The Downside Of URL Shorteners And One Answer

The recent announcement by Pierre Far at Cli.gs that the company would be shutting down its URL shortening service raises questions about the viability of this sort of company.

As Far said,

• Short URLs are a feature, and are definitely not a business on their own. This assertion is fact in my mind given what I’m seeing in the market and what my customers and users are saying. There comes a point when you need to actually hear the message the market is telling you, and not just listen and ignore it. This is the core reason.
• Cligs is a big service (for some definition of big). It has tens of thousands of registered user accounts, and does tens of millions of forwards a month. Not bad for a one man show doing this on the side. However, a service this size requires customer support, which is time. It’s not much time in absolute terms (an hour a day usually), but again, for me, it’s a context switch I can do without.
• I have a nice plan for how to add premium features on top of offering short URLs. When I look at that plan as a business, it’s quite separate from a short URL play. Building on what I have now, and integrating the two might work but will require a lot of effort (i.e. time) that I cannot afford to give. This is the other major reason.
• Finally, it all costs some money. It’s a bit of money out of pocket every month. By necessity, the Cligs architecture is extremely efficient for what it does.

Note that Cli.gs will be supporting redirection of its URLs through the end of November, which may give the company time to find a buyer. But, as Far notes in an update to his first announcement, the company is a founding member in 301works, an organization that is still in the works, but which is intended to ensure against the possibility of individuals or organizations creating and using shortened URLs that might someday not be redirected because a URL shortener has shut down. He writes

What about 301Works? As a founding member of 301works, of course it’s something I know about and thinking how to best use it in this instance.

This comes only a few months after Nambu’s similar announcement about the tr.im URL shortening service, which is now headed toward an open source future, which is one possible approach to shutdowns like these.

But Tr.im’s management have decided that 301works is a ‘publicity stunt’ being orchestrated by Bit.ly, a leading competitor. I have spoken with John Borthwick about this episode, and I think that his actions were motivated by a concern for the impact that a tr.im shutdown would have on the perception of URL shorteners generally. Note that he is Bit.ly’s CEO, and it is the leading URL shortener. So his efforts weren’t altruistic: he stands to benefit if people trust URL shorteners.

But we, the community, stand to gain if we really can trust them. And so I am willing to work to make sure that we can.

Nambu’s plans for Tr.im may match the hopes of an active and willing community that will take over the direction and hosting for the Tr.im service. However, there are a growing number of companies that are entering this area, especially companies with a different business model, one in which URL shortening is just a feature. I had a demo from such a company last week, and when I asked if they were licensing their URL shortening from another company, they said no, they were going to do it themselves, because they wanted to capture the analytics, and in such a way as to directly match their use, which they saw as a competitive advantage.

For this reason, I expect we will continue to see new shorteners appearing every week, and I don’t expect that many of these new entrants will have an incentive to get involved with the Tr.im open source effort, or to buy up an existing company like Cli.gs. Why inherit the headaches and costs of supporting a batch of old URLs? Therefore, we are likely to see a continuing drop out of URL companies, as a part of the normal ebb and flow of business.

I have started to become involved in 301works, through my activities in Microsyntax.org, the non-profit I set up over the past few months. The 301works organization is still in the process of being formalized, and its relationship to Microsyntax.org and other organizations is still in the process of definition.

Its goal is to ensure that those using URL shortening services can do so without worrying that these URLs will someday stop working.

However, for 301works to stand as that guarantor, the participating members will have to agree to cede control of their redirection in the case of a shutdown. Otherwise, the organization and the guarantee to the community that 301works represents will be hollow. And if it is only that, just PR, I won’t be a part of it.

Certainly companies can opt not to be a part of 301works, and maybe the community won’t mind. It won’t stop companies from raising money, or selling their service off. But it will mean — once we hammer down the official details — that users of shortening services will know that any company bearing the 301works seal has contractually agreed to hand over control of any shortening domain to the organization in the case of a shutdown.

Obviously, 301works will have to structure the technical side of things so that the redirection for these shortened URLs will work. That is a technical challenge, but one that can be managed. And the organization will have to canoodle bandwidth and servers so that URL redirection on these retired domains will work, or raise money to do so.

And 301works must be structured so that it is not perceived as a publicity stunt of any single participant, or of the participants collectively. We are working to create a governing board where all participating companies have an equal voice and a seat at the table, but where the interests of no one company dominates.

Most importantly, 301works — in whatever form it takes, finally — has to run on our behalf, we, the users. We are the ones potentially harmed if a service shuts down. We are the ones harmed by link rot.

And it is our trust that these companies need. So in the near future, once 301works has a seal that participating companies can display, it will stand as a guarantee that your URLs will work. And we can get back to business.

[disclaimer: I consult with companies that create shortened URLs, including Bit.ly.]