Vodafone - Device Trade-In
Greenfield end-to-end device trade-in project from conceptual design through to MVP release
-
My role
I was the Lead Product Designer on the project, my time split across 50% hands-on design/research, 50% leading the team. My Lead responsibilities included:
Setting up processes & ways of working
Mentoring other team members
Reviewing and signing-off work for the rest of the team
I guided the team from project conception through to MVP delivery and beyond.
Background
A 2019 study estimated that as many as 40 million unused gadgets were languishing in UK homes alone, so the potential value across the world for users to trade in their devices is vast. Vodafone sought to enter this space with a trade-in proposition for Vodafone customers only, which aimed to increase acquisition and retention of Vodafone customers by allowing them to benefit from trading their device in and also generated stand-alone revenue from future refurbishment and resale of traded devices.
Project goals
Unlike the other competitors in the market who could only give users an estimated quote upfront (which may then be revised after the users send their device in to the warehouse for final review), we sought to differentiate ourselves by applying the following core principles to the proposition:
Provide users with a guaranteed quote upfront - this was to prevent scenarios where the user sends their device into the warehouse and then they get a revised quote once it is discovered their device is not in the condition they said it was. This would be a lose-lose situation for the user and business: unhappy users and increased costs for the business with potentially damage done to the Vodafone brand at the same time.
Make the online diagnostic journey as quick and simple as possible - to improve user experience and increase conversion rate as much as possible.
Device diagnostics
The user experience that most competitors provided was to just ask users what the condition of their device was in a self-report style, which gave a quick but not guaranteed quote. Our key differentiator in the market was the technical capability in the back-end. The way we sought to provide users with a guaranteed quote was to:
Conduct some clever back-end technical assessments of the device while the user was completing their digital journey - e.g. checking battery and CPU health.
Require users to take a picture of their device using a mirror - which would allow some clever algorithm to scan the picture for any cracks or scratches on the device. NB. we knew this was potentially going to be the most tricky part of the user journey in terms of usability so we had to make sure we did all we could to make it as easy as possible.
The device diagnostic part of the test was managed by a third-party technology company and so we had an additional challenge to manage them and provide them with UI specs to update their designs to align with the rest of the flow.
Process
Throughout the project we adopted a Double Diamond approach which worked well as this was a greenfield project and we could use it to draw upon valuable inputs from the rest of the delivery team. Not all members of the team were clued-up on Design Thinking so we took responsibility upon ourselves within the design team to educate and lead the rest of the Device Trade-In team and bring them along with us on the journey. We started out gathering insights from the business, existing Vodafone user profiles, and competitor analysis and brought them into a kick-off ideation workshop.
Competitor analysis
We started out with competitor analysis of the existing device trade-in market. Most of the trade-in companies out there had very quick and simple trade-in journeys that just asked users to select the condition of their device and then provide the user with a quick quote. Our hypothesis was that while this was good for conversion rate, in the long run this might not be good for the user or business as they risked being disappointed after sending their device into the warehouse and getting a revised quote when the devices handlers find out the device wasn’t in great condition. This means users could potentially get disappointed and company costs go up through having to send the device back.
Kick-off workshop & roadmap planning
To define the future roadmap we initiated a multiple disciplinary ideation workshop with the rest of the team. Together we were able to:
Absorb the findings and insights from competitor analysis and user research
Define our proposition, market positioning and ideal user experience
Ideate and generate as many cool product features as we could using lateral thinking methodologies inspired by Edward De Bono
Prioritise product features and define roadmap for MVP and future releases
During the session we broke the group out into smaller groups and used lateral-thinking techniques such as provocations to open people’s minds and think without any restrictions. This helped us to come up a high volume of product features that addressed the user needs. We voted on the best ideas and grouped them into themes using an affinity map. Then we went through a feasibility-checking process with the whole team to identify priority features for each phase in the roadmap. Everyone had to assign a value to each idea for:
User value
Business value
Technical feasibility
Understanding the user
To start to build up an understanding of user needs we conducted some up-front user research, interviewing users who had previously traded-in their devices with competitors. Our goals of the research were to understand users’ motivations before going to the trade-in site and to understand what their pain-points were during and after the digital trade-in journey. Some user insights that impacted our thinking about the direction of the product were:
Disappointment after quote revision - Our hypothesis was validated as we discovered users liked having a quick and simple online trade-in journey but some were also frustrated when they completed the online trade-in journey but then later discovered their quote had to be revised after sending through to the warehouse
Multiple devices - Some users have multiple devices left in their draw and would like to trade-in multiple devices at once
Devices that don’t work - Figuring out how we could allow users to trade-in devices that didn’t switch on anymore was something we hadn’t thought of a part of the proposition so food for thought in future releases
Environmental motivation - some users indicated the value of trading their device in as part of a general benefit to society; recycling old phones, giving them a new life and helping to save the planet were all positive by-products of trading in their device for money
User flow mapping
We took the requirements from the kick-off workshop and started mapping-out a basic user flow of the potential user experience from beginning to end. This was useful to help us think about the bigger picture (service design) as well as running our thinking past product owners and back-end developers at an early stage so as to validate business and technical feasibility before progressing to the design phase. The key phases of the user journey were:
Entry point via My Vodafone App > Device selection and diagnostic check via web > User gets a quote then sends their device off to the warehouse for processing an quality checking > user gets final confirmation of price once device is checked in warehouse
Sketching / design jam
After the requirements workshop the design team had a design jam where we were able to rapidly explore multiple low fidelity design ideas in quick succession. We brought all our user insights and materials from the competitor analysis into the session for inspiration. Low fidelity sketching was useful to help us quickly try out new ideas and iterate. Doing this in a group helped us bounce ideas off each other to inspire further creativity.
