It’s been a few weeks away from the Experiment, what with operations and the like, but I’ve not forgotten the task in hand… except some rather important things have changed since the last time we looked at stats. They are nothing to do with me, and everything around how Twitter itself does business, and they have had a significant affect on the client’s reporting tools at a basic level.
I knew this blog post on the 17th was coming well before it arrived. The standard infographic data stopped working for a few days beforehand. Tweets would not register any hits, the real-time reporting went tits up and some key figures were royally screwed.
As you can see, I didn’t gain 3389 followers this month, but for some reason this is is how the details on my account have been ‘reset.’ You can work out when that was: that dip in followers corresponds in the 55% and 47% dip in profile visits and mentions. Something major happened mid month that (I suspect) may also have removed a lot of bot accounts and automated followers, as the number of random new people I used to get as a result of using hashtags in posts has effectively vanished overnight. The meat of the Twitter blog is however far more concerning: focussing on advertising data, syncing accounts across devices and effectively tailoring your ‘advertising’ experience if you don’t have the good sense to opt out.
The reality of my engagements over the last couple of weeks is that, like it or not, I’ve just not been trying that hard. Mostly that’s been because of recovery from my operation: there was also a Tweet last week that acted as a salutatory reminder that reach is not everything. I made a comment on a message by Duncan Jones (@ManMadeMoon) which he then subsequently retweeted. He justifiably commands a massive online audience, but getting them to interact?
So I manage 38k views, and only 17k of those are self-generated. The rest are that one tweet, that garnered a less than 1% engagement rate. I can’t even be confident that these numbers are correct, considering that I didn’t do anything last week, according to my metrics. Until there’s confidence that what I’m seeing is in fact correct, it is probably an idea to not place too much store in what I’m presented.
It’s fine, there’s plenty to do in the meantime instead, including addressing a significant issue with a tool I was beginning to enjoy using:
CoPromote has been broken since April 29th, and the last active tweet from the company was on May 13th. If I were a betting woman I’d say that that interruption I saw happen pretty much put pay for good to this form of promotion, which made money that should really by rights should have been Twitter’s to begin with. It reminds me to go and see if my credit card was charged for May, and if it was to make sure that before June that amount is denied. I’m betting that we’ve seen the last of this Company, and that other such ventures will be quietly yet firmly shut down by Twitter going forward.
It means, starting in June, that there’ll need to be some serious consideration on how I use Social media for effective promotion.