We had one year to complete the project under the Knight Foundation's guidelines. We spent the majority of that deploying and testing, deploying and testing, adjusting, then deploying again to see how the audience responded. We did this on a web-based version of our platform, which is a content-management system for hyperlocal communities to organize information around interest points and then geocode it. Our first implementation was for the burgeoning light-rail transit community in Phoenix.

During that time, we found that our users really responded to events near the train. Like other publications in the market, we had a hard time gaining traction from commuters using the rail, but we were more successful with college students and suburbanites using the train to seek entertainment in an urban environment (downtown Phoenix and Tempe, Ariz.).

From this, we organized "crawls" along our route with itineraries and exclusive promotions at participating bars, restaurants and retail stores near the train. This helped us improve our visibility with both of our user bases: light rail passengers and small businesses around the stations.

Interestingly, few users actually found value in the other types of content on the site. This included micro-level classifieds (sort of a Craigslist for the neighborhood around each station); networking (a resume-posting tool for each station to help passengers network with other riders they may see on the train every day); and fix-it projects (posts where people could garner support for neighborhood-improvement projects around each station). News posts had some traction, but the events were by far the most popular form of content.

Find out what's popular

So that's lesson one: Take the most popular part of your product or content and make it the "big carrot" in a mobile app.

We noodled this idea around and came up with a pretty solid version 1. First, we cajoled the local transit authority into giving us the complete schedule for train arrival times and departures, making us even more accurate and effective than the agency's own website. We posted these in the app so that they are searchable by station and day of the week. Second, we plugged in events, and then ran with a few ideas that will carry us into version two (and probably beyond).

Adding Features and functions

Which brings us to lesson two: See how far you can take the content from lesson one before you add new features/functions to the app.

Since several of the businesses near the train are involved in our events, we added a searchable business directory. Many of the events are also tied to public places (i.e., parks, libraries, common areas), so we included itineraries for each station that described these places, their location, and their significance to the community. Finally, users of the app can mark "favorite" places, businesses or events for future reference. Our next implementation will include a bridge that allows users to save events on their digital calendars (i.e., iCal, Google Calendar, etc.). See how far we took our events feature? You can do the same thing.

Which brings us to our final lesson: Don't try to be everything to every user in version one of your mobile app. Let the popular features/functions/content make the decisions for you. The audience is giving you the answer, so run with that to get something out the door.

What a difference a year makes. Last year, the CityCircles team was solely focused on building an audience for our unique, web-based publishing platform for the Phoenix light rail community. CityCircles has one of the slickest, easiest experiences for posting geo-tagged content (at least we like to think so).

That said, think about everything that has happened in the past year from a technology and user experience standpoint. Safe to say that if we had a crystal ball when we filed our Knight Foundation News Challenge application in 2008, we'd have put our focus on the mobile space instead of on the web. We figured that if the audience loved the web version of CityCircles, it would translate well to the mobile version.

We were wrong. We went left, and the trends went right. But now we're on the right track with our first mobile application for the iPhone (awaiting review -- torture!). We've been able to learn from our mistake and play with some new toys that may grow in popularity as they become more sophisticated.

Adding A 'Cool' Factor

I'm talking about animation. For version 1, we wanted to display the train stations on a map and list businesses, events, itineraries and arrival times for each of the light rail's 31 stations. It would have been easy to present this on an established mapping platform such as Google Maps, but we wanted to do something unique. What if you could open the app, and watch an animated train crawl down the tracks to see if it's getting close to your location? This would be different than many other transit-related apps out there, and it would add a "cool" factor to it.

Here's a scenario we used: You're at a restaurant or bar near light rail. You open the app; and you leave it open so every now and then, you can glance at it and see if the train is getting close. You pay, exit the business, and check the arrival times on your way to the platform using the app's train schedules.

Our developer, Micah Jameson, wanted to play with a new tool called Raphael to see if the animated train effect could work. In short, it worked very well. Here's a screencast with an explanation from Micah:

http://www.youtube.com/watch?v=Kw9hUNBQLCI&feature=player_embedded

CityCircles iPhone app with Raphael

Looks cool, right? There's just one problem: The animation really slowed down the app. And we know what happens to slow apps, don't we?

