Tekz (redacted) is a clean energy provider that serves the Texan deregulated market (residents can choose any eligible provider).
Summer is coming, and so do complaints about high bills. Tekz doesn't know how their customers engage with the weekly emails that are supposed to inform their usage. Are these emails useful? Do they read it? What do they do after seeing the notification in their email inbox?
Investigate the current email's pain points to deliver a redesign that offers "more" to different plan groups and significantly improves helpfulness ratings.
My role: UX research, UI design
Collaborator(s): Design and Customer Retention teams
Figma
UserTesting
Going into this project without any analytical data for the current email was a challenge. The requirement for the redesign was based on UI inconsistencies and overflow of numbers that don't point toward a next step. Currently, the weekly email shows them what they used last week and projected usage for this week in kilowatts per hour (kWh) + the difference between the two. For more data in a bar chart format, the user would need to sign into My Account (online portal) or click on "View in browser" hyperlink.
But being in the dark brings up a lot of questions for me. What do customers think about this email's presentation? Is it helpful in any way? Let's find out.
My design instinct pointed towards a need for giving users "more" – more useful data, more account insights ("Account Snapshot"), more things to do after seeing the email. To test this out, the design team conducted 2 rounds of A/B testing to (1) inform the initial design and (2) finalise the new email for development. I will focus on the 2nd round as I took full ownership of the test and the final design.
My hypotheses were the following:
The 2nd A/B test had a unique comparison matrix as it tested 2 sets simultaneously – current v. new email and Tekz users v. non-users. This was done to remove familiarity bias among current users and test a plan-specific feature that is believed to be helpful to bill credit plan holders. Due to platform limitations, I had to create 2 separate studies for each user group on UserTesting.
Round 2 - Tekz users group methodology:
Round 2 - non-Tekz users group methodology:
Study highlights:
As suspected, Tekz customers gave a higher rating for the current email due to familiarity (6.4/7 v. 6.8/7) but 8 noted that the visuals gave a much clearer picture of their usage and their impending bill:
[Account Snapshot/AS] is easier to read. The way WU's presented is hard to read, a lot of numbers. Not everyone has the brains to grasp statistics like that. I really like current usage and current cost and how much you need to hit that threshold. Most people don't know how their electricity works, just sign the contracts. Most people are surprised when they get their bill and think if they use less, they should have lower bill. But it's not always the case. Some people have contracts, people rely on hitting that big threshold. AS did loads better than 2nd email.
Although Round 1 solved major pain points, there were still areas for improvement:
I worked closely with customer retention team to understand the inner workings of Tekz' suite of plans and each group's energy priorities. It presented a challenge for me as it meant the "more" of the Account Snapshot should be tailored to maximise the email's value. "Your Cost So Far" chart and its phase 2 equivalent "Your Bill Estimate" are the key feature, so any special feature would logically be placed afterwards.
A/B tests prioritised Bill credit customers since it is a popular plan in Texas to use. Since they need to stay in-between 1000-2000 kWh at the end of each billing cycle, having "Your Usage Credit" chart that compares their current and projected usage help them adjust, whether to use less or more. Anywhere outside the range gives a warning in orange with additional messaging. Test findings showed that all current Bill credit users approved the new feature.
EV owners want to take advantage of off-peak electricity rates since they need to charge their cars (blue). A visual breakdown would best show how well they're hitting that goal.
Solar owners want to get ROI on their panels so knowing how much they have used and sold this current billing cycle would be most helpful. The data is also readily set up as it is used in Tekz app.
As I learned more from customer retention team, I found out that Tekz currently does not have any insights on how much thermostat tweaking saves on their customers' electricity bill. To ensure there is enough time to develop the emails, I had to scale down the scope for thermostat saver plans to omit any custom feature, like the fixed rate plans.
The new email will be deployed in 2 phases to accustom users and measure their engagement. Phase 1 holistically tells a story on their current usage, usage trends and plan-specific features, but will not include bill estimates until back-end implementation becomes feasible in the future.
Once the email's performance can be measured with current customers, the next step will be to introduce bill estimates in the mobile app to test its effectiveness before replacing the Your Cost So Far chart in the email.
The weekly email was a complex UX problem but that's what made it fun and rewarding to solve! The expansion of the email's scope to an "Account Snapshot" more accurately captured the users' need to know how they are doing in relation to their usage/cost so far in the billing cycle.
I worked hand in hand with development and there are many limitations that had to be considered, incl. chart generation, missing insights for Thermostat saver plans, and week segment width in Your Cost So Far chart (AS). Although there was set back with introducing bill estimates for phase 1, its value was proven by A/B tests and will be implemented in phase 2 with approved design (see Final Product demo).