I’m a huge Twitter proponent, and it’s frustrating to me when I see people under utilizing it because they don’t see the potential that a 140 character maximum size message can yield, or are just flat out using it incorrectly. To start, let’s analyze a very simple, seemingly innocuous tweet. The kind that people point to when they say Twitter is stupid:

“I’m at the Kona Bistro having a cilantro pesto chicken sandwich.”

This is a tweet. The basic unit of measure in the Twitter ecosystem. From that simple 65 character message, some of the following reactions can be elicited:

1. Kona Bistro? Where is that? Is it good?
2. Kona Bistro. I like/don’t like that place.
3. Kona Bistro. It’s close by, I’ll join him./It’s far away, lucky bastard.
4. Cilantro pesto: either that sounds good or it doesn’t. Regardless, I’m hungry.
5. Chicken sandwich. Bill is not a vegetarian.
6. Bill’s been out of town but is now back in Cincinnati.
7. Bill’s eating a sandwich.

To people that don’t get Twitter, they see “Bill’s eating a sandwich”. The reader can choose to ignore the statement – going on about their life, using Twitter, but never really understanding what it is, what it can do, simply tweeting about how insipid Twitter is because they don’t care about what someone is having for lunch. To people that get Twitter, they see everything outside of the obvious statement. This reader, who has some semblance of intellectual curiosity, can explore the myriad of other options presented and learn a little bit more about Bill, find a new restaurant to try, or meet a friend for lunch instead of eating in their cube that day.

Twitter is one of the most popular, yet most misunderstood of all the social media platforms out today. While on the surface, Twitter is indeed about communicating with your followers (who may or may not be your friends), even if you’re on board with the above example, that’s still only scratching the surface of the functional value of Twitter. What about:

Data Entry (Command Line Interface)
For an individual, Fuel Frog is a cool little web app that performs one primary function: tracking your fuel economy. For the aggregate, the data it collects can be used to track the fuel economy of everyone with the same type of car, determine fuel pricing by region of the country and determine when people take driving vacations (MPG goes way up on long trips). The only data entry looks like this: @fuelfrog 342.0 2.479 15.33. That’s the number of miles traveled on the tank of gas, the cost per gallon, and the number of gallons filled. That’s it. The FuelFrog processor then picks up the transaction and adds it to your history. From the thousands, if not millions of data points it’s received over the last 24 months or so, all sorts of trending data is available about fuel consumption that can be extrapolated into other areas. Is your car getting 5 mpg less than everyone else with the same model? “Hello Mr. Klos, this is Toyota. We’ve noticed that you may be having a problem with your Camry as you’re not getting near the gas mileage you should be. If you want to bring it in, we’ll take a look at it for you.”

Twitter lends itself to quick data entry. Your follower is simply an observant program designed to handle whatever simple task it was designed to do. Adding a trip to your itinerary or billing your client or can all be performed from whereever you happen to be. Don’t want to use the command line/text interface? Putting a more attractive graphical cover on it is just a smart phone away.

Application Communications
Use Twitter to provide up-to-date news about features, new versions, tips-of the day, or a quick-and-dirty means for in-app communication. In the next version of Stonage we’re implementing the ability to fight using Twitter as the means for sending both the fight request and the fight response to handle the need to be able to fight asynchronously. As an added bonus, Twitter is the means to send the user a “tip-of-the-day” every time they start the game. Yes we could build that statically into the the game, but this gives us a way to actually provide fresh content without having to build the infrastructure to support unneeded robustness.

Messaging Processor
For app developers (especially mobile apps), Twitter provides a very cheap, relatively stable way to communicate between applications if you don’t want to go through the hassle of standing up a full blown message queue. While maybe you wouldn’t use it for mission critical applications, and definitely not for high-volume purposes, using it for proof-of-concept work and other lightweight needs would be quick and simple to do.

Transaction Monitor
Waiting for a job to finish? How about you want to know when your credit card is charged? Did you flight gate just change? These are all examples of how Twitter can be used to notify you of something current and relevant. Sure, you can use SMS, but the ubiquity of Twitter means it can be woven into whatever applications you want.

Corporate ‘Bot
An automated rules engine makes a much better mouthpiece for a corporation than picking some random marketing person to spew out mission statements periodically.  This bot can be used to answer queries about your company, pull contact information from the corporate directory, be used to update project status, broadcast the cafeteria menu, and whole list of assorted functions.  Yes you can give it a corporate voice too and actually use it to have it speak for the entire company, rather than just the marketing department.  This one of the things I’ve done for my company and it’s actually quite cool.

Online, Realtime (almost) Help
Need to ask the developer a question? A custom Twitter interface links the application developer and the user together. The user’s Twitter settings are part of the preferences page and the application knows the Twitter account of the developer. All communications can be handled in the app so the experience can be seamless.