Going with Google Maps

So we went back to drawing board. We ripped out the Raphael coding and replaced it with Google Maps. The experience is more static, but much, much faster. And since our audience may have a minute or two to make a decision to board the train, this functionality was superior to the Raphael implementation above. Here's a screencast with an explanation from Micah:

http://www.youtube.com/watch?v=KcJdw1BSKwQ&feature=player_embedded

CityCircles iPhone app with Google Maps

I asked Micah for his opinion overall on the Raphael experiment. This is what he had to say in an email:

Although it's fairly common practice to display content using a UIWebview in an iPhone app, JavaScript performance still needs to gain some traction before it can be used as an animation platform. I've experienced a performance difference between the iPhone simulator versus actual hardware in the past, but on the scale of 1.5X-2X slower. Raphael inside a UIWebView with any substantial amount of data to display was more like 10-20X slower. I did run tests with Raphael on an actual iPhone before I started programming, and it ran fine. It wasn't until I started trying to display a lot of data that it hit a performance wall. This leads me to believe that Raphael uses a lot of memory to run effectively, so a memory-constrained device like an iPhone won't run it well.

For more updates on our progress, you can follow us on Twitter (@citycirc) or read our blog.

Adam Klawonn, a native of Phoenix, has spent the first part of his career working for metro newspapers and a city magazine. He started ZonieReport.com in 2006 to cover statewide issues in Arizona and previously taught online journalism classes at the Walter Cronkite School of Journalism at Arizona State University.

As we discussed CityCircles during the meeting, the inevitable question arose: How much traffic are you getting?

The answer, in all honesty, is not much at the moment.

But "hits" -- or page views, or unique visits, or whatever traditional web metric you choose to use here -- is not what we're looking for at CityCircles.

Our project is less about "how many" people are using the service and more about "how" people are using it: How they are interacting with it, with each other, and with the light rail community at-large as a result of our existence. I bring this up because it will inevitably be part of any early discussion you may have about your own startup.

The Battle For the Top of Search Results

Your answer will obviously be critical to how the project is perceived. For us, we do our best to follow how web usage is developing as new startups go live. One particularly interesting development is the mile-wide content creators like Demand Media, Associated Content and other related sites. (See the ongoing feature on these kind of content farms being published this week at MediaShift.)

In general, these companies pay writers of a general skill level to write about almost every topic under the sun for an extremely, ahem, modest fee. They are essentially choosing quantity over quality as their business model. (However, that is in the eye of the beholder, as any piece of content is capable of being high-quality to a particular user if it's exactly what they're looking for at exactly the right time. It just tends to be something that won't win any major journalism accolades.)

A really great story on this topic -- with a really great volley of thoughtful comments -- came out earlier this month on The Wrap.

There's a lot there to contemplate, but what I prefer to ponder is a post written by FoundingDulcinea's Mark Moran in December 2009.

He argues -- successfully, in my mind -- that sites like Associated Content and others will, over time, kill search engines' usefulness (if the search companies don't address this issue). The deluge of content from thousands of writers on multiple topics will come to dominate the top search rankings, thus diminishing the utility each user gets from that search. As some have noted, certain searches require you to wade through posts to get to the deeper results on a topic you are interested in, and this equates to being invisible in search because few users click past the first page or two of search results.

Metrics to Consider

Why do I bring this up for potential startups?

The impact these sort of sites can have may force you to re-think your own metrics. If page views work for you (and you should think beyond that), then that's great. Just remember to follow developments that impact search engines, because that is where validation for your project will come from as you talk to potential stakeholders.

If you'd like to consider other options, here are a few metrics we are tracking under our grant:

• Number of registered users
• Number of posts
• Number of comments on those posts
• Number of community improvement projects completed
• Most frequently visited landing pages for light rail stops
• Length of time spent on our mobile website (train schedules)
• Interviews with users at our light rail events (anecdotal stories about light rail use)
• Number and type of merchants participating in our light rail events

Food for thought -- especially at a time when startups have to score millions of page views to attract a whiff of advertising money, if that's your business model. Our model is based on a deeper experience of use, not just information consumption.

Start thinking ahead to answer that inevitable question.

By "pitch" I mean: How do you explain this to your audience in 10 seconds or less (i.e., an elevator pitch)? How do you explain this to people who may want to work with you after the grant ends?

We've finally found an approach that seems to be working for CityCircles, so let me save you some time by presenting two options to consider.

Pitching the Audience

First, let's start with our audience -- the people we want interacting with the site via their phones. These are folks who either ride light rail, live by it, or have a business near the tracks (within five blocks, to be exact). We've built a website and mobile site that lets them post news, events, classifieds, promotions, resumes, and community projects. Journalists can also contribute stories to the community.

Earlier in the grant period, I would give people what I thought was a clean, concise-yet-explanatory pitch about what we are and how we work. As time went by, it became shorter and shorter, and the look in their eyes seemed more attentive. I was on the right track. At some point, I tried this pitch:

It's called CityCircles. We're like Craigslist for the light rail community -- with a splash of journalism.

That's it. Done. Pause two seconds to see it sink in, then add a few details to fill in the blanks.

