Robin Goadâ€™s post yesterday on what a Google-Twitter marriage might mean has some terrific data on Twitter traffic, like this chart which shows the growth in UK visits to twitter.com over the past year and especially in the first quarter of 2009.
That Q1 2009 line is dramatic, isnâ€™t it? I bet this is reflected in many other countries, not only in the biggest market, the US [see update, below].
While Robin discusses a range of extremely useful stats from Hitwise research on Twitter downstream traffic, this statement especially caught my attention:
[â€¦] I should also add the usual caveat: the service is probably even more popular than our numbers imply, as we are only measuring traffic to the main Twitter website. If the people accessing their Twitter accounts via mobile phones and third party applications (such as Twitterrific, Twitterfeed and Tweetdeck) were included, the numbers would be even higher.
Thatâ€™s exactly what I always wonder every time I see stats on Twitter traffic where those stats are measuring growth: What about mobile devices and 3rd-party apps on a computer?
How I use Twitter is probably typical of many, certainly among many users I know â€“ I rarely go to the Twitter website, and donâ€™t use the site for actual use of Twitter, eg, sending tweets, â€˜@â€™ replies and direct messages.
Thatâ€™s when Iâ€™m at my desk. If Iâ€™m out and about, or even wandering around my house, I’ll use Tweetie on the iPhone or Twibble on the Nokia N95 8GB depending on which mobile device I have to hand.
Whatever actual application or device I use, the point is that probably close to 95% of my use of Twitter is via 3rd-party applications that donâ€™t involve me visiting the Twitter website.
So the traffic by people like me wouldnâ€™t show in data like Hitwiseâ€™s reporting. Project that to the downstream-from-Twitter stats â€“ nothing included there, either.
Now Iâ€™m not suggesting that Iâ€™m wholly typical of all users. But if, say, 30% of people use Twitter like I do â€“ and I think 30% is a conservative figure â€“ then you can imagine what Robinâ€™s charts would look like if those numbers were taken into account.
[Update:] Stats from web analytics company Compete.com on visits to twitter.com by Twitter users in the US covering a similar period as Hitwiseâ€™s stats for UK visits show almost a mirror-image growth trajectory:
These US figures come today in a post by Nick Oâ€™Neill, who says that Twitter has surpassed 14 million users, up over 76.8% since last month when he wrote that they had surpassed 8 million users.
While the figures are expressed differently to those from Rob Good â€“ his chart shows percentages while the US stats talk about actual visitor numbers â€“ Oâ€™Neill reckons itâ€™s realistic to estimate that Twitter will end 2009 with 50 million active users (although he doesnâ€™t define what â€˜active usersâ€™ might mean).
Based on all the numbers you see at the moment, such an estimate does look credible.