I’ve been watching the controversy and community panic around the Tr.im shutdown and re-emergence with a bit of bemused detachment.
The problem stemmed from the fact that Tr.im couldn’t seem to think of any way to make money with their URL shortening service, despite the fact that they’re pulling in close to a million monthly unique visitors, and ostensibly even higher pageviews.
In one of their farewell posts, Tr.im boo-hoo’ed a little bit over the fact that they just couldn’t make money with a URL shortener, and I know I’ve seen more than a handful of pundits parrot the sentiment that there is simply no business model for URL shorteners whatsoever.
This, of course, is silly, but we’ll get to that in a minute.
Late last night, I caught a short post from Mike Elgan’s Raw Feed, posing a modest proposal: Twitter should just build their own URL shortener, a move that comes with a variety of upsides:
Twitter’s 140 character limit has got users scrambling to find third-party URL shorteners. You have to copy your URL, find your shortener, paste in the URL, copy the shortened URL, go back to Twitter, paste in the shortened URL. (Or, you have to add yet another Firefox plug-in.)
Another issue surfaced this month when Trim announced that it was going out of business, which would have meant the breaking of user links. (Trim has since announced that they’re staying in business after all.)
Twitter will often change and shorten URLs if they’re already short enough to fit into 140 characters. Why not accept URLs of any length as long as the words plus the shortened URL fit?
Of course, this wouldn’t work sending tweets directly via SMS. But who does that? Everyone’s either using a Web page or a cell phone app. If the URL shortening service was as open and extensible as Twitter itself, the phone apps could tap into the URL shortening service as well.
If Twitter suddenly got an attack of conscious when it came to preserving the integrity of the link economy, why not just rewrite all the URLs on the system with their own short URLs – it’s the only way to ensure the links will continue to exist.
Beyond Mike’s reasoning and the moral debate of Twitter taking all other URL shorteners out of business, I can think of another reason why Twitter should do this – 400 million reasons, in fact.
It’s time for some extrapolative math…
According to Tweespeed, a service that attempts to graph the amount of Tweets that race across Twitter per hour, on average, there are 800,000 tweets per hour on the service (with peaks at around 1.2 million per hour, and valleys around 400,000 per hour).
According to the various studies I’ve read, which are imprecise on this particular datapoint, the percentage of tweets containing a link (shortened or otherwise) sits at around 5-18% of all tweets, which would equate to a range of 20,000 to 324,000 links per hour, or an average of about 172,000 links per hour.
Based upon the extensive usage patterns I’ve monitored from users of my personal URL shortener over the last six months or so, the average link posted to Twitter gets about 40 pageviews, which bears out regardless of the size of the user’s social graph. Obviously, users with larger social graphs get more retweets, and thus more clicks, but when you factor that into the total number of pageviews, it usually ends up in the neighborhood of about 40 pageviews per tweet (ranging as low as about 16 and as high as 100).
There are exceptions to these findings, undoubtedly, but for the majority of users these are pretty good rules of thumb.
For those of you who have been monitoring my URL shortening project, you know that I monetize the traffic with display ads. I’ve put together a special concoction of ad types, with two thirds being flat rate CPM, and one third being engagement based advertising. Generally, my ads on the URL shortener perform in the range of $3-5 eCPM, with dips as low as $1.50 eCPM and spikes as high as $15 eCPM (dependent mostly on what campaign types are running on the engagement based ads).
When you take all these numbers and throw them into the right math equation, you quickly get into what are some real dollars and sense.
I’ll run the numbers on worst case scenario data, best case scenario data, and average data for your benefit here, all assuming no growth whatsoever in Twitter’s overall volume:
Worst Case Scenario
20,000 links per hour * 16 pageviews per link / 1000 * $1.50 eCPM = $450 / hr.
$450 /hr * 24 hrs * 7 days * 52 wks = $3,931,200 yearly revenue.
Best Case Scenario
324,000 links per hour * 100 pageviews per link / 1000 * $5 eCPM = $162k / hr.
$162k / hr * 24 hrs * 7 days * 52 wks = $1,415,232,000 yearly revenue.
Average and More Likely Scenario
170,000 links per hour * 40 pageviews per link / 1000 * $4 eCPM = $27.2k / hr.
$27.2k / hr * 24 hrs * 7 days * 52 wks = $415,833,600 yearly revenue
The numbers more than speak for themselves.
Obviously the ad networks I have been utilizing for my piddly little URL shortener wouldn’t be able to fill the inventory of something of this magnitude – so let’s dial back these eCPM numbers a little bit to something more along the lines of Google AdSense.
Most folks I talk to who monetize through Google AdSense typically see eCPMs ranging from $.50 eCPM up through $2.00 eCPM, and Google certainly has the ad inventory to fill those orders. Let’s run the numbers on something like that with our “More Likely Scenario” figures.
Low End AdSense Estimates
170,000 links per hour * 40 pageviews per link / 1000 * $.50 eCPM = $3.4k / hr.
$3.4k / hr * 24 hrs * 7 days * 52 wks = $29,702,400 yearly revenue
High End AdSense Estimates
170,000 links per hour * 40 pageviews per link / 1000 * $2.00 eCPM = $13.6k /hr.
$13.6k / hr * 24 hrs * 7 days * 52 wks = $118,809,600 yearly revenue
The costs to operate a URL shortener are marginal – TinyURL is a one man shop. To have a URL shortener and monetize it in this manner doesn’t even require the advanced stat tracking Bit.ly provides.
Throwing ads up on everyone’s links, though, may run afoul of some people’s sensibilities, but with a “Step 1, 2, 3” profit model like this within one day’s worth of development effort’s reach, can Twitter afford not to cash in now?