What technique(s) would you use to evaluate and [re]define fields in a lengthy and complex form?

Question

We were commissioned the redesign of a web application for real estate agencies. During the first stage of our user research, which was a contextual inquiry, we found that the form fields (or data structure) for entering a new Property has many problems. There are fields that should not be mandatory, malformed fields, fields with incorrect options, maybe missing fields, etc.

We are looking for a user-research technique or method that allows us to evaluate the existing fields and determine the necessary changes in order to adapt the forms to current market practices and customer needs. The usual methods such as card sorting would not be very useful here, as it's not a matter of labeling or sorting the fields.

Since we have seen differences of opinion between the real estate agents and some level of consensus must be achieved, I suspect we should use a quantitative technique rather than a qualitative one, but we are open to all possibilities.

Any ideas?

(Context: This is an application for the housing market in a latin american country with thousands of real estate agents, closed to the general public, which has been running for more than a decade with little to no UI or UX updates).

Answer 1

You may find it helpful to separate business goals from user goals. In the case of a form, business goals can be used determine what information needs to be collected by the form; user goals determine how the user can enter this information easily and successfully.

For example, a business goal of your form might be to collect as many leads as possible. This business goal would dictate that first name, last name, and a least one contact method (let's say email address) be required by the form.

When using the form, one of the user's goals is to complete the form successfully. Since business goals dictate that first name, last name, and email are required, you'll want to make sure the user knows this so they can successfully complete the form. In order to help the user know what's required, you'll want to:

• Clearly indicate which form fields are required/which are optional to prevent mistakes
• Provide clear error handling to help the user recover from mistakes
• Provide validation to verify that the user's email address is valid, and provide clear directions for the user to fix it if it is not

After you've determined all of the form's business goals and translated them into requirements and user goals, you'll then be ready for usability testing. It may serve you well to try contextual inquiry again since you already have a baseline with that method.

No matter what you do, don't make the mistake of asking your users to identify your business goals for you. Users will serve you better by reacting to knowns than to describing unknowns--and you know your business goals best. Consider Walmart's $185 billion dollar mistake. Walmart relied on what customers said in a survey and ignored what they actually did in the stores, and they lost many billions of dollars as a result.

Answer 2

I would recommend the following approach:

Conduct a Usability Test with atleast 8 users. It should be focussed on how they define and interpret the various real-estate terminologies. What this means is, ask for:

1. What are the critical fields of information that need to be provided.
2. What fields are mandatory and what are "good to have"
3. What are the typical values encountered in these fields (this should give you an idea of the length of the fields)

8 users should give you a fair idea of the critical info to be captured, the field labels, lengths and the mandatory fields. Armed with this knowledge, create a new form. Create a help text for each field which you think is ambiguous or has different meanings for different users. This help text can be visible to the user when he highlights a form field, right next to it (eg. tool tip)

Once you have this new form ready, again test it with users. This user research should be a Survey. Give out the new form to atleast 50 users and look for confusing labels and terms. This can be your Quantitative data to validate your design. Incorporate changes that you get from the feedback from this survey and what you create should good enough for atleast 90% of your users.

Answer 3

Your question is confusing. You are asking designers that may or may not have real estate expertise to decide what is necessary information to complete a form for listing submissions?

I suggest you come to some kind of consensus within your field as to what is manditory, meaning "You will not be able to proceed without this information". From there you can determine 2nd, 3rd, etc tiers of information. These would be preferred fields and then peripheral information.

Answer 4

While I agree with the usability testing suggestion wholeheartedly, you could also conduct an expert review with just a couple of colleagues and the right knowledge of the key users. The beautiful thing is that you don't need any ACTUAL users to do this, and you can do it fast.

While I'm not a huge fan of heuristic evaluations (mostly because I think people often just grab Nielsen's list of 10 and try to apply them to everything; you really need heuristics that match your project and goals), there is definitely a place for them if done well. Just make sure you are using the right heuristics.

You could also employ something like the PURE evaluation (https://measuringu.com/pure/) which similarly allows you to evaluate the usability of a site or piece of software.

The important thing with an expert review, IMHO, is that you don't just do one in isolation. You will get the best results if you have two or more experts reviewing the same thing, then comparing their identified themes, issues, and severity and working through the final evaluation and recommendations together.