Click the above image for the interactive version
We’ve been looking at characteristics of break-out tweets and memes. All memes are NOT the same. One interesting approach is to use our Salt library to plot time (horizontally) and number of retweets (vertically). This creates effectively a line chart made up of millions of lines. Each line grows from left to right as it is retweeted. Some tweets take off as a vertical line, some tweets take a while to grow and the vast majority of the data is stuck at the bottom never breaking out.
Instead of looking at Trump, we decided to look at tweets surrounding a commercial event - and what bigger commercial event is there than the Super Bowl! The visualizations above and below show retweets on Feb 7, 2016 leading up to and during Super Bowl 50. You can click any line to see the tweet. One interesting observation is that you can find A LOT of commercial activity surrounding Super Bowl 50, some humour, and very few offensive tweets (unlike some other Twitter analyses we’ve done).
Within a few clicks, you’ll probably find one of Esurance’s tweets “retweet for a chance to win $50K,” such as this one - interestingly, quite a few people re-tweeted this before the start of the game, in spite of instructions to retweet during the game, although most people waited until game time as seen by the big jump at 6:30.
Some tweets take off quickly but then plateau pretty quickly too. A good joke or a great photo sent to your fanbase. Other tweets just keep building - how about Betty White - that’s the kind of awareness marketers must be hoping for.
One interesting aspect to all these lines is that they are a lot like “accumulating graphs” that are used in various applications. Financial traders, for example, might compare accumulating volume over the course of a day compared to historic patterns to estimate how the volumes will continue to grow. Same approach is relevant for Tweets - if you’re a marketer, you want the growth pattern of Betty White - which suggests ongoing relevance on longevity to your message as opposed to the growth pattern of Coldplay - a quick flash, quickly forgotten. This approach has many great applications - e.g. look at the twitter chatter before a movie opens and during the reviews to predict movie performance; or the launch of new product introductions; or the stickiness of multiple ad campaigns.