Now, I'm not advocating for making your project sound like a cocktail with a weird twist. But I am advocating for using a comparison to something that already exists, closely relates to your project and is widely known. It helps get the point across quickly. Choose wisely.

Detractors might ask, "Why would you want to promote someone else in your pitch?"

To which I respond: If the other project already has an established community "feel" to it, then it may help your chances. In our case, Craigslist is a universally known example, and it resonates immediately. The "splash of journalism" adds context and meaning.

Investor Pitch

Now let's look at the investor pitch. These are the folks who will keep your project going after the grant expires. We're solely focused on making our project a success in Phoenix, but reality dictates that we plan ahead. To do this, we're considering "sponsors" in other markets.

Instead of trying to find one large-scale investor, you might consider a "sponsorship" from someone who needs/wants some additional street cred.

One thing I am looking at is last year's green company rankings from Newsweek. Since we're a platform that encourages civic involvement and mass transit in the inner city, someone toward the top of the list could add us to their "green portfolio" if they support our expansion (in exchange for a promotional advertising presence on the site, like NPR's "brought to you by..."). Conversely, those companies toward the end of the list may want to use us as a means to advance up the rankings.

And like the audience pitch, choose wisely.

That said, one person can't carry the entire load, especially as the project inevitably evolves from its humble beginnings and wire frames.

One solution that's worth considering is sinking some funds into a test suite -- a closed environment where other developers who share a vision for the project can develop new features with the approval of the "master" developer. This is the approach we recently took with CityCircles.

Test Suite

In March, we contracted with a local development shop called Integrum Technologies to build a test suite. The project is connected to our code base and includes simulated tasks that other developers can build toward and "test." If these features pass muster in the test suite, then we can push those changes to our code base permanently. If they do not, then the developer can tweak them until they do without ruining the live site.

The test suite took almost three weeks to build and cost us roughly $9,500. (That may seem pricey to some, but good Ruby on Rails developers are not cheap. In our case, Integrum specializes in test suites.) However, for startups, this is a very helpful option for reaching goals of new features and functions on a budget. Open-source software developers that are looking for a "portfolio" piece and are attracted to the project's mission can participate at a fraction of the cost to the project. In return, they receive publicity and, in some cases, a promise of future paid work. The idea is that everyone wins.

Once your test suite is completed, start poking around your local area for developer meetups. Go online and subscribe to developer forums and Google groups. In our case, the project is built in Ruby on Rails. I have joined the Rails community's leading Google Group with the intent of marketing this test suite to developers.

I've also been invited to attend Integrum's weekly "hacknight" meetup in Chandler, a Phoenix suburb. Tomorrow night, I'll be there to spread the gospel of the project and hope that our handy test suite attracts the right crew.

Use these test suites to your advantage, as simulators like them can also help create an organic "buzz" around the project as well. Include the developers' names on the open-source software license, too. That will also help.

But be mindful of the pitfalls. Just as there are several developers that may want to participate, they may not have the chops to complete the work in a timely or accurate manner. It helps to have a strong master developer to sign off on their work.

Since most of you probably haven't heard of TZR, here's a quick recap of my post: In my digital farewell, I talked about why I did what I did, outlined a few things I learned, and shared what I planned to do next. Since then, I realized I should have elaborated more on my lessons learned because I feel they have been misinterpreted. I could do that on The Zonie Report, or I could do that here for the greater good of online journalists. I chose the latter.

What follows is a top five list of things I learned along the way of running a critically acclaimed but commercially unsuccessful news website in the great state of Arizona. I am applying what I learned to CityCircles, a hyper-local information tool for light rail riders and nearby residents in Phoenix, and I hope this comes in handy for your respective online journalism projects as well.

