Challenges

A few challenges came up during this project. Because this redesign is only being rolled to iOS in the short-term, inconsistent platform parity between iOS and Android would potentially impact marketing and customer support efforts. Users may realize the visual differences in promotional campaigns and be confused or upset. On top of that, both platforms are leveraging same areas of the backend for different elements and needed to be heavily tested during QA to ensure proper alignment.

Approach

This redesign was part of a company-wide initiative to drop everything on the roadmap and work together in attempt to increase conversion rates in just a few quick sprints. During this time, we were able to borrow top talent from across the company to work on solving problems and creating a better upgrade & downgrade experience for the user. Due to limited bandwidth, we were only able to deploy these updates to iOS at first.

Internal Research

After looking at current fallout point data, we identified that there was an 83% fallout between starting population and upgrade, significantly impacting conversion rates. We validated this fallout data with existing customer reviews.

Due to these fallouts, we saw the average weekly cohort request rates drop 76% over the course of almost one year.

Discovery

Collecting and analyzing competitive data allowed us to make more informed decisions for the redesign. We mainly evaluated visual design language, UX copywriting, content, and timing of marketplace products that have high ratings & a large user base.

Inspiration from successful companies like Spotify, Strava, Lifesum, Headspace and Duolingo led our direction.

We researched SaaS best practices and took them into consideration to structure a streamlined and uncomplicated flow–hammering further simplicity with the reminder that "you don’t need more space, you need less stuff."

Testing

Utilities recruited customers for us to conduct on-the-ground user research to validate causation.

VOC Surveys, Behavioral Analytics, and A/B tests.

Ideal User Flow

After mapping out our current flow from post-login to Subscriptions, we found that we were not only lacking in pathways to Upgrade, but hadn’t even considered all potential edge cases to begin with. We worked towards several ideal user flows and came up with one that checked all of the boxes–offering more routes to the end goal, and a business-driven exit ramp that still remains user-centric.

Solutions

As part of the effort to holistically improve the Customer Journey experience, we integrated a smoother user-centric path to the Subscriptions flow (both upgrade & downgrade).

Additional Upsells

After analyzing the most-used pathways to the upgrade flow, we found that the main 3 routes came from upsells on our Usage page, My Devices page and the Menu page (where Subscriptions live). As these have proven to be useful entryways to Premium, we aim to have a small reminder to upgrade on every page.

It’s clear that users will pay for a little convenience. This means that in order to increase engagement, the benefits of the Premium plan needs to be extremely clear right off the bat.

We offer an upsell in the Onboarding flow that displays clear value props alongside an actionable CTA–if the user declines, they will see continuous soft sells throughout the experience (most dismissible). These soft sells are well-timed (triggered by user behavior) and strategically placed, so they become persuasive to those ready to upgrade, and unobtrusive to those who aren't.

The majority of soft sells reach the user only when the content will add value to their experience. Messaging shifts from general value props to tangible use cases, pushing for deeper user personalization, and helping familiarize features by showing a glimpse of functionality.

On the Usage (landing) Page, if the user hasn’t scrolled to where UI-related soft sells may be hidden after two visits, a notification banner upsell will display .

Redesigned Upgrade Flow

Plan names had previously been Lite, Live, & Link, reflecting the base plan, the plan to connect an Energy Bridge hub, and the plan with smart home capabilities. This was an obscure and unnecessary naming system, but we were able to persuade stakeholders to get behind more recognizable terminology. Plans have been simplified to Basic & Premium, alleviating the user from avoidable confusion.

We removed the ability to choose between multiple plans.–now the terms ‘Upgrade to Premium’ and ‘Downgrade’, are used in place of displaying side-by-side options.

The Address Confirmation page being it’s own step in the flow was previously responsible for 5% of authentication fallout. We removed this page and now pre-populate user’s address on the Confirmation page, allowing them to move on without second guessing, but edit if need-be.

Messaging and iconography in plan descriptions were vastly simplified, and CTAs now infer little work is required of the user. We figure that if someone needs to click around, or needs some kind of dictionary to understand what’s being shown, we’ve already lost them.

Redesigned Downgrade Flow

We added several steps to our cancellation flow, as to provide a more interactive concierge experience. Since the flow has been elongated, there is a clear option to cancel in each step as to not upset an already out-the-door user.

We’ve added continued learning, live support, and a peace-offering to provide relevant information in hopes we can help solve the issue prior to cancellation.

Copy was carefully thought out, wanting to show value of the product while still respecting the user. We also let them know we’ll be around if they choose to come back.

Technical improvements and additional error states

Users can be rejected from getting a hub for various reasons. For the past few years, we never notified them of their rejection, an experience that left our users in the dark.

Now, once the user has hit confirm, a dialog is displayed, informing them of their rejection and potential next steps to take.

Marketing

Utilities contribute to relevant promotions such as social advertising, customer emails, and App Store/Google Play designs are all updated to reflect the new ‘Upgrade to Premium’ model.

Results

Much like the Customer Journey flow, this product was only deployed on iOS. After one month, request rates increased by 25%, which represents a large expected net impact in Hardware/Software revenue over the next year.

In addition, iOS had been traditionally lower performing than Android with respect to conversion rates. Of the 73 weeks from January 2019 to May 2020 (just before launch), iOS only led in request rates 27% of the time, with only one instance of two consecutive weeks. Post-launch, iOS has had a higher request rate than Android for all 4 weeks.

Next steps

• Roll to Android. The impact to iOS is evidence that the new upgrade flow has an improvement on conversions and although the population of Android users is less, the total net impact to Android over one year can be expected to be more than that of iOS, justifying the effort.

• Evaluate and finesse the upsells & soft sells we decide to display

• Gather user feedback on the experience and make necessary updates