For provocateurs of social media data analysis, the most interesting trend on election night actually have happened after the race was called.
As of 2pm November 13, 2012, the Tweet garnered more than 815,000 retweets.
We can take these numbers at face value, but the most salient questions should be, “Are these numbers good?” and “What does it mean?” and “How do you benchmark against these?”
Context is the key to understanding relative performance.
A for Amplification
If we are looking solely at the high number of retweets, focusing on amplification is vital.
Amplification is the percentage of a fan base that engages with content via a retweet. This metric helps measure how engaged and interested fans are in content, based on whether they want to share it with their own audiences. To get a sense of amplification, compare the number of retweets to the overall reach of those tweets (i.e., the combined number of followers for all retweeters).
Though there isn’t access to the reach numbers, use followers as a proxy to examine the amplification percentage: As of November 13, President Obama has 23.3M followers and approximately 3.5% of them amplified the election night tweet. Let’s compare this to the previous record holder, @justinbieber (30.2MM Followers).
Following the same rules as the President’s official Twitter account, this post was amplified by only 0.74% of Bieber’s followers.
The President looks pretty good compared to the Biebs.
For the sake of context, look at one more example from @TJLang70, an NFL offensive guard on the Green Bay Packers who has 124,836 Followers. In September, the Packers lost a football game on a controversial last-minute touchdown call made by replacement officials.
He tweeted his disappointment following the game receiving 98,680 retweets, good for third all time, which means 79% of his followers amplified the message.
Granted, the collective amplification would look different using actual reach numbers in the calculation, rather than followers, but there are overarching key learnings that can applied to social data analysis:
- Big numbers don’t necessarily mean big results
- Ask questions when analyzing social data, because even small numbers can be meaningful
- Add the missing layer of context to foster greater understanding (i.e., dig deeper into the metrics rather than just analyzing follower numbers or the number of re-tweets, understand a community’s voice, interests and user habits)
- Know the fan base; all three examples above were either timely or directed toward a rabid base of fans
- Apply relevant benchmarks to understand performance
Photo Credit: Barack Obama