The outputs from the session could also be quickly communicated with stakeholders in the team to validate business and technical feasibility. A lot of the ideas we produced during these design jam sessions were ‘future thinking’ ideas that would not necessarily be feasible for MVP release but would still be valuable for future releases on the roadmap. Some of the features we kept for future releases were:
Single Sign On - so the users didn’t have to log in after leaving the app
Auto device-detect - so the user didn’t have to manually input their device at the start
Auto-detect IMEI number - so the user didn’t have to manually authenticate their device
Rapid prototyping & usability testing
We wanted to validate our sketches and design ideas asap so our next step was to create a clickable prototype of the user flow which we could then test with users and iterate the designs as necessary. There were several rounds of user testing and design iterations throughout the project and we triangulated methodologies across moderated and unmoderated tests for qual and quant. Overall we conducted:
Five rounds of unmoderated usability tests using UserZoom (approximately 75 users in total)
Two rounds of moderated usability tests (10 users in total)
I was a bit unsure about using unmoderated usability tests at first as they lacked the qualitative insight of moderated user tests (where you observe first-hand, question and probe the users in real-time). However, on retrospect they worked very well as part of a triangulation of methodologies for the following reasons:
Speed of results - we could upload a study at the end of the day and get the results in the next day
Quantity of users - it was possible to gather feedback from a larger number of users
Natural environment - users were not in a lab setting so they could conduct the study at home as though they would in a real life scenario
Consistent delivery of questions and tasks - every question was given to users in questionnaire form so the delivery 100% consistent, impartial and so therefore not prone to moderator bias
Animations to improve usability
One of the rounds of usability testing indicated that some users were having some difficulty with the mirror screen test. Of all the five diagnostic tests it was perhaps the most tricky as the user had to take a picture of the phone using a mirror. In order to improve usability of the diagnostic tasks in general (and conversion rate as a result) we took the initiative to commission the design of a series of GIFs which would be positioned at the start of each diagnostic test, demonstrating to the user what they were required to do on the next task. Our testing showed that as well as improving usability the GIFs also improved engagement and user experience by providing an element of playfulness. Here are some examples of the GIFs we put into the journey.
Higher fidelity designs
As we progressed through multiple design iterations and user tests we continually captured and documented the required functional behaviour of each part of the user journey. Below is an example of higher fidelity designs for the MVP launch that had been validated by users, the business and the technical team. We used Sketchcloud to host the designs and share them with developers who could then inspect the design specs before building.
More design iterations
There were many design iterations throughout the project, some happened as a result of usability testing, some as a result of changing business requirements, and some as a result of descoping for MVP. Here are a few examples of some design iterations.
Device selection
We explored multiple ways of allow the user to select their device. The initial idea was to prompt users to select manufacture, model and memory separately. In the end we went for a search field option with predictive search results as this offered the best usability. However, it wasn’t easy to persuade the dev team to take this onboard as there were some bespoke development features that needed to be setup to make it work, but we used findings from user research to demonstrate its value.
Test visuals
The visuals for the tests went through various iterations, eventually resulting in GIFs that demonstrated to the user what they needed to do next. We found in user testing this helped assist with usability.
Quote summary
The initial quote summary page had some results feedback from the test with ticks or crosses next to each test so the user could see some indication of what might have affected their quote. User insights showed this to be a valuable feature but it had to be descoped for MVP as it would take some technical effort to build so it was put in to a later release on the roadmap.
End result…
We took the project from concept design through to MVP in 9 months, which was an achievement considering the corporate nature of Vodafone and the speed at which projects typically move.
The initial design was released in Germany and since then we have been monitoring the analytics of the user journey to track conversion rate, task completion, general user behaviour and any areas of considerable drop-off within the journey.
Once we have enough data from the MVP release we plan to review the findings together as a team and plan for improvements and iterations in the next phase later in the year.