Introduction

Surprisingly, just 14% of customers enable self-service. Why?

This was an important question. Unanswered, it would miss an opportunity for 86% of customers. That was the real underlying business strategy: drive down the TCO for all of our customers.

I worked with my stakeholders to answer these needs.

Deeply embedded assumptions

The main barrier was thought to be “security risk.” I interviewed our Kit persona (i.e., IT Administrators and the decision-makers for my research topic, self-service) to validate this.

Kit said:

• “We’re not worried, we already use a form of self-service.” - IT Administrator

• “We would use self-service if we knew that existed.” - IT Administrator

This required a major shift in our thinking. If security risk wasn’t a barrier, what was?

Results

Crystal clear deliverables

I created a user journey barrier map.

There’s a neat backstory on why I chose that specific deliverable.

Duo has a mature research practice and a trove of documentation. I wanted to make sure I didn’t miss anything. So, in my first weeks, I conducted a literature review. This helped me build the formal and informal connections needed.

More importantly, I discovered an old journey barrier map from several years ago. I riffed on that, creating a new, evidence-based one.

Then, I socialized the new map into several workflows with managers, designers, engineers, and research on the Mobile, Prompt, and Admin Panel teams.

This led to several evidence-based, tactical design changes.

Far reaching design impact

Turning foundational insights into tactical design impacts is important but easy to miss. I worked closely with my product partners to align research with design.

Here are a few examples:

• Designing nudges: customers want self-service, but a main barrier is “where it is and how it works.” Design partners are now designing nudges in the Admin Panel to make sure customers know “where it is and how it works.”

• Unifying self-service: self-service lives in several places, with confusing dependencies, and this violates users’ expectations. Engineers are looking into fixing this to unify the experience. A very specific example is eliminating SSO-as-a-prerequisite to self-service enablement.

• Revisiting traditional prompt: self-service performs better on the universal prompt than the traditional prompt. Managers want to know if we should commit resources to improve self-service on the traditional prompt, even if it will end-of-life on March 30, 2024. Research is continuing to partner with managers to answer this question.

Adding new theoretical knowledge of our Kit persona made my research team happy, and creating evidence-based product plans made my stakeholders happy—but, you may be wondering, what does this have to do with the original brief?

Tying it back to the brief

Researchers sometimes revise original problem statements to reflect new understandings of “the problem” as they study it. That was the case here.

If lowering TCO for all of our customers was our business strategy, improving the end-user experience wouldn’t cut it. We needed to understand why only 14% of customers enabled this feature. There’s a great napkin sketch my supervising manager and I made to this point.

I didn’t leave the original brief hanging.

I answered the original brief by leading a preference test on UserTesting.com. This wasn’t the focus of my writing here because it’s a fairly straightforward task.

In this test, there were two options: the current experience and a new concept my design partner and I workshopped together. (Here are 5 of my initial concept sketches.) The main difference was we added self-service to the Mobile app. 7 out of 11 participants preferred the new concept, with self-service on Mobile. The team was excited. Implementation is slated for next quarter.

Last word

In this way, I met the original brief’s objectives.

• I workshopped and evaluated a new design concept to improve self-service for end-users (i.e., by adding it to the Mobile flow), lowering TCO for our 14% of customers with self-service enabled. This directly answers the original brief.

• I worked with stakeholders to demonstrate why this action is not enough: 86% of customers went without self-service, and we didn’t fully understand why.

• To close that gap in knowledge, I captured the motivations and mental models of enterprise customers in the Admin Panel through interviews. I uncovered the barriers to self-service enablement. This is a step further from the original brief.

• I delivered a user journey barrier map, and worked closely with stakeholders, to ensure that a solid representation of this research resulted in real, tangible benefits for users.

As I mentioned, several tactical projects followed (e.g., designing nudges, unifying self-service, improving traditional prompt) and are in-progress today.