Creating a web application

In June, our driving theory test website www.passyourtheory.co.uk celebrated its second anniversary with a brand-new site design. As we plan for the future of this web application, I’ve spent some time reflecting on what we’ve learned over those two years, during which time Passyourtheory (PYT) has, after a slow start, become a profitable, mature product with an independent business identity. As one of its authors, it’s easy for me to become obsessed with the detail of web application development, but all that effort would be wasted if we were heading in the wrong direction. In many ways, actual coding is the easy bit, since programming is governed by a set of inflexible syntax rules that are burned into the subconscious of all experienced coders, whereas planning is a less well-defined process that is, frankly, far less comfortable.

This month, I want to begin by sharing with you my real-world experience of planning, creating and running a successful web application. The aim isn’t to cover any of the issues in full depth, but rather to point you in the right direction if you’re interested in web-application development. Let’s start at the beginning: the planning process.

The idea

The most crucial step is the first one, coming up with a good idea, and the best source of ideas is your own experience. For example, I came up with the idea for PYT when standing in a queue at WHSmith – in front of me was a young woman clutching a CD-ROM that covered the theory test, and it struck me then that this service could be much better delivered online. Why? First, because access could be granted instantly, thus saving that trip to the shops; second, because a web application is much easier to keep up-to-date than a CD-ROM; and third, by charging for access based on the length of time it’s used, we could offer prices that suited each user’s need rather than a single fixed cost for everyone.

Usually a five-minute search on Google is enough to discount any new idea that comes to me, because the likelihood is that this niche has already been occupied. In this case, however, I found only one existing competitor, which puzzled me. On further investigation, it became obvious why: the theory test a learner driver sees on test day is made up from questions randomly drawn from a pool of more than 1,200, created by the Driving Standards Agency. The text of these questions can be licensed from the DSA, but this isn’t a trivial process and it involves negotiating a contract that must be countersigned by the Minister for Transport. A royalty is then payable on all memberships, and it was this double-whammy of a complicated licensing process and having to pay royalties that had put off others. Plus, of course, the time and cost involved in building the web application itself.

PYT is an example of an information-based web application. A theory test costs £21.50, but around half of all tests are failed, so many people consider an investment of £2.99 represents good value. Other examples include the quite wonderful Lynda.com, which provides online application training for a fixed monthly cost and, of course, Britannica and Encarta.

Google Docs, on the other hand, takes a different approach. It offers word processing and spreadsheets, previously available only on the desktop, and delivers them online. Much of the functionality of Word and Excel has been lost in translation, but other functionality that takes advantage of online delivery has been added. This is the key: if you’re thinking of creating a web version of a desktop application, there needs to be some overall benefit to the end user. It’s unlikely you’ll be able to provide all the features of Photoshop, for example, but can you compensate for the lack of features (in itself an advantage for some users) by adding, say, user commenting or sharing capabilities.