Go back
Glovo
The goal of this project was to increase the number of couriers joining Glovo through the referral program.
March - May 2022
Prodcut Designer, Data Scientist, UX Researcher, UX writer, Product Manager, Software engineers (x3)
+ 900bps referral contribution to CFO (courier’s first order)
+ 200bps lead to CFO conversion
4 weeks (design) + 16 weeks (engineering)
In many of its markets, Glovo is suffering from a fleet gap problem i.e. there are not enough couriers on the road. So one of the top level business goals of Glovo is try to reduce this gap.
One of the most cost effective ways for Glovo to bring new couriers into the platform is referrals. However, the current system is built outside of the courier Glovo app, through email. This method was full of pain points and was not an ideal experience for couriers.
The goal of this project was to improve this experience for couriers and therefore entice couriers to refer more people.
Old referral user journey with pain-points
The project has kicked-off by conducting UX research and data analysis to gather extra information about the current referral program.
Research objectives:
Referred couriers had a higher retention rate
What could be improved in Glovo's referral system
Pain points around Glovo's referral system
See user profile canvas below to see the summary of pains and needs.
This acted as an extra reason of why we should dedicate more resources to improve the referral program.
I’ve analysed referral programs of the direct competitors as well as products from a different industries in order to understand what kind of functionality they were offering. I saw that most of the referral features were very simple and only included basic description and ability to copy/share the referral link. Very few for example included a feature to track the live status of your referral.
Analysing referral program solutions offered by the direct competitors as well as other products in the market.
Now that we had clarity on the user pain points and needs we wanted to tackle, it was time to kick-off the solution phase. I first began by organising a brainstorming session with stakeholders to explore the broad range of solutions to the identified needs and pain points.
Screenshot of different ideas that were presented during the workshop.
One of the most common dangers during the solution stage is to depart from the insights identified from user research. To avoid this, I have ensured that the ideas that carried forward were answering pain points/needs identified from the research.
Value proposition canvas helps to ensure there is a clear link between the identified user needs & pains and the proposed features.
It would be an unreasonable approach to build all of the listed features at once, as it would require an enormous amount of time and effort. In collaboration with stakeholders we have collectively prioritised features (gain creators/pain relievers) based on the perceived impact and engineering effort involved. Features that we believed would bring the most value to the user and at the same time involved less effort to implement were prioritised.
2x2 matrix used to help prioritise gain creators/ pain relievers. The ones that are marked with the star were prioritised.
Although, some ideas like increasing the referral reward we believe would have had a big impact on the user and would involve minimal engineering effort, these were consider out of scope as they would incur a high business cost.
Next, we could break prioritised features/epics down into smaller user stories and place them on the feature roadmap. This helps to build clarity of the scope of each iteration. Roadmap serves two main purposes:
1) Creates an alignment on what should the solution contain.
2) Priorities stories by deciding what goes into MVP and what would be implemented in the following versions.
Breaking down epics into specific user stories and placing them on the version roadmap.
Having an alignment on the scope of the solution enabled me to proceed to the next stage. During the wireframe stage, I first like to experiment and try multitude of different ideas. Experiments range from trying out different flows, layouts, information shown to the user etc. By going wide before narrowing down on a solution, I maximise the chances of finding the better solution. Below you see an example of different options I have considered for displaying the status of the referral.
Exploring different ideas for showing progress of the referral. I went for the Option C, since it allowed to view progress of multiple referrals at the same time.
Having explored variety of options, I have then presented several to the design team and stakeholders for feedback. After receiving feedback internally, I have iterated and narrowed down on the options.
To validate the solution further, I have conducted a very quick round of guerrilla user testing by speaking to 5 couriers. During these sessions I have shared with them interactive prototype, gave small missions and asked for feedback. Overall, users perception of this solution was positive.
I have taken on board the feedback that was provided during these sessions and made sure that necessary amends were done in the next round of fidelity.
Now that I had the high-fidelity designs, I wanted to test them with users. I created usability script and designed a test on Maze. With this test I wanted to primarily test discoverability and understanding of certain information. With the help of research ops we recruited over thousand testers across 3 markets i.e. Kenya, Portugal and Spain. All of the testers were active Glovo couriers.
Screenshot of the Maze results report.
Further processing the results from the Maze test and defining necessary iterations based on the results.
The results came overall positive, majority of the users had no problem with completing the missions. Nevertheless, there were several small changes that I wanted to implement based on the test results e.g. some copy iterations. Receiving positive results from user testing gave me confidence to proceed to the handoff phase.
Generally, for the handoff, the goal I pursue is to document designs in such way that if I was to go on holiday developers would still be able to implement the solution solely relying on the documentation. As part of the handoff, I have mapped out screen-flow of the solution for both MVP and V2 versions.
Full experience map for MVP and V2 versions (including all the surrounding communication.
I have documented layout of designed screens. Documentation included guidelines on spacing as well as the components referenced in the designs.
Example of the screen prepared for the handoff.
Also, as part of the handoff process, I have documented all the new components. Having clear documentation will ensure minimum discrepancy between the design and the coded version.
Screenshot of the newly documented component
The new referral feature was successfully rolled out in 18 markets and had the following impact on the defined KPIs:
The hypothesis was proven to be successful and the feature had surpassed the expected impact.