3 ms·

It's from his email: https://twitter.com/TechEmails/status/1590750376118341632 https://twitter.com/TechEmails/status/1590750376118341632 "Performance advertisi

It's from his email: https://twitter.com/TechEmails/status/1590750376118341632 https://twitter.com/TechEmails/status/1590750376118341632 "Performance advertising" is a marketing term, as distinct from "brand advertising". Brand advertising: feel good stuff, like the Coca-Cola polar bears. You can't measure an immediate impact, you just saturate the market with it to ensure you're top of mind. Brands often have a big pool of money to spend here, and are somewhat indifferent to whether they spend it on Twitter, Facebook, Youtube, etc, etc, so long as they hit certain metrics about demographics and reach. Performance advertising: you're trying to get a specific action from the user. You often build these campaigns to be ROI positive. For example, you spend $100 to get clicks on "house insurance" ads, you expect to get $200 of sales. All else being equal, advertisers prefer to run ads on sites where they're not adjaced to controversial things. For performance ads, you might swallow your distaste and tolerate the ad being next to controversial things -- after all, you can directly measure if they're producing the ROI you want. For brand ads, it's much squishier, so you're more likely to take your spend somewhere "brand safe".

legitster 4y ago
A bit of an oversimplification, but generally true. Depending on your market, you may love controversial topics as it gives you more niche targeting. It's just that broad consumer brands with the deepest pockets are so diversified that they don't need any particular platform. There's also an open secret in advertising that most of this money is wasted. Coca-Cola doesn't nearly need all the advertising it gets - but the entire brand advertising industry engages in heavy rent-seeking. So when Musk says that "activists" ran advertisers off of the platform (despite never being able to run them off of Google, Facebook, etc), I think there may be some truth to it. Most advertisers never found much value on Twitter, and largely only because the admins themselves were proponents of Twitter. The irony of Musk running the bots off of Twitter is that they consisted entirely of his audience who would have been interested in performance-based advertising.
- moistly 4y ago
  > "activists" ran advertisers off For that definition of "activist" that means "people dropping the n-bomb all over twitter," I suppose. In all seriousness, though, it is in fact hate speech that has driven advertisers away. It's not at all difficult to measure using keywords and sentiment analysis, and what has happened is an immediate order of magnitude more hate speech after Elon took over. That kind of environment does not attract conventional advertisers. > ...a range of terms were examined to see if frequency of use increased in correlation to the change in ownership and stated platform focus. The terms included vulgar and hostile terms for individuals based on race, religion, ethnicity, and orientation. Collection of the totality of Tweets using these words in the period immediately before and after Musk’s acquisition was done to assess both frequency and the rate of with which this content was posted. Tweetbinder’s sentiment analysis tool, which measures positive or negative tone in Tweets, was utilized to assess potential hostility in the collected data. > Specifically, the seven-day average of Tweets using the studied hate terms prior to Musk’s acquisition was never higher than 84 times per hour. From midnight on October 28 to 12 noon, however, hate terms were Tweeted some 4,778 times. Calculation of potential reach (i.e., the potential number of times a term posted in Twitter could have been viewed) was in excess of three million. As such, this research suggests that there was an immediate, visible, and measurable spike in hate terms on Twitter after Musk took over as CEO. https://www.montclair.edu/school-of-communication-and-media/wp-content/uploads/sites/20/2022/10/Montclair-State-SCM-Study-Increases-in-Twitter-Hate-Speech-After-Elon-Musk%E2%80%99s-Acquisition-1.pdf https://www.montclair.edu/school-of-communication-and-media/... https://www.montclair.edu/newscenter/2022/10/31/study-finds-hate-speech-on-twitter-increased-following-elon-musk-takeover/ https://www.montclair.edu/newscenter/2022/10/31/study-finds-...

legitster 4y ago

A bit of an oversimplification, but generally true. Depending on your market, you may love controversial topics as it gives you more niche targeting. It's just that broad consumer brands with the deepest pockets are so diversified that they don't need any particular platform. There's also an open secret in advertising that most of this money is wasted. Coca-Cola doesn't nearly need all the advertising it gets - but the entire brand advertising industry engages in heavy rent-seeking. So when Musk says that "activists" ran advertisers off of the platform (despite never being able to run them off of Google, Facebook, etc), I think there may be some truth to it. Most advertisers never found much value on Twitter, and largely only because the admins themselves were proponents of Twitter. The irony of Musk running the bots off of Twitter is that they consisted entirely of his audience who would have been interested in performance-based advertising.

moistly 4y ago
> "activists" ran advertisers off For that definition of "activist" that means "people dropping the n-bomb all over twitter," I suppose. In all seriousness, though, it is in fact hate speech that has driven advertisers away. It's not at all difficult to measure using keywords and sentiment analysis, and what has happened is an immediate order of magnitude more hate speech after Elon took over. That kind of environment does not attract conventional advertisers. > ...a range of terms were examined to see if frequency of use increased in correlation to the change in ownership and stated platform focus. The terms included vulgar and hostile terms for individuals based on race, religion, ethnicity, and orientation. Collection of the totality of Tweets using these words in the period immediately before and after Musk’s acquisition was done to assess both frequency and the rate of with which this content was posted. Tweetbinder’s sentiment analysis tool, which measures positive or negative tone in Tweets, was utilized to assess potential hostility in the collected data. > Specifically, the seven-day average of Tweets using the studied hate terms prior to Musk’s acquisition was never higher than 84 times per hour. From midnight on October 28 to 12 noon, however, hate terms were Tweeted some 4,778 times. Calculation of potential reach (i.e., the potential number of times a term posted in Twitter could have been viewed) was in excess of three million. As such, this research suggests that there was an immediate, visible, and measurable spike in hate terms on Twitter after Musk took over as CEO. https://www.montclair.edu/school-of-communication-and-media/wp-content/uploads/sites/20/2022/10/Montclair-State-SCM-Study-Increases-in-Twitter-Hate-Speech-After-Elon-Musk%E2%80%99s-Acquisition-1.pdf https://www.montclair.edu/school-of-communication-and-media/... https://www.montclair.edu/newscenter/2022/10/31/study-finds-hate-speech-on-twitter-increased-following-elon-musk-takeover/ https://www.montclair.edu/newscenter/2022/10/31/study-finds-...