An analysis of my article on Social Media Today
I was recently delighted when Social Media Today accepted an article I wrote about a Twitter experiment on engagement — it’s on SMT here.
As we all know, distribution is a key aspect of having content read, or at least mentioned in some way, to help it find an audience. It was interesting to watch as my article went through their large network. This is an analysis of what I observed over the 72hrs after it was published.
As with the first article, this is purposely low-tech. It’s not from a macro data download or from an all singing all dancing analytics engine. I wanted to get a personal feelabout what was happening by reading the threads in sequence and not be distracted by additional or collated information.
So, here we go…
72hrs in the Life of a Twitter Article
The total number of tweets of my article grew quite steadily over the time frame.
Okay, this doesn’t say much; let’s see how it breaks down:
Within the first hour it had received 76 tweets representing 27% of the total it would garner over the three day period.
Right, so the shares were front-ended upon publication. Makes sense. Here’s how it did for the rest of the time:
Whilst the largest bulk happened within the first few hours, within the first 24hrs it had received 68% of the tweets it would. After 72hrs both tweets (and other forms of sharing) had crashed to a nominal level.
Following initial publication, the distribution was extended and perhaps gained further eyeballs by tweets from the Social Media Today Twitter account (which has an impressive 321k followers):
Tweet Variant #1 from SMT
The tweet structure was the title of the article, with a widely used hashtag (perhaps not interest specific enough).
It’s worth mentioning that tweets originating directly from the article were also generating a significant number of shares during this period, perhaps as many as 70%. As a caveat to this, it’s not possible based on the available data to know how many of those people originally found the article due to the tweet.
Whatever the true picture, it didn’t look like this tweet was driving the shares in isolation or incrementally improving the distribution during the post 24hr period, although it did appear to help maintain it.
Tweet Variant #2 from SMT
It became hard to track after this using Twitter, however you can see that the reach was diminishing. It could be that the natural lifecycle of the article, the more ambiguous copy in the second tweet, or (most likely) a combination of both were inhibiting it’s being shared further.
This demonstrates just how quick the news cycle is; as well as that the lifecycle of any content — even published on large distribution networks — is limited. Time waits for no writer in the Twitterverse!
A way to overcome this of course, is to utilise multiple distribution networks. As this article was an exclusive, I didn’t do that, although I will post it on my Medium page once the exclusivity has expired to give it a second life.
Let’s try to track the geographic reach of my article by the tweets it inspired. Just for fun, around the world in tweets we go:
Admittedly not the nicest looking picture ever…
The widest geographic distribution by far came from the initial publishing of my article, as the large network of SMT sprung into action. However, both of the subsequent tweet variants supported some dispersed activity, although the volume naturally reduced as time progressed after publication.
In total, the article gratefully received tweets in 33 countries (that I could see).
The Final Result
Tweets represented 75% of the total Social shares of my article. It seems that not all sharing is created equal, with a quite low level of sharing on other social networks.
The ratio of shares across all types was: Twitter 43%, Email — 50% , LinkedIn —5% , Facebook — 2%, Google+ —ahem. I’ve had a look at a few other articles on SMT and this split seems relatively normal, although other networks do seem to score a little higher as an average. The article was about Twitter so it makes sense that it might not inspire people to share it elsewhere.
What this does say about how we use Twitter differently to the other networks is that it’s really easy to push out tweets and the nature of it means we often don’t think twice about tweeting. It requires less thought if something is within our sphere of interest to share it. Whereas, perhaps there is an extra layer of decision before posting it on narrower networks.
Watch Out For The Sharing Event Horizon
Not strictly to scale…
In particular email sharing seems something of a black hole. It’s likely tough to attribute and have visibility of the people that were sending, to where and for what reason.
Surprisingly, the high number of emails suggest that email shares are still quite widely used.
It’s again highlighted to me the importance of creating and distributing content with open eyes, whilst tracking the crap out of it. In addition:
These points are, of course, more important for companies looking for ROI than individuals. However, good data unlocks the obscured reality for all.
All in all I was happy with the lifecycle of my article considering that it was published in such a high volume and content rich location, although longer would certainly have been nice. Some final thoughts for your consideration:
What’s the lifecycle of your content and how do you extend it?
If you liked this article please share it!