How Data Helped Us Take a Design Decision
“We live in a day and age where data is king” — I recently realized this fact when data helped us take a big design decision for a project.
A gist about the project
Townscript is a Do-It-Yourself event-ticketing and registration management platform. We have an Identity Verification procedure for event organizers which is to prevent frauds and keep the platform safe for everyone, also known as the KYC (Know-your-customer) process and Townscript makes payouts to only those event organizers who complete their KYC. So this project was about a design revamp for the whole KYC process itself, and we were planning to conduct a usability study for the same. Shweta Chaurasia, Assistant Product Manager at Townscript, and I were working on this project.
How did we identify the problem -
We saw a lot of drop-offs in the number of people completing the KYC process. For instance, let’s say 100 people (event organizers) signed up on Townscript; only 15 people were able to complete the KYC process. That’s an 85% drop off, which is a concerning figure.
Expected outcomes -
• A significant increase in the % of people completing their KYC process.
• Simple and easy KYC so that our users don’t dropoff from the platform because they couldn’t complete their KYC process.
Our hypothesis
At this point. Our intuition says it should be the KYC process itself that could be the issue here because we had received some complaints from the Customer Success team. But the thing about intuition is that it’s trained from past experiences, it’s not 100% correct always.
We laid the plans for the next month to test out a high fidelity prototype that is to be designed for the improved KYC process.
User Research
A crucial step before designing any feature is to gather insights from the users that are going to use the functionality and the key stakeholders of the project. We decided to talk to at least 10 of our events organizers who have gone through the KYC process.
First up, we decided to talk to event organizers who recently got KYC Verified. To give you some context, when event organizers complete filling their KYC details, after submitting their application goes for a verification process. I’m referring to the ones who’s application got approved, and now their KYC status is verified. Here are the findings -
- 5/5 Users felt the KYC process was easy and very straightforward.
- 3/5 Users mentioned that they had difficulty in finding where to fill the KYC details.
Surprising results! Although we spoke to only five users so far, our hypothesis about the KYC process itself being the problem doesn’t hold yet. People completed it without any confusions or doubts. Anyway, we moved on to the next set of users that we had decided to talk to — users who got stuck or had dropoffs in the KYC process. We spoke to 10 such users and got some interesting insights.
Here are the findings -
- 8/10 Users couldn’t complete filling their details because they thought a specific form field is mandatory and they didn’t have that information with them so left it incomplete.
- 6/10 Users mentioned that they had difficulty in finding where to fill the KYC details.
Our initial hypothesis isn’t ultimately proven wrong, there are issues with the KYC process too, but the bigger issue if the findability of the KYC process in the event organizer dashboard. We turned to data to validate this intuition.
The validation
Product metrics are critical in situations like these where you’re looking for validation for your intuitions. We decided to observe the metrics from the month from which we took the 85% drop-offs figure earlier.
Data doesn’t make sense unless you visualize it accurately. At first, glance, looking at the product metrics, we were not able to connect the dots. I decided to visualize it in a flow diagram and then it was clear!
Key Insights from Data —
- There were dropoffs, but not much in the KYC process but people were not able to find the feature, which is an information architecture problem but a bigger one to tackle before going into revamping the KYC process.
- More than 90% of event organizers who could find the KYC process, were successful in completing it.
Then we realized, instead of putting efforts into improving the KYC process, we should first work on the findability of that feature. All in all, significant learning is to use data as a guiding light which helps you make decisions and helps you validate or disprove your hypothesis.