Marketing Mouthpiece
Not something I recommend normally, but if it’s done right, I’ll follow it. By being done right, I don’t mean broadcasting advertisements to me every hour. That will get you blocked. Rather, information that’s important to me. If you’re a nationwide pizza company, hit me with information that’s relevant to my location. I’ll follow @papajohns_cincy vs. @papajohns if you give me different offerings locally than you do nationally. Mmmmm, Goetta pizza. Or if you’re one of my vendors announcing system outages, software provider: new features to your software, a game store: “We have 3 Wiis in stock.”, local sanitation department: “Christmas trees picked up tomorrow” or anything that is hyperfocused to my context. You will be very useful, and you’ll get followed.

Backchannel Feedback Gatherer
One of the most powerful, but under-used features of Internet infrastructure and the capabilities we have today is backchannel discussion. These are the murmuring conversations that go on in big meetings, conference calls, and other group events. On podcasts or webinars where there is a live audience, these usually take the form of a chat room where the moderator can see questions being asked and various relevant comments and can choose to include those points-of-view or questions into the main presentation or conversation. This allows the overall experience to include a more evolving context than what the general presentation may have provided.

What Twitter can do, is to form impromptu backchannel discussions anywhere and at anytime. Through the use of hashtags (a “#” then whatever keyword is agreed upon), anyone with a Twitter client can monitor and contribute to the alternate conversations going on surrounding a common occurrence. This method avoids the need to build the infrastructure required for a realtime audio/video feed with integrated IM, and can actually occur later in date/time if the event were something like a podcast being listened to weeks after it was originally broadcast. The hashtag would allow the producers to gather comments and feedback as far after as made sense, and still incorporate whatever ideas or suggestions into future endeavors.

Additionally, this method could be used for a quick “show of hands” for large audience. For example: “Tweet #retreatrio if you agree the company retreat should be in Rio de Janeiro next year, or tweet #retreatscranton if you want it to be in Pennsylvania.” The votes are tallied in pseudo-realtime (give or take Twitter performance) and you’ve got measured feedback. Why American Idol or any other voting based variety show hasn’t picked up this yet I don’t know, but it’s perfect for large corporate meetings, sporting events, and church.

Building It
The architecture to put all of these types of applications together is not a monstrously difficult proposition and they actually work relatively well in a relatively short amount of time. Getting them to be rock-solid takes a little more time, but in the end, you have an automated solution that simply responds to the rules you tell it, and the requests it receives from the Twitter timeline. There’s no need for a fancy interface (Twitter provides the ins and outs), you just have to parse and build tweets. C’mon, they’re only 140 characters. Let’s build something cool!

You’re Doing it Wrong
OK. So you’re happy with just using Twitter as a communication means for keeping up with people with shared interests. Fine. You get all the talk described above, but you just want to tweet, retweet, and read tweets. That’s perfectly valid, but if any of these apply to you, you’re doing it wrong:

1. You follow more people than you could possibly keep up with unless tweeting is a full time job for you (like a marketer). There’s no real number that acts as a threshold because it depends on the type of people you follow. Do they tweet on average 3x a day, or 20-30? A couple of hundred I can see. If you’re following over 1000, sorry, you’re looking for quantity, not quality.
2. You’re a marketer.
3. You tweet 30+ times a day. This is called spamming. Stop it. Unless you’re providing original content for value then it better be interesting because you’re monopolizing your followers’ time. Unless your followers fall into the first category in which case they have no idea you’re bombarding them because there’s so much noise on their screen. If you are simply RT other people’s comments and they’re doing the same for you, then you’re part of a twisted Mutual Admiration Society where as long as you people stay self contained, have at it. I’ve seen it. It’s creepy. I’ve stopped following people who are in such a group simply because the spillage of retweet traffic got to be annoying and I felt like I was being brainwashed.
4. You auto-follow. Do you follow someone just because they follow you? Needy much? Following someone should be a conscious decision. How often do they tweet? What do they tweet about? Do you get value from their tweets? If they’re geographically close, work for the same company, have similar interests, or are flat out just interesting, then by all means, follow them. But, if you wouldn’t follow them around town, why follow them online? Personally, I try to reduce the amount of connections in my life that don’t share my context.
5. You have an abnormally high celebrity percentage in your list of people you follow (>50%). Here’s my first rule of thumb for following celebrities: they have to do they’re own tweeting. My other one is the same as described above: there has to be context that they provide to me. I follow a bunch of cyclists: Lance (@lancearmstrong), David Zabriskie (@dzabriskie), Levi (@levileipheimer) and I really enjoy seeing their perspectives on races after they happen. I think Nathan Fillion (@nathanfillion) is cool and hilarious to boot and I’ve always been a huge Kevin Smith (@thatkevinsmith) fan. Round that out with some authors, scientists, and online personalities and my overall celebrity percentage is about 15-20%. Of course I define a celebrity as anyone more famous than me – so people I count as famous, other people may have never heard of thus my percentage may be construed as high.

So try looking at Twitter through the lens of these suggestions and comments.  I’m hoping to see a lot more of the less obvious uses start to come to the surface.