Since the end of December 2014, Twitter has been rolling out an analytics tool for its mobile app.
This is fantastic news for those of us who are out and about, live-tweeting hilariously named brands in the international supermarket or a joyless bus-ride and wondering if anybody actually cares.
If you haven’t spotted where to find it, it’s just below the text once you’ve clicked through to the body of the tweet.
Clicking this link leads to the following screen, with quick, easily accessible insights on impressions, engagement, and whether people clicked on any links in your tweet or your profile itself.
This is essentially a mini, simplified version of its desktop analytics tool that can be found in the drop down menu on your profile thumbnail.
Or alternatively you might see ‘Twitter Ads’ here, in which case click the link…
Then head to the top left and click ‘Tweet activity from the drop-down menu’.
Previously this was only available to those with an advertising account, but since August it has been made available to all users. Hence why you may not have spotted it before.
Last July, our head of social, Matt Owen, wrote a handy walk-through guide as to what insights users can expect, including a focus on impressions, strength of content and multichannel measurement.
Here I’m going to take a look at the analytics from my own personal account and one that I run for a music blog, to see if there’s anything I should be learning from the data. If anything else it may stop me from harassing Kasabian at 4am on a Sunday morning. (3pm on a Wednesday is obviously much better).
Blindingly obvious point alert: being better at Twitter = more followers
Although I’ve had a Twitter account for approximately six years, I was pretty lacklustre at tweeting. When I joined Econsultancy 16 months ago, I had around 100 followers. This was laughable compared to my colleagues’ legions of 4,000 to 10,000 follower counts.
Of course it’s not about the size, it’s about the engagement. Any brand or individual with 100,000+ plus followers can look mightily impressive, but if you’re not providing interesting or entertaining content or information that people are interacting with, you might as well have zero followers.
You’ve all heard this famous mantra of content and social strategy a thousand times before, but here’s what you don’t hear that often… it sucks if you’ve only got a handful of followers. In fact it even hurts a bit. Yes it’s a vanity metric once you get over a certain amount, but if you’re working as a writer for a digital marketing company, than it’s kind of embarrassing.
So I tried a few things with my accounts to improve matters: tweeting more often, sharing articles I’ve written on the blog, tweeting Instagrams directly from my Instagram app (a rookie mistake).
After a year I doubled my following to… 200. Yeah great. Clearly my heart wasn’t in it.
In September I made a concerted effort to turn things around.
Change your Twitter handle if it’s rubbish
The first thing I did was change my Twitter handle. Possibly a dangerous thing to do late in the game, but being as I hardly had much of an empire to oversee, this mattered very little. Plus my original handle was awful: @christopherRCLF. Part man, part illiterate football club.
I changed to the much more dynamic @Christophe_Rock, which despite my lack of Frenchness or similarity to comedian Chris Rock, at least meant I had a slightly more memorable name which was composed of actual meaningful words.
The next thing I did was lessened how much I broadcasted my own articles. This may seem counterintuitive, but all I was doing was automatically tweeting my headlines plus a link without any kind of optimisation or tailoring. People didn’t care anymore and were ignoring what was essentially an RSS feed for my own work. Even I was getting bored.
By only tweeting specific articles sporadically that I was genuinely proud of, it meant I had more of an emotional investment and worked harder at tailoring the tweet used to promote it.
Other obvious yet vitally important points
I also began tweeting more often. Following more people. Varying my types of tweets. Sharing more articles, videos and images from a wider variety of sources. Retweeting and actively talking to the people I followed… Heck, actually enjoying using Twitter.
This is nothing more than the basic, honest-to-goodness stuff that makes Twitter work. Again this isn’t anything new and may be blindingly obvious to most. but… the results work.
Here’s how my personal account’s following has increased in the four months since ‘being better at Twitter’…
And here’s how my music blog Popdin’s account has increased…
It’s heartening stuff for someone who sucked at Twitter a few months ago.
Here’s another secret though, I still suck at Twitter.
