Twitter has a problem, they fail at their own use case.

Twitter has always struggled to communicate what its use case was. Not being an IM application, or a social network in the ilk of MySpace of Facebook. It has flirted with a few concepts as to what Twitter is and what benefits a user could expect to get when using the service.

At first, it was a tool to notify your friends via SMS and the web as to what you were doing.

Twitter’s second use case was pitched to us as a service that provided access to a direct line to allow everyday people to communicate to the celebrities they admired. The publicity around Oprah Winfrey joining was a significant draw card that brought massive growth to the Twitter platform. People though were disillusioned after signing up when the celebrities they signed up to communicate with either treated the service as a one way press release syndication tool, or (simply due to the number of people tweeting them) were not able to meet the communications demands their followers were asking of them.

This lead to a re-branding of the Twitter site, to focus on the sites potential to tap into the pulse of the globe at this very moment. Changing the question posed by Twitter to users from “What are you doing” to “What’s happening” is a very important evolution in the psyche that powers the Twitter universe. Additional to this has been location based twitter trends so users can tap into the up to the minute news and hot topics from locations of interest.

This is a great step to realising the true potential of Twitter as a service. There is a problem however, which is that if Twitter is positioning itself to be the pulse of the world. That when significant events occur, the service needs to be able to handle the additional load on it’s systems that major world events can put on it.

On Wednesday 20th of Jan, an aftershock measuring 6.1 hit Haiti, almost immediately after. Twitter was unresponsive for 90 minutes, suddenly the use case of Twitter, that of connecting us to what is happening right now fell to pieces, and showed us that depending on Twitter to deliver on the use case pitched at us is not wise, and raises questions over the services long term survival if hey cannot deliver on the core of their business.

The official word is that there was a “failure” somewhere (possibly hardware) behind the outage, however it is entirely possible that the outage was triggered by the surge in tweet volume after the 6.1 aftershock. Twitter already has trimmed the number of tweets it will cache significantly for search.twitter.com results citing performance issues, so it is reasonable to assume that a considerable spike in traffic might be enough to tip Twitter over the edge.

Regardless of whether this outage was caused by a hardware fault or capacity limitations, if a service pitches itself as being the pulse of the globe as Twitter does, it simply cannot be unavailable for a significant period of time during a crisis like this.

This tells us,  that in its current form, Twitter can not deliver the service required to meet their own use case to us. This undermines the brand, and until Twitter can iron out the bugs and capacity issues they face, they need to re-examine their use-case message and branding to users.

12 thoughts on “Twitter has a problem, they fail at their own use case.”

  1. Twitter does a remarkable job given that the service is provided for free.

    Any shortcomings are typically solvable by cash but I cannot see where Twitter currently makes money, I might have misssed Twitter’s source of funding but it is my contention that the service is reasonable.

    In the meantime, via Twitter, much money is being raised for the Haiti.

    Further, mainstream information sources are also badly flawed , I give you as exhibit one – the Herald Sun.

  2. Indeed, owners of social network / social media / messaging sites are generally quite clear about this being the transaction. They own your data now, and now they can do what they like with it.

    An excellent example of this came from MySpace CEO Owen van Natta yesterday, as cited on Twitter by Robert Andrews:

    One of the things we’ve never disclosed is, we’ve got over 180m playlists created by users.”

    As he says, “We’ve got.”

    I’ve often thought that even if Rupert Murdoch makes a massive loss on the buying and selling of MySpace itself, while his organisations have access they have access to this treasure trove of personal data. data mining firms can do all sorts of analysis on this.

  3. I actually think that the lack of a specific use case, and the simplicity of the service offering, is the BEST thing about Twitter, not its big problem. I see Twitter’s big problem being the lack of a solid business model from which they can derive significant, sustained revenue. The search engine deals are a good start, but I really don’t see them being enough to keep the company going indefinitely.

    Currently, Twitter’s biggest issue from the point-of-view of its users is availability. I (perhaps incorrectly, but bear with me) put this down to scalability problems. The only way you solve scalability problems is to throw money at them: more hardware, more datacenter locations, refactoring your code, etc., etc.

    For me, the more features Twitter has added, the less I’ve liked it. If I follow someone, I _want_ to (at least be able to) see their mentions to everyone. And I’d much rather metadata like mentions, hashtags, and retweet syntax be taken out of the 140-char (or whatever it is this week) limit than have fancy-yet-broken features.

  4. Even without the Haiti earthquake it is easy for a new user to find out that they can’t handle the “pulse of the world” use case. It is extremely frustrating to tweet a few things (esp. for work purposes) and then find out that some (if not all) do not appear in search for no obvious reason, and there is no channel to feed our frustrations back to Twitter’s developers.

  5. I think the problem is we are expecting too much of services like Twitter. These tools were never designed to be used for disasters and relying on them in an emergency is risky at best.

    The situation in Haiti is a good example of this, Web2.0 and cellular phone services are swamped in times of great demand and should there be damage to infrastructure, such as in Haiti, the Victorian bushfires or the September 11 terrorist attacks, the systems become even less effective.

    We need to keep the limitations of these tools in mind. This is why we need to be very careful on how we use them for disaster management.

  6. Yes, twitter has been crashing a lot lately. It’s a curious concept, and I don’t see how it can keep going without some form of advertising. This may not appeal to many twitterers, but how can it succeed without revenue? It’s often over-loaded even now. It needs more servers or support , and this means money.

  7. Great post David, and good to see informed commentary above as well. One of the most powerful features of Twitter is it’s Search, which most advanced users would agree is “broken” because at best you get about a week of history. Eventually that tweet you were looking for just isn’t appearing any more.

    The good news is that the tweet is not lost – have a read of Chris Brogan’s post “How to Listen for Opportunities on Twitter”

    http://www.chrisbrogan.com/how-to-listen-for-opportunities-on-twitter/

    The ability to use RSS to track a search result can be very useful. I also use FriendFeed – it has powerful search and once a Twitter account is added, it will remember all the tweets. Interestingly FriendFeed supports “imaginary friends” so you can add people not even on the FriendFeed network to track their tweets and keep them searchable.

    We are fortunate in that Twitter’s open API allows other networks to provide functionality that is lacking in Twitter’s own service – search and the use of FriendFeed to overcome the limitations is just one example.

    Cheers,
    Tony Hollingsworth

  8. Hi David,

    Nice post. Hit on 2 oft-asked issues:
    Why tweet? Where’s the money in it?
    That post fetched some well-informed comments.
    Good read. Thanks for that.

Comments are closed.