I’d like to have a look at yesterday’s Twitter traffic data from a different angle.
As I wrote in the piece, February is a short month, and this definitely accounts in part for the drop in Twitter.com’s month-by-month numbers. Furthermore, Compete’s data is very US-centric, and with Twitter.com possibly now accounting for less than 20% of all Twitter usage (different source here), it should, as I’ve consistently mentioned, be taken with a pinch.
But it still has value. Even with February being a shorter month, Twitter.com’s numbers were, at best, flat. Indeed, the website hasn’t really gained any ground with new visitors since June 2009, continuously hovering around the 22-23 million uniques mark ever since reaching that peak.
So, what about daily averages? Using Compete’s data, I’ve put together this chart.
Again, this is US-centric, but as we can see Twitter.com hasn’t seen any day-on-day growth since June. Indeed, it seems almost as if the website has become pegged against its users, like a currency. Is it consolidating for the next move upwards, or have we already seen the highs?
What about other sources of Twitter data? Let’s take a look at Google Trends.
For all mentions of Twitter:
(see here for an explanation of the reference points)
What about mentions of the word tweet?
Finally, you may recall that Twitter recently edged up to 50 million tweets per day.
The problem is, if we assume that the network has around 25 million users, that’s only two tweets per user, per day. If it has 75-100 million or so as some have speculated, it’s a lot less.
It’s all a bit of a mixed bag. As we can see from the Google Trends data for ‘twitter’, there’s a definite up-channel there, but it’s fairly measured, whilst being prone to pretty sharp corrections downwards.
It’ll be interesting to see Twitter’s numbers for March. One would think that if the Compete data could break above the 25 million area, and the other indicators also moved higher, Twitter likely will have emerged from this dormant period and climb upwards throughout the rest of 2010.
If, however, March sees another dip, a more disturbing trend may be in development. Even if Twitter’s daily tweet totals continue to edge upwards, that simply means that existing users are tweeting more, but even thatÂ doesn’tÂ really matter all that much – even an extra half a tweet a day per user would have a massive impact on the overall figure.
As with anything like this, it’s very much a case of ‘wait and see’, and likely we’ll need a few more months under our belt before we can make anything close to a confident prediction of what the future holds. But one thing is increasingly clear: Twitter.com as the favoured portal for Twitter interaction appears as if it predominately appeals to first-time and very inexperienced users. And even that is fleeting, as those that take to Twitter often quickly move up to (the admittedly superior) clients such as TweetDeck, Seesmic Desktop and HootSuite.
The main website still has lots of hardcore, veteran and power-users, too – it’s a daily part of my Twitter usage, alongside several other clients – but the bulk of users favour getting their fix elsewhere. And overall, and like-for-like, Twitter.com is definitely stuck in the virtual mud.
One could argue that Twitter, Inc does not care – usage is usage, after all, assuming the overall numbers continue to rise – but this would somewhat clash with Twitter engineer Alex Payne’s comments about planned improvements to the home page that he madeÂ last month. Even if he subsequently backtracked (and deleted his tweet), clearly something is afoot, and if Twitter wants to attract advertisers and business it makes sense that the core product should be able to (at least) compete with corporate-friendly packages like HootSuite.
Perhaps this re-modelled, brand-pleasing version of the main site will be the premium version of Twitter that we’ve long been promised. Suffice to say, unless we see some significant traffic improvements, and soon, they’ve very much going to need it.