Stop being lazy
For Popdin, I know I’m not tweeting nearly enough to take advantage of the 2,500+ followers I’ve accrued.
Looking at the site analytics, a tweet publicising an article tends to direct around 80 people to the site, which is 3.2% click-through rate, so not too shabby at all. Then again it’s a niche site that has a Twitter account with followers who are obviously interested in that particular niche, otherwise why would they follow? I don’t have to ‘hit and hope’ as much as more general blogs or publishers.
So why don’t I tweet more often? Why don’t I retweet popular articles that I’ve published in the past? Why don’t I engage more with my followers?
To be entirely hand-on-heart honest, it’s two-parts time constraints and one-part laziness. That last part I can definitely rectify with a simple self-kick up the arse. The first one I can fix by allotting a few minutes at strategic points during the day to find some relevant content to share and commenting on similarly relevant music news.
It’s really not a lot of effort, and if I have this built into my diary then I don’t have to suddenly worry about it at weird times of the day or at 4am when I wake up in a cold sweat (which is when Kasabian normally bear the brunt of it).
Let’s see what else I should learn from the last few months worth of data.
@mentions of influencers can work (if done well)
What accounts for this sudden spike in 11,269 organic impressions?
Is it that I have a huge number of hip-hop fans in my following? Probably not, it’s in fact down to the frankly rampant @mentioning of various related Twitter accounts with large followings. This was retweeted by the artist @therealelp himself, who is an inveterate retweeter.
Knowing your audience obviously helps, but also knowing the right influencer to target can provide a huge boost. Kanye West is never going to retweet your self-produced mixtape, however a smaller artist more in touch with his fanbase will respond favourably to a positive mention of his gig you attended.
However this tweet didn’t direct anyone to my own personal site, and only managed to succeed in generating one new follow out of 11,691 impressions. It’s far from a blistering success.
But then that wasn’t the point. The real point was just to provide something of value to the followers, give some credit to an artist I admire, and promote some positive brand perception.
Also so does knocking U2. With an engagement rate of 5.8% this is one of the more successful tweets of the last few months.
‘Engagement’ also includes the number of times people have clicked on the image in order to expand it, so the teasing letterbox format of the image worked well here.
Images really work
Making your tweets more visually interesting obviously makes them stand out from the crowd, but when you look at the analytics it becomes all the more clear.
There’s a definite jump in engagement in posts that feature an image. The one of Mr Burns didn’t even contain any text to contextualise it. (I don’t recommend this).
Times to tweet
Looking at my own personal account, which I tweet a lot more from than my blog account, I can see which times and days are more likely to see engagement.
On Monday, just before lunchtime around 12:50pm, managed 815 impressions, which is about half my following impressively. Anything before this is dead time, as everyone is focusing on getting their head back into work mode. However as soon as lunchtime approaches, that’s the time to start providing an escape.
Saturday lunchtime is bad. Just go back to bed, or do the thing that you’re supposed to be doing instead.
Surprisingly I have managed some fairly high engagement at 1am on a Saturday, but they may say more about my followers and me than anything else.
Ultimately though my analytics tell the common tale that tweeting content between 11am-12 and 2pm-4pm on Tuesday, Wednesday and Thursday, will engage followers the most, particularly if it’s within a couple of minutes before or after the hour.
If you’re a brand or a publisher, you should definitely look at your analytics. It will provide so much clear, concise information about when to promote your content and whether your audience cares or not. It’s invaluable data that you need to use, not just for driving traffic to your own website, but also in providing the right content for your audience, therefore improving their experience.
If you’re a regular Joe just like me, it’s probably best if you don’t dwell too long in your analytics. It’s primarily important just to be yourself and not fret too much about ‘who you want to be’ on Twitter. Sounds glib, huh?
Looking at the best times to tweet, and whether or not you should be commenting on a news event even though you’re primarily a fashion blogger will just lead to anxiety. Your Twitter feed should be an accurate account of who you are and what you are feeling, followers can smell a fake from a mile off.
Now with all that said and published, I’m going to sit here and watch my follower count take a nose dive.