Build it they will come AND replace existing process

[fgregg]

If an existing process/activity is well defined in users mind AND it's hard to do, then sometimes they will come.

http://cpstiers.opencityapps.org/ is the most trafficked app that I've helped make. It just came from feeling "This shouldn't be so hard" when I looked at the official process http://cpsmagnet.org/apps/news/show_news.jsp?REC_ID=184188&id=0

Most of our traffic comes from google with people searching for information about cps tiers.

[fgregg]

maybe's what's different here is that it works with google and people's pattern of problem solving

another example is http://wasmycartowed.com/

[bensheldon]

Yes! There is a lot to tease out of the CPS Tiers example. For example:

• Did you do a Keyword analysis beforehand about how many people were searching those particular terms/phrases and try to SEO towards that? I think you guys also did a ton of outreach to parents groups and social workers to let them know this thing existed too.
• Were you trying to do behavior change? (e.g. bring new people into the school selection process). You definitely made the process easier, but did you actually make it more accessible? (I know a lot of this was also hashed out on the OpenGovChicago mailing list too).

Definitely this: it works with google and people's pattern of problem solving. I think this is key, and is actually a good idea pattern: market or marketshare analysis: how big is the actual market of people who are actively searching for answers, how fragmented is it, how are people being served (in the CPS Tiers example, are for those who already are trying to figure out the Tiers system, how are they doing so? just rereading the CPS website till they grok it?).

This isn't to imply that problems with only a few potential users (or very difficult to acquire users) shouldn't be tackled, but that that analysis should be done upfront so you can weigh potential breadth and impact.

Thanks for opening this issue; I'll work on getting it integrated.

[fgregg]

We didn't do the keyword analysis (although that's a great idea). However, I did see that, on the parent blogs, the tier system was something that parents were really interested in and trying to understand.

However, you can use the site to get the piece of info you need for your school application (your tier), without having to learn about the policy. 3/4 start with that, although about 20% come directly to the policy explanation page. about 1/7 go to policy page after hitting the top page.

So, to zoom out a little. We are first trying to serve parents and kids going through the application process, and who are just trying to figure out this one piece of info, and then a smaller but still sizeable number of folks who want to understand what this thing is.

Behaviour change: We were not thinking about increasing the number of people who tried to look up their tier, but about increasing the number of people who succeeded in that. We saw it as worthwhile to make this application marginally easier to get through (in part because it wasn't a ton of work to make this app).

Also, these days I think of the tier app as a MVP. It demonstrates that there is a lot of demand for help in navigating the enrollment process, and I think an effort that did that would be really popular and make the process much more equitable.

[bensheldon]

However, I did see that, on the parent blogs, the tier system was something that parents were really interested in and trying to understand.

How did you narrow down your universe of possibilities to wanting to build an app that targeted schools/parents/children? Especially since (I don't think) any of you OpenCity-ers have kids. Cause I think it's really important to point out how you were "engaged" (gah, why did people have to make that term bullshit!) with the community by seeking out the blogs/forums where parents were talking.

Also, these days I think of the tier app as a MVP. It demonstrates that there is a lot of demand for help in navigating the enrollment process, and I think an effort that did that would be really popular and make the process much more equitable.

That's a really cool statement: splitting the "process improvement" from the "behavior change". I think a challenge of a lot of civic apps is that they try to change behavior and fix a broken process at the same time. You're demonstrating "this is a problem that can be addressed with technology" (which I don't think we can take as a given for so much in the civic space) before you try to optimize/generalize the actual technological intervention.

[fgregg]

In Chicago, schools are an enormous public issue, and regularly in the media. But, I don't really remember why I started looking at cpsobsessed.com. Once I started I started looking at the tiers, and it really seemed like a no-brainer. The right process was clearly an address lookup tool, which Derek already had with his searchable map template. Originally, that's all it was going to be, but we got very useful push back from some of our collaborators that we needed to explain what these 'tiers' were.

They were right, but the research and the writing of the explanation took 3-4x longer than the technology part of it. The tech part took almost no time at all because we had all the pieces already.

[Mr0grog]

They were right, but the research and the writing of the explanation took 3-4x longer than the technology part of it. The tech part took almost no time at all because we had all the pieces already.

This could be another good pattern: often the biggest reason better tools for civic processes are needed is really just because the process is complicated or obscure and has never been clearly explained. Sometimes a guide or explanation is all you really need! OpenCounter in Santa Cruz was a good example of both of these things.

[jpvelez]

+1

I tend to the think of CPS tiers as improving the usability of government. On that count, explainers are incredibly important, but so are old-school design principles of keeping things simple and not hating your users.