Is AI the Remedy for Brand Safety Woes?
When it comes to programmatic advertising today, companies are focusing on brand safety as much as they are on impressions, click-throughs and revenue generation, but it is virtually impossible to monitor brand safety due to the scale and speed that programmatic offers.
However, with the increasing adoption of artificial intelligence (AI) in digital marketing, such technology will not only help marketers better target their ideal audience, it might just be the cure for protecting advertiser dollars.
Programmatic Could Compromise Your Brand Safety
Brand safety came under the spotlight following a number of advertising mishaps in early 2017. Alexi Mostrous from The Times revealed that many household brands were unwittingly supporting terrorism on YouTube by placing their ads on hate and Islamic state videos. Later the same year, ads from some of the world’s biggest brands were seen to be running alongside videos that sexually exploited children. This led to widespread panic, with many brands pulling programmatic spend until they could be assured by publishers like Google that measures were being taken to filter out such content.
In 2018, brand safety has broadened to cover any offensive, illegal or inappropriate content that appears next to a brand’s assets, thus threatening its reputation and image. This could include controversial news stories or opinion pieces, as well as fake news or content that is not aligned with a brand’s values – for example, a fast food company’s ad appearing next to an article about heart disease.
In a recent survey, 72 percent of marketers stated that they were concerned about brand safety when it came to programmatic. Also, more than a quarter of respondents claimed that their ads had at some point been displayed alongside controversial content.
Why are brands running scared? Because this definitely has an impact on consumer psyche. Nearly half of consumers are unequivocal about boycotting products that advertise alongside offensive content, and an additional 38 percent report a loss of trust in such brands.
To be fair to brands, they are not choosing to support such content.
Traditional Techniques Come With Limitations
Brands understand the damage that offensive content could cause their image, but it is not feasible for them to implement customized brand safety measures across each ad placement.
Digital advertisers do not have direct relationships with publishers. Ad exchanges receive inventories from thousands of websites that are auctioned off within milliseconds, on the basis of demographics, domain and size of ad. Hence, there are no checks for context or appropriateness, only audience relevance.
While the explosion of programmatic may be offering more opportunities to reach the right audience, the sheer volume also makes it difficult to monitor.
Of course, there are some topic areas that no brand will advertise around – terrorism, pornography, violence, etc. And brands can stay away from content around these by using blacklists, whitelists and keyword searching. However, these have their own limitations.
A blacklist, for example, details individual words that a brand does not want to be associated with – but this ignores nuances and context, letting some ads slip through the net, or blocking placements that may, in fact, be safe.
Finally, safety is subjective. A washing machine brand will have nothing to lose by advertising next to content on prevention of tooth decay, but this could be problematic for a biscuit or chocolate brand.
In the long term, such techniques that are nuance-agnostic cannot completely assure brand safety. Neither can manual methods and checks keep up with the volume, scale and speed that are characteristic of programmatic today.
AI Introduces Context to Content
In this context, artificial intelligence, which offers a solution through algorithms that can understand nuance and context, is fast becoming an answer to marketers’ brand safety woes.
Although AI solutions might not be able to eliminate false positives or avoid the damage entirely, such solutions, specifically those that use machine learning (ML), natural language processing (NLP) and semantic analysis, can offer the nuanced contextualisation that programmatic is lacking today.
ML can ‘learn’ how people approve or blacklist content and then use this to automatically deem content as appropriate or offensive. NLP and semantic analysis assess brand safety at a granular level by understanding the context of a page rather than only look at the keywords or domain name.
Using AI tools that can process large volumes of data at speed to analyze inappropriate placements, advertisers can benefit from the scale and targeting efficiency of programmatic, while avoiding potentially damaging ad placements. Simultaneously, AI can also unlock the potential that false positives undermine by recommending content that brands would otherwise be blind to.
Post the YouTube-debacle, Google confirmed that it was using AI to make YouTube content safe for brands, stating that using ML allowed it to flag offensive content more efficiently and faster than manual methods.
Also, when it comes to brand safety, post-campaign analysis will simply not cut it. Brands have to combine programmatic tools with AI to ensure that the ad placements they are bidding for do not contain content inappropriate to the brand message.
Last but not least, brands should take note that AI tools are only as good as the rules that drive them. Hence, brands must first understand safety within their own context, and what they deem appropriate or offensive. Brand safety rules need to be re-examined periodically as context evolves so they cannot completely do away with human intervention, but AI can help to deal with the sheer volume and scale at which brands operate today.
WE ARE HERE TO HELP
YOU MIGHT ALSO LIKE
Artificial intelligence (AI) is revolutionizing how marketers operate – if you don’t embrace it now, you risk being left behind the competition. Read on to find out how it can make your marketing campaigns more effective and efficient, maximizing your return on investment (ROI). What Is AI Marketing? AI marketing leverages tools powered by artificial intelligence techniques such as machine learning and deep learning to analyze and predict customer behavior in order to market to them more effectively. AI marketing takes many forms, but they all rely on data to reveal customer action or inaction. This data-driven approach to marketing is grounded in reality – how your customers are actually behaving, rather than how you imagine they are or would like them to be – and so generates actionable insights that marketers can use to create more compelling marketing content and campaigns. The Benefits of AI Marketing It is more efficient. AI tools like machine learning let marketers crunch huge amounts of data very quickly, generating actionable insights much faster than analyzing manually, saving your enterprise time and money. It is reliable. With a grounding in hard data, you can rest assured that the results generated by AI
Regression analysis is an incredibly powerful machine learning tool used for analyzing data. Here we will explore how it works, what the main types are and what it can do for your business. What Is Regression in Machine Learning? Regression analysis is a way of predicting future happenings between a dependent (target) and one or more independent variables (also known as a predictor). For example, it can be used to predict the relationship between reckless driving and the total number of road accidents caused by a driver, or, to use a business example, the effect on sales and spending a certain amount of money on advertising. Regression is one of the most common models of machine learning. It differs from classification models because it estimates a numerical value, whereas classification models identify which category an observation belongs to. The main uses of regression analysis are forecasting, time series modeling and finding the cause and effect relationship between variables. Why Is It Important? Regression has a wide range of real-life applications. It is essential for any machine learning problem that involves continuous numbers – this includes, but is not limited to, a host of examples, including: · Financial
VIPs are some of the most profitable customers for e-commerce players, but keeping them loyal and engaged is an ongoing challenge. With emerging artificial intelligence (AI) tools, however, it is now possible to understand your VIPs better, predict their behavior and create hyper-personalized messaging to win their hearts and wallets. Consumer spending can be said to follow the 80/20 rule, meaning that on average, 80 percent of your profit comes from 20 percent of your customers. This 20 percent are your VIPs who represent the highest customer lifetime value (LTV) or profit potential. To ensure these VIP customers spend more and stay loyal, e-commerce companies usually offer special incentives, such as free delivery, early access to sales, invites to special events or birthday vouchers. While these measures can be effective, they are sporadic. Implementing them more often is simply not feasible or practical, and can even damage your brand perception. In order to engage VIPs more effectively, you need a different strategy – one that not only delights on infrequent occasions, but also on a daily basis. This is where AI can help you better target and engage your existing VIPs, and even create new ones through hyper-personalization and prediction.