We love Tumblr and we especially love the diversity and creativity that goes into making it the wonderful place that it is. That means we know that not everyone works at a company with the budget for these plans. We’re working on more analytics offerings for everyone, including a version for anyone interested in measuring their personal blogs. That’s coming soon, so stay tuned!Union Metrics’ Tumblr analytics system — which uses the Tumblr firehose — is now available, though most users can’t afford the $499/month starting price (most companies totally can, though). The company says they’re working on it.
Did you see those silly cats on Tumblr, that breaking news on Twitter, and those photos of your friend’s kids on Facebook? Different social networks have their own distinct personalities. Bitly links are shared across all social networking services, giving us a unique viewpoint on how these…
This is fascinating, but I think it gives very little attention to the most important part of this, which is: When are the people that you’re trying to reach most active on that service? The fact that Tumblr gets a lot of sharing use in the evening hours isn’t surprising, because it has a high concentration of younger users. But those might not be the people you are trying to reach; Tumblr also has a good audience reading it at work, and if you’re targeting those people you’re probably better off posting earlier in the day.
“Friday evening, a no-man’s land on other platforms, is an optimal time to post on Tumblr.” This actually corresponds with what we’ve noticed. A lot of mornings might be a little slow for us, but post something at 10 p.m. on a Sunday and it blows up. Fascinating trend.
inothernews says: I think we broke it.
insaniyat asked: Too much exposure, too fast? It’s not working for me :(
» SFB says: It’s very finicky about addresses. When setting yours up, type in your address using the format “yourname.tumblr.com” without a slash, even if you have a custom domain. It will come up with an error if you don’t. Either way, it’s still pretty beta, but it looks cool. Only real downside is that the information getting analyzed is public, so just know that going in. — Ernie @ SFB
For example, our notes-to-posts ratio is 113:1, based on an early sample. Also, we have over 12,000 posts on Tumblr. Which is a pretty useful to know, right?
Fashion is one of the platform’s most-popular industries, and they just got another major brand on the service today: Calvin Klein. Karp gave the keynote at the Decoded Fashion one-day conference. Next question: When will end-users get access to such analytics?
Google Ripples: Concentrated awesome. This is the future of analytics right here. You can get as broad or as granular as you want, and in a completely non-linear way. Imagine if Tumblr did this — a map of your reblogs. That would be some bad-assery right there. But Google got there first, and so they get the high-fives.
NewsBeat opens in public beta: We’ve tried a few analytics services in our day, and one of the ones we tried was ChartBeat. We liked it, but it perhaps wasn’t granular enough for us — particularly because it mainly gave a snapshot of the specific moment — making it hard to track data from a contextual, historic perspective. (Our current favorite is Clicky, which is like Google Analytics on steroids.) Apparently realizing that the tool wasn’t the best for journalism, they went back to the drawing board and created NewsBeat. Looks pretty cool and seems to work around some of the problems we had with the earlier service. Might try it out.
According to our awesome real-time analytics software, a Fox News Channel employee has looked at this particular post. (Also, Fox News apparently forces its employees to use Internet Explorer 7. Upgrade to Chrome, jerks!) We hope it’s Bill O’Reilly. Paging our anonymous tipster!
Why people unfollow @breakingnews: We’re lucky enough to have access to a beta analytics tool from Twitter that lets us see which tweets on @breakingnews trigger more “unfollows” than usual. Any guesses? Surprisingly, it’s not the occasional tweet about Lindsay Lohan’s legal troubles (which we try to keep to a minimum). But it tends to happen when we tweet in rapid succession, or when we post a local news story about a crime or tragic accident.
For example, we sent 4 tweets over the course of 5 minutes earlier this week while Ghadafi’s son was speaking to the nation in Libya. Each of those tweets triggered more unfollow events than usual. A school bus accident in Arizona and an Amtrak-pedestrian accident in New Jersey both sparked more unfollows, as well.
This lines up with a recent survey of our @breakingnews users. “One person shot in Seattle? I don’t care,” wrote one user urging us to cut back on local stories. “Why is 95% of the stuff you share so negative?” added another. “I know that’s the news in general, but geez, can you move past kids shooting each other to bigger pictures?”
These are all great points. Sure, it’s impossible to make everyone happy — as evidenced by the often contradictory replies we see moments after we send out a tweet — but certainly there are common themes where we can improve our coverage. Stay tuned, and keep the feedback coming!
Fascinating to see how Breaking News is using analytics to figure out the kind of news people want to read. Unfollows = lessons. (Also, how do we get access to this analytics app?)
Copyright 2009-2013 Ernie Smith • Ask us stuff! • E-mail us • Follow us on Twitter • Follow us on Facebook
