanings. However, it does not generate per-word reports, requires user education, and requires a new way of thinking about campaign optimization. Contextual AI is different from search and key phrase matching.

The speaker compares contextual AI with hyper-manage, which is based on simple key phrase matching. The latter lacks an understanding of the topic and can generate irrelevant results. Contextual AI can create sophisticated rules that specify inclusion and exclusion criteria, leading to relevant results.
The talk discusses the spectrum of contextual advertising options and their advantages and disadvantages. Contextual AI is easy to scale and adapt to the changing internet, but requires some education and optimization. Hyper-managed is based on simple phrase matching and provides full control, but requires a lot of manual effort and rule creation. AI assisted is a middle ground option that uses AI to suggest word and phrase expansion, but still requires some manual effort and has limited reporting.

The talk emphasizes that there are many contextual options available to fit different advertiser needs and performance and reporting requirements. The key takeaways are that contextual strategies perform well in practice, and there is room for industry innovation across the entire spectrum of options.

In conclusion, contextual advertising provides a range of targeting options for advertisers to choose from. While behavioral targeting has seen a decline in performance due to changes in privacy laws and technology, contextual targeting has stepped up to fill the gap. There are different types of contextual targeting, each with its own advantages and disadvantages. Advertisers can choose from various options based on their needs and requirements for performance and reporting. The talk highlights the need for industry innovation to keep up with the ever-changing landscape of contextual advertising.