Lessons Learned

1. Identify the community you are serving and get really, really close to it. On The Zonie Report, I tried to write about "the rest of the Arizona story" (thank you, Paul Harvey) by focusing on regional and statewide topics that I felt mainstream media outlets missed. Great idea, except I was entering the "being all things to all people" territory that challenges metro newspaper coverage today. With CityCircles, we are drilling down to everything within a five-block radius of Phoenix light rail stops, which run for 20 miles. This is much more efficient and doable -- and more relevant for a higher concentration of folks.
2. Differentiate your presentation, voice and method as much as possible. This one may seem obvious, but shouldn't be overlooked. At TZR, I tried to write with more of a folksy style, and I think people enjoyed it. It engenders conversation, comments and trackbacks (which help site visibility) while appealing to readers. If they want drier or more official writing, there's a lot to choose from. Experiment with a weekly editor's video/podcast. Make yourself "human" - and not just an unseen editor - as much as possible.
3. Identify your competition, then link to them whenever possible. First, use of the word "competition" in online media can be debated, but any journalists reading this will know it's a commonly used term in the industry. This may seem like heresy, but don't be afraid to link to other sites with overlapping work. It's good practice and a good service to readers, and it can help with site visibility in the long run. One of the most successful things I did at TZR was find the most interesting statewide news stories of the day from rural news outlets and point to it. I plan to do the same kind of linking with CityCircles.
4. Use social media for more than just marketing. On New Year's Day, David Carr of The New York Times wrote a great post about Twitter. He quoted Steven Johnson, a tech observer who wrote about the 140-character phenom for Time, as saying that Twitter "is looking more and more like plumbing, and plumbing is eternal." I agree. My point is that a tool where millions of people can write about anything at any given moment on almost any device is incredibly more powerful than something that you'd just use to market your stories on your website. That's mostly how I used it at TZR; at CityCircles, we've bent Twitter so that train passengers can "tweet" all types of content - from news to events to classifieds and more - across any of the train stops. Be creative with these tools.
5. Put together a business plan that acts like online ad revenues don't exist. Banner ad networks continue to drive online advertising prices into the floor. Unless your startup is the next Facebook and can produce volume and deep targeting, you will need a more diverse plan for sustainability. I tried this with TZR and it went almost nowhere. At CityCircles, our initial plan is to bundle online advertising with other marketing services for light rail merchants, and to make it very affordable. This may change, but, for now, we're rolling with it.

And on that note, I'd like to add more thing: Never be afraid to change your strategy on anything. The biggest mistake I made with The Zonie Report was taking too much time to let go of ideas/concepts that weren't working. Don't do this. It will bog you down - and keep you from the ideas/concepts that really do work.

This was a crucial event. ASU has a huge footprint in the Phoenix area because it has 69,000 students. They buzz around the Valley in cars, on bikes, on foot and yes, on light rail. This makes them a huge group for us as potential users and collaborators.

As we talked to them, we realized that an assumption we made early on -- one that many other projects probably make as well -- was not accurate. In the hype surrounding the race to 3G and 4G and touch-screen phones and do-it-all applications, only a fraction of the market uses these smartphones. The rest of us have, for lack of a better term, "dumphones."

This means that designing a platform for smartphone users could lead one astray. Every student we spoke with had different versions of phones that used a native web browser. We initially considered this reality and were encouraged by our advisors to do so, but we didn't realize how dead-on the advice was until we interacted with some of our users.

So there are two takeaways here: organize events that allow you to interact one-on-one with different groups of your core users; and gear your mobile development towards dumphones. You can always build that slick mobile app later.

For more on our ASU experience, check out our video:

CityCircles at ASU from Adam Klawonn on Vimeo.

It was an incredible opportunity and an enlightening experience that really helped us out. For those of you who are thinking about doing a new media project or are already working on one, the TV interview format can be a boon as it forces you to sum up your project in small sound bites (while under the extreme pressure of being on live television).

Apparently, we did a great job. How do I know this? Well, when we laid out the basics of the project (our pitch), the anchor, Tess Rafols, thought it made perfect sense and, in fact, was surprised no one had thought of it yet. (The Internet is a pretty big place, so trust us, Tess -- this idea is floating around elsewhere.) Still, we were very encouraged, and we think we are bringing a great concept to the table.

Now for the fun part: The interview also made us re-think our name, Daily Phoenix.

This has been a source of angst for us since the interview. Why? Well, the word "Daily" doesn't really add much utility to the name and isn't very descriptive. It's on the Web, so, of course, anything can be "daily." Originally, we thought it made sense because it gave people the sense that this is a publication -- which, indeed, is the exact reaction we have received. But others have said that it sounds like "old media."

The word "Phoenix" is also troubling. The bulk of our light-rail system is in Phoenix, but it also runs out to the suburbs of Tempe and Mesa. Although newbies use the word "Phoenix" to refer to the region, using "Phoenix" in our name could alienate local folks (our primary users).

As a result, we've decided to brainstorm a new list of names over the next two weeks. This will be a collaborative tool for the community, so, in that spirit, I'd like to toss out this opportunity to the Idea Lab audience for help. We can't offer fame or fortune, but we can offer that cocktail party "coolness" factor if your name -- or some mashup of names submitted below -- actually sticks.

Here are our basic guidelines:

• Our project uses web, print and mobile technologies to deliver information by light-rail stop to urban audiences. They can collaborate or use it to guide themselves around the city's news, events and services.

• Because of this, they should feel like they have control over their environment. This is an empowering tool for them.

• It's location-based knowledge -- meaning it has an awareness at a certain point and time.

• Whatever the name is, it would be nice if it could become a verb among our users (i.e., "googled").

That's it! Although we're coming up with our own list, any suggestions below in the comments are welcome. Thanks!