How brand suitability can improve your ad targeting
October 27, 2020
SCMP Insights
About four months ago, we decided to build and launch SCMP Signal, a product which has evolved brand safety into brand suitability for marketers and also provides them with customisable contextual, keyword and sentiment targeting options.
Initially we worked with our data team to create a South China Morning Postnatural language processing project, with the aim to understand: 1. the emotion of the article delivered; 2. whether the article contains the given sensitive words and; 3. how easy it is to be understood by the audience.
The three components – sentiment analysis, sensitive keywords tagging and article readability analysis – work together to provide a complicated profile of the written articles, which looks not only at the context, but also the sentiment, keywords and readability. All of these innovations , plus the use of a lexicon-analysis tool, called Valence Aware Dictionary for Sentiment Reasoning (VADER), gave rise to SCMP Signal.
The product, which was built directly into our content management system to ensure a 100 per cent coverage on all of our articles, supports not only our advertising unit, but also plays a big part in our editorial DNA structure. Its ability to understand what kinds of sentiment scoring or keyword pairing will drive pageviews means we can focus and produce more topics and content that expand our readership, increase page-session depth and lengthen user engagement.
Why is this the ‘new norm’?
At SCMP we believe the application of brand suitability for all direct campaigns to be the “new norm” to address the following industry concerns;
Evolution of brand safety
Over the years, we have seen the industry try to protect the image and reputation of brands from the negative or damaging influence of questionable or inappropriate content through the use of various methods, including whitelisting, which still allows access to approved applications and websites, blacklisting, which blocks unwanted entities, and keyword targeting. While they have been effective, they also have been known to limit reach and not be the best solutions.
For content producers such as SCMP, keywords without context could lead to brands missing out on key audiences: an article headlined “New Canon camera shoots multiple coloured images”, could be negatively flagged because of the keyword “SHOOTS”. We are seeing an increase in marketers leaning towards suitability, and we believe this is a pivotal point in our industry as suitability is more about inclusiveness than exclusion. The ability to match brands’ values to context they may appear to be against exposes them to their audiences when it matters.
Death of the third-party cookie
Since Google's announcement that third-party cookies will be blocked on its Google Chrome web browser and Apple’s announcement about restrictions on the Identifier for Advertiser – its random device identifier which enables advertisers to track data so they can deliver customised ads –, the industry has been faced with many futures, including the rise in publisher consortiums and a buyers’ focus on contextualisation. We expect contextual buying to be a key focus for buyers as we lose some behavioural targeting capabilities.
The results so far
We believe publishers have always been best placed to help brands find audiences and content to build relevance and engagement. It took us three months to create it, but now we have independent tech that adds value to advertisers, with early results in some cases indicating a more than 40 per cent increase in performance.
A health and fitness campaign, for example, which ran from May toAugust this year, showed that some strategies which applied sentiment targeting helped to drive a more than 35 per cent increase in performance compared with some run of site strategies
The graph above showed us that campaign strategies with sentiment applied drove a significantly higher performance than other lines. But the most interesting aspect to us was the sentiment breakdown. This gave an insight into the role emotion can – and does – play in digital advertising.
We saw that articles with a fairly negative score drove a higher performance than those with a fairly positive score, which was astonishing to see during these challenging times. We were able to understand and confirm that users react differently, not only to content, but also their state of emotion when consuming ads.
These key findings across our campaigns have given us as publishers a lot of confidence in brand suitability – and for brands a key opportunity to tap into their audience’s feelings and emotions without having to sacrifice on performance.
We are not stopping here. SCMP is currently adding engagement metrics and a viewability measurement to help prove campaign return on investment. You, too, could build one – and take back some control.
Initially we worked with our data team to create a South China Morning Postnatural language processing project, with the aim to understand: 1. the emotion of the article delivered; 2. whether the article contains the given sensitive words and; 3. how easy it is to be understood by the audience.
The three components – sentiment analysis, sensitive keywords tagging and article readability analysis – work together to provide a complicated profile of the written articles, which looks not only at the context, but also the sentiment, keywords and readability. All of these innovations , plus the use of a lexicon-analysis tool, called Valence Aware Dictionary for Sentiment Reasoning (VADER), gave rise to SCMP Signal.
The product, which was built directly into our content management system to ensure a 100 per cent coverage on all of our articles, supports not only our advertising unit, but also plays a big part in our editorial DNA structure. Its ability to understand what kinds of sentiment scoring or keyword pairing will drive pageviews means we can focus and produce more topics and content that expand our readership, increase page-session depth and lengthen user engagement.
Why is this the ‘new norm’?
At SCMP we believe the application of brand suitability for all direct campaigns to be the “new norm” to address the following industry concerns;
Evolution of brand safety
Over the years, we have seen the industry try to protect the image and reputation of brands from the negative or damaging influence of questionable or inappropriate content through the use of various methods, including whitelisting, which still allows access to approved applications and websites, blacklisting, which blocks unwanted entities, and keyword targeting. While they have been effective, they also have been known to limit reach and not be the best solutions.
For content producers such as SCMP, keywords without context could lead to brands missing out on key audiences: an article headlined “New Canon camera shoots multiple coloured images”, could be negatively flagged because of the keyword “SHOOTS”. We are seeing an increase in marketers leaning towards suitability, and we believe this is a pivotal point in our industry as suitability is more about inclusiveness than exclusion. The ability to match brands’ values to context they may appear to be against exposes them to their audiences when it matters.
Death of the third-party cookie
Since Google's announcement that third-party cookies will be blocked on its Google Chrome web browser and Apple’s announcement about restrictions on the Identifier for Advertiser – its random device identifier which enables advertisers to track data so they can deliver customised ads –, the industry has been faced with many futures, including the rise in publisher consortiums and a buyers’ focus on contextualisation. We expect contextual buying to be a key focus for buyers as we lose some behavioural targeting capabilities.
The results so far
We believe publishers have always been best placed to help brands find audiences and content to build relevance and engagement. It took us three months to create it, but now we have independent tech that adds value to advertisers, with early results in some cases indicating a more than 40 per cent increase in performance.
A health and fitness campaign, for example, which ran from May toAugust this year, showed that some strategies which applied sentiment targeting helped to drive a more than 35 per cent increase in performance compared with some run of site strategies
The graph above showed us that campaign strategies with sentiment applied drove a significantly higher performance than other lines. But the most interesting aspect to us was the sentiment breakdown. This gave an insight into the role emotion can – and does – play in digital advertising.
We saw that articles with a fairly negative score drove a higher performance than those with a fairly positive score, which was astonishing to see during these challenging times. We were able to understand and confirm that users react differently, not only to content, but also their state of emotion when consuming ads.
These key findings across our campaigns have given us as publishers a lot of confidence in brand suitability – and for brands a key opportunity to tap into their audience’s feelings and emotions without having to sacrifice on performance.
We are not stopping here. SCMP is currently adding engagement metrics and a viewability measurement to help prove campaign return on investment. You, too, could build one – and take back some control.
