Content note: This project had to do with allowing people to buy sexually-transmitted disease testing online. While this portfolio item doesn’t include any discussion or images of diseases, we have found that some are not comfortable with the brief mentions of STD testing or the client’s characterization of their target audience. If that includes you, please skip this item.
The client wanted the website flow redone to match their new call center script, which asked a variety of (rather personal ) health questions. They only sent the call center script and our instructions were to make the best thing we could. They told us their target customers were “sexually promiscuous gay men,” who would want STD testing but (according to the client) not want to see their doctor or go to a free clinic. The client believed they would want to pay cash to a lab service. This made no sense until we uncovered that the client was a startup funded by at least one health insurance company.
This portfolio item is the story of what a client wanted vs what was the best match for their customers’ real needs and likely behaviors. For this project, Ptype CEO Debbie Levitt was freelancing for a San Francisco agency.
Some of the Process and Deliverables
Flow and Wireframes – Version 1
Debbie had suggested a process that first walked people through the health questions, then suggested tests with pricing, then choose your lab, create an account, and pay.
If the invasive health questions were enough to drive people away, then let them leave early before they’ve created an account or chosen a lab. These should be before we ask people for their name, address, insurance information, or other sensitive personal details.
Flow and Wireframes – Version 2
When the client saw version 1, they were unhappy. They told Debbie there is no way their legal and medical advisers will allow the health questions to go first. We must collect personal information first and force them to create an account.
This seemed like something the business would want but something no user would naturally do. Some companies believe that if you can get the user to create an account, they feel committed and are more likely to convert. However, if creating the account drives them away, then what have you really accomplished?
Debbie used progressive disclosure to make the personal info questionnaire seem shorter (and get longer based on your answers as you went). It was still a lot of very personal information that Debbie hypothesized most users wouldn’t be ready to give up.
User Testing Supports Debbie’s Hypotheses and Design
The client hired an outside company to do live, moderated testing with eight participants (separately). The client had decided that participants should first be shown their desired flow, Debbie’s version 2 (here named A). They would then be shown her original flow (here named B) to see which they preferred.
The testing showed that Debbie’s flow was absolutely the one to use. Additionally, despite not being directly asked, all participants volunteered that they wouldn’t use this service at all. Why pay cash for STD tests when they can go to their doctor or a free clinic?
Clients and UX are sometimes at odds, but the final word should be what is learned from user testing, and it certainly was here.
Flow and Wireframes – Version 3
The project ended with Debbie producing a final version based on making changes to the first, original version.
Ultimately, this is also a story of a client who didn’t bother with UX personas. They created a target audience that wasn’t real and didn’t exist. They assumed their audience was so embarrassed to get testing that they would give up any personal information and pay any cash amount to get testing.
They were wrong and user testing validated that. They were unfortunately wrong about a lot of things. They tried to pivot their company into a lab finding service. They closed shop in 2018.