In an increasingly competitive car-sharing market, the need for customer loyalty is growing. SHARE NOW having been the first to market over 10 years ago was in the luxurious position of having a significant user base with about 1,4 million active users. To ensure users stay loyal to SHARE NOW as well as to attract more users in the future, a rewards program was created...
2020 - 3 Month
Responsibilities
UX Concept
Wireframes + Prototype
Visual Design
Stakeholder management
Tools
Pen + Paper
Figma
Stakeholder
UX Research
Engineering
◇ Starting Point
A lot of conceptual decisions had already been made before the product team got involved. Rewards and the mechanics of how to earn them were decided and a strong preference for colors and related tiers names existed from our marketing partners.
What had yet to be done was a verification of those assumptions through user testing as well as an evaluation of technical feasibility.
The timeline for the validation was tight as user testing had not been viewed as essential initially. However, I raised concerns about the pull effect and was able to negotiate a few weeks for qualitative user testing. We quickly scrambled to create testable UI prototypes to fit as many rounds of testing within the tight window as possible.
Some pain points to overcome and risks to mitigate were identified:
- Is the rewards structure of points and tiers understandable and does it create the desired pull effect (make unseres use our service over competitors’) with users?
- Internal tools for communicating with users are very limited → how to create awareness?
- To comply with Germany’s GDPR rules and the latest T&Cs, an automatic opt-in for the rewards program was not possible.
The Mechanics of Getting Rewards
◇ The Goals
◇ Initial Concept
Tiers
Based on extensive market research our marketing team created 4 tier name. They were expecting to convey an inherent hierarchy. The expectation towards design was to find colors within SHARE NOW’s schemes which revolves around shades of blue. This was a challenge and we needed to test whether or not users would understand that e.g. Royal Gold was a shade of blue.
Assumptions
As the availability of rewards is dependent on a person’s tier level, the general assumption was that people would feel encouraged by moving from one tier to the next to unlock more/better rewards:
- Displaying the tiers in a pageable element in the top section will encourage the user to drive more in order to move to the next tier.
- Displaying the available rewards underneath will provide positive reinforcements to the user for using our service.
UI details
◇ Testing
Round 1
We wanted to validate the thought up mechanics as well as get a better understanding of the perceptions of the tier names and colors in context.
With the first iteration ready for testing, I collaborated with our UXR team to support me in a first qualitative study.
Scope and test setup
→ Play with Prototype
Learnings
Users lack the emotional investment to care about gaining or losing a tier as there is no emphasis on the return on investment.
The focus on tiers weakens the relationship between points and rewards.
Tier names and color schemes were not supported by existing mental models.
The user’s goal is always to learn more about what the program will do for them. Instead, they are guided to first understand the tiers.
◇ Iterating
The findings from the user testing helped us identify some key elements to improve on.
Sketching and wireframes
When it comes to quickly iterating with potential improvements, pen and paper still is my weapon of choice. Drawing a sketch and getting fast feedback from peers as well as colleagues far away from the project helped narrow down the options.
UI
Supporting the unusual tier name / color combination
Attempt to support the tier name and colors provided by creating a conceptual watermark pattern.
Header
Using the header of the dashboard screen to help the user understand their status quo in terms of tier and points.
Reward cards
With reward cards indicating potential gains or losses in rewards, we wanted to learn if this would incentivize users to drive more with us.
Teaser Page & Confirmation
Shifting the focus from tiers → to rewards on the teaser page could cater to the non-member users. Again highlighting the newly gained rewards right after sign-up should emphasize the benefits gained to freshly converted users.
◇ Testing
Due to severe time pressure on the project, we quickly moved to prepare another round of testing. Our assumptions around the next iteration were based on internal testings and discussions and needed to be further validated.
Round 2
→ Play with Prototype
Learnings
Participants found the new teaser page much clearer but were still a bit overwhelmed. Important information that was either missing or overshadowed included how to earn points, the Initial starting costs of the program, immediate progress, and the rewards themselves.
Participants were able to better understand the link between their investment and the rewards they received.
Participants demonstrated far less confusion when discussing attaining rewards through the tier level, and were quickly able to identify their status and necessary next steps to maintain a level, gain a new level, or prevent loss.
However, users still struggled to build a clear mental model of their progress with the program. Users were motivated to maintain their levels but missed instantaneous gratification.
◇ Iterating
We took the findings and created another iteration with which we went live. We were confident that the improvements made with this version were a significant step to solving our set how might we‘s. Further iterations would need to be based on not just qualitative, but also quantitative research.
After the interview sessions, we were able to convince the marketing and branding team to reconsider the tier names as well as the constraints around color choices and come up with more options supported by existing mental models and easy to grasp.
→ Play with Prototype
◇ Implementation
We started the implementation without any further testing to make up for some “lost time” (for the better). Unfortunately, a few features did not make that cut and will have to be brought in at a later stage. The most noteworthy one would have to be the lack of nudging. Even though it performed extremely well during testing when it came to the question “Would this motivate you to drive more with Share Now?”, we had to out-scope it to reduce complexity with our internal microservices. It certainly would be a great addition to drive more user behavior changes in the future…
◇ Success?
Looking at the data coming in through real data points and with our initial How Might We questions in mind, we quickly realized where we were successful and where we needed to maybe dig a little deeper.
Here we could see an increase in revenue of 20-22% (depending on the market) with customers who have signed up for the rewards program. So that is a stellar success.
On the other hand, with a sign-up rate of only 28% of our existing users, the area for improvement was clear.
We are taking a multi-angle approach to improve here:
- Give contextual hints about the existence of the rewards program (e.g. at the end of a rental stating how many points could have been earned by signing up)
- Improve the visibility of the program for all users
- on the Map directly (default screen when opening the Share Now app)
- In the Menu (with rewards not being the main focus of the app, the account menu remained the designated entry point for active users and soon to be’s alike)
- Opting into the rewards program when registering with the service
◇ Next Steps
While providing contextual hints to the users is being worked on by a fellow designer under my supervision as I am writing this, it is too early to say how valuable this avenue will turn out to be.
We have since iterated on the entry point for the rewards program making it more prominent for members and non-members alike. We have just finished implementation, so it is too early to tell for sure, but early indicators point to success with minimal engineering efforts.
A glance at the numbers
A sneak peek into the click rates for the rewards entry point shows that the new presentation had the desired effect of leading more people into the funnel for signing up.
◇ Key Take-Aways
Don’t UX test only when the train is about to leave the station
Sounds obvious and we all probably preach that day in and out. In this case, I found myself in a situation where testing was not done at an earlier stage. Forcing a few rounds caused delays in engineering and initial frustration amongst the stakeholders. The results proved us right though and the feature is widely considered a success.
Prioritize what is must have and what’s nice
With this project being under immense time pressure due to an unusual ramp-up as well as the late testings, shaping an MVP while iterating on the concept proved to speed up development time significantly (and with the feature being built as a web view, iterations are quick and not dependent on release cycles)
It takes a village
Coming to this point was quite an undertaking and I am certainly not the lonesome hero here. Without the extensive research and creativity from the marketing department, the negotiation skills of our partner sales team, and last but certainly not least the UXR department, this would not have been possible to turn around in the time and quality that it was