Fixing forms is easy…no?

We’re not sure who coined the idiom first, but it’s definitely true to say “the less you know about something, the simpler it seems”. A good example is other people’s jobs. Have you ever asked someone what their occupation is, only to find it hard to imagine that they have enough to do all day? It’s only when you probe further that you get an appreciation of just how complex different jobs can be.

Forms are no different. On the surface, designing a form that is efficient, effective and satisfying seems like something almost anyone could do. You make it short, ask clear questions and pretty it up a bit, and Bob’s your uncle!

In reality, the process of designing and implementing a great form often means overcoming some serious obstacles. In this article we’re going to step away from specific design guidelines for a moment to explore some of the less tangible challenges you are likely to face. These are the sorts of things one learns the hard way from experience…unless you have an article like this to save you the trouble!

Design is a balancing act

Search the web and you’ll find plenty of articles listing the “X” things you should do to design a good form (including some of ours!). Without a doubt, these guidelines can be very useful; if you’re working with forms it is important to know the principles that result in a positive form-filling experience. What most of the articles don’t say is that as a designer, you’ll regularly encounter situations where two or more guidelines are in conflict. For example, conflict may arise between the desire to keep the form as concise as possible, while still being clear about what the form-filler needs to do.

In our experience, the most common tension is between consistency and suitability. Consistency is an ideal, regardless of whether it is between different forms from the same organisation, or within a single form, because consistent experiences facilitate learning. That is, if your forms are consistent, it will take users less time to work out what’s going on, and thus less time to complete the form. But of course there’s no point having a consistent form if it isn’t suitable. Suitability is about the extent to which a form serves its intended purpose.

Figure 1: Website banner logos for three agencies in the Australian Government. The requirement to use the crest and the words "Australian Government" as shown helps consistency but harms suitability. Users have to work harder to identify the agency than they would if each agency had their own unique logo.

Clearly, bringing consistency to a suite of forms may mean the suitability of each individual form in the suite is reduced. For example, an organisation may insist on collecting addresses in the same way on every form, yet one particular division needs addresses in a different format (perhaps for compatibility with a legacy system). Conversely, if every division is allowed to collect addresses in their own unique way, the poor form-filler will not be able to use their experience with one form to speed up completion of another. This will reflect negatively on perceptions of the organisation’s professionalism, trust and desirability.

Getting comfortable with the need for compromise, even while following established best practice, may require some personal growth. We designers — especially those who also do user experience — tend to be perfectionists with high ideals. This perspective can work against us, though, as there is no perfect design. By definition, the diversity in humans and organisations means no single tool is going to work all the time.

Your respondents aren’t always going to do the right thing with your form anyway. They are humans, not machines, so they will do things like skip over instructions and self-select based on section headings. Therefore, even if you could design something that was perfect — from a first-principles point-of-view — human nature would lead to less-than-perfect data. All the more reason to be prepared to compromise.

Fear is the enemy of progress

Fear of change

The aim of producing quality forms is to make everyone’s job easier: not just the form-filler’s job of completing the form, but also the organisation’s job of using the form answers to provide goods or services. Given this, you’d think your colleagues and clients would need no convincing about the value of quality forms. Yet in reality you’ll rarely achieve full and immediate buy-in for your form project.

The problem is that redesign of a form or associated process means that things are likely to change. Change is scary and unknown. Humans have a natural — evolutionary, even — aversion to change because predictability means safety. So even if the change means things will (at least eventually) be better, until then it is likely to meet some resistance.

Again, we can make the best out of this situation by adjusting our perspective on the project, and accepting that any improvement is a step forward, even if it isn’t quite the huge leap we had wanted to take. Initial, incremental changes will demonstrate their value, making more significant changes feasible over time.

User testing is also a great way to really bring the problems with existing forms home to stakeholders. In our experience, observing a couple of user sessions, even if it’s only 15 minutes of each, can do much more to convey issues than reams of reports ever will. When stakeholders see first-hand the extent to which respondents struggle with a form, it is hard to deny that change is warranted.

Fear of removing questions

Fear of change is a factor in any design project, no matter what the output. There is one fear, however, that is specific to the redesign of forms: the fear of removing questions.

Before an existing form is redesigned, you’ll need to do an inventory of its contents. This means identifying how each question is used, and by whom. Sometimes this process will identify a number of questions that are apparently not used for anything, by anyone. So what should you do with these questions?

Every question on a form incurs a cost: more questions means lower completion rates; lower user satisfaction; and more errors; not to mention greater expense in the case of printed forms. In addition, every question has a corresponding entry in a database. While storage costs may not be such an issue these days, database items need to be managed over time (which takes time and money). Databases are also likely be subject to privacy and other legal constraints (which also take time and money).

Figure 2: A mandatory title field is an all-too-common example of a question for which organisations probably pay too high a price. In this case, the user is registering a new appliance and so their motivation to complete the form is low. It’s likely that many customers give up at the start, as soon as they see that they are required to choose a title. The organisation needs to seriously examine whether having title on every entry in the customer database is worth losing a number of registrations.

Given all these costs, the obvious response to identifying unused questions is to remove them from the form. And if the resulting data is truly not used at all, there is no good reason to keep such questions. However, the failure of the inventory to identify any users doesn’t necessarily mean the questions are no longer needed.

For example, we have seen cases where data from a certain question is used just once a year. Therefore, depending on when and how you conduct your inventory, uses like this may or may not be detected. In particular, staff turnover often means the (new) employees you’re talking to about the form don’t (yet) realise that they will need certain answers from it.

So, you may need to keep some questions for a year, or a single sales/service “cycle” (if the business has one), before you remove them. But make sure they don’t end up staying there for perpetuity: enforce a strict deadline so that everyone knows the questions will be dropped if no legitimate need is identified before that time.

A final word on removing questions: “it’s just always been there” is not a legitimate reason for keeping a question in a form. Forms should reflect the needs of the organisation that owns them. Organisations are continually changing, so their forms should too. After all, you wouldn’t keep using a typewriter just because that’s the way things have been done in the past.

A form is just a tool for doing something else

The principle of the form changing with the organisation is an important one. It reminds us that forms don’t have a purpose beyond the organisational goals that they help achieve. Only the truly passionate folk like us here at Formulate collect forms, and even then we still don’t hang them on our wall like art. For everyone else, the form is just a means to an end.

Consequently, when you get stuck on a form design or process issue, the answers can usually be found by stepping back and asking:

“What are we trying to achieve here and what do we actually know about the situation?”

Come back to the original purpose and see what would best respond to that need. Sure, you’ll have to make compromises, balance priorities and manage fear, but you’ll also achieve something wonderful: fixing bad forms.