Online Applications shouldn’t make you pick a favorite child

A few simple questions to avoid this ↑

I’m sure there are a lot of people out there that have experienced system design and lived to tell about. I had a scrape with it this morning and I thought I’d share because I think it highlights some of the type of thinking that absolutely has to occur with more frequency in both the public and private sector space. I think anytime you are contemplating the delivery of a service you really need to be prepared to answer these three questions convincingly:

• Have I thought about the big picture?
• Have I thought about what happens if I’m successful?
• Have I thought about how people will actually use the system?

I will talk about this and use as context Arlington county Virginia’s registration system. Arlington county is extraordinarily progressive in its provision of activities and facilities for its residents providing a wide array of programs that are, as I can attest to, extraordinarily popular. So popular in fact that I can rarely get all three of my children into any activities, which brings me to today’s blog.

Have I thought about the big picture?

I think a lot of organizations fail to think long enough about the actual use cases they will face when they implement a system. In Arlington county’s case, the system is clearly designed to help individuals sign up for activities. Unfortunately, this doesn’t help families who have more than one child. In my case, with three kids and only two parents, the inability to sign up more than one child at a time combined with the popularity of the classes and the poor workflow means that we have only once been able to actually have all three of our kids in the same class. Essentially, Arlington makes no provision for the needs of parents with more than one child as a use case.

Have I thought about what happens if I’m successful?

Anybody who has ever built a system or application to automate a process hopes to make a big splash and the same applies in the public sector. Arlington has a parks and recreation program that is absolutely exceptional. The depth and breadth of programs is amazing and the few programs we have been able to get our kids enrolled in have been exceptional. Unfortunately, those are few and far between because the combination of poor workflow and slow system performance have meant we almost never get to put all three of our kids into an activity. This means we either choose which kids we love most and send them or opt out of being able to leverage the capability my tax dollars support. Since this is a registration system, I am guessing that most of the traffic occurs just a few times a year, however I can also guess that Arlington currently pays for the resources and infrastructure supporting this system in a pretty level fashion. This type of requirement is perfect for a cloud based system and in fact, as I clicked through the screens this morning hoping my kids would be able to take swimming lessons together all I could think was: Salesforce…

Have I thought about how people will actually use the system?

One of the most frustrating things a user can experience is having to process multiple repetitive workstreams, i.e to order something more than once I need to move through the same workflow multiple times rather than simply ordering in multiples. This would never happen on an e-commerce site because people vote with their feet. In the public sector it’s a bit more challenging because the profit driver isn’t there forcing the issue. Arlington’s website asks you about special needs every time you fill out the form as a separate step rather than remembering or incorporating it into a user profile, etc. It also forces you to take multiple trips through the very slow queue to sign up for classes. This means my kids end up not getting to use these community services, because by the time I have kid number two signed up all of the slots are gone for kid three. If I could simply sign all three up at once, I’d be able to use the services that people with only one child get to use.

Anyway, I hope I stayed out of rant territory and provided a useful example of the type of thinking that needs to occur when you begin looking at process automation and application development. Nothing I said above requires a lot of technical insight. It simply requires the ability to put yourself in the users shoes before you start writing code and building systems. I’d like to point out again that I love Arlington’s services and for the most part you couldn’t interact with a friendlier or more devoted team, I just wish their technology matched the great service they provide so that even those of us with more than one child can enjoy them. I’m sure that most of you have had an experience similar to mine. I’ve been on teams that have made similar mistakes so I’m not guilt free in that regard. It is very hard to anticipate how users will leverage your technology and sometimes something that should have been obvious sneaks through despite the best efforts of all parties. What are the questions you ask yourself when you get ready to change the way your organization does business? How do you approach business requirements, use cases and scenarios?