Open Government Products
Spinning up a payment service in 3 weeks to support $85mil worth of payments within 4 months of launch
My Role
My Contribution
About the team
When a government agency wants to collect payments, they will look for vendors to help them propse a solution, build a website, set up the payment form and related infrastructure. This might take months or even years. How might we empower government agencies to set up payment links? How might we make the payment experience a positive and quick experience for citizens?
We interviewed a few agencies and this is what we found -
I was the lead and sole designer. I helped out with user research and testing. I did all the design work (mocks, prototypes, branding, visual design, ui design) and also wroked on the frontend styling of the landing page, email templates for payment links and receipts.
1 Product Manager, 1 Product Designer (myself), 3 Software Engineers, 1 Product Support Specialist We also worked with public servants at the various government agencies for the operational details.
In June 2020, Singapore wanted open her borders to travellers and returning residents. The government needed a way to collect payments for COVID Swab Tests and compulsory quarantine accomodation. Multiple government agencies were involved as there were students, employees and other travellers seeking to enter Singapore.
Singapore want to open our borders and we intend to charge inbound travellers for their swab tests and hotel stay if they quarantine heres. Is your team are able to develop a system to help us collect the payments? We need it running in 2 weeks.
The team that I worked with is most familar with Stripe (amongst all the different payment gateways). We haved worked with it (and a few more providers) on previous projects. Hence, we are confident that if can leverage on Stripe's robust APIs and admin capabilities , we can put together something quickly.
We only have 2 weeks and this is the idea we came up with. The journey starts with a government officer will collate an excel file consisting of the user information and payment details that and send to the developers each day. And our system will create payment links and send them to the user emails so that they can make payment via the payment links. The user information can come from internal systems/ online forms etc.
We worked with multiple government agencies including the Ministry of Manpower, Ministry of Education, Immigration and Checkpoints Authority. And all of them have very different requirements. To be able to deliver something in two weeks, requires focus and we narrowed down to this core requirements in green checkmarks and deprioritise those labelled as red crosses.
Each payment, in this case, is in the ball park of thousands (due to the 14 day quarantine requirements). The team decided to focus on the payer experience and prioritise these screens.
This also means that we deprioritise things like a customised admin view, ability for officers to self-service, ability to make refunds, accept other payment types, payments in bulk, integrations with internal systems etc. As most of the refunds/finance tracking and audit can be handled in Stripe admin out of the box and others are less critical at this juncture (e.g. integration) compared to the base experience.
Collecting payments online require building trust. How does the payer know this link is from the government? How do we make sure that travellers, citizens and companies who sees the payment links, trust it to make payments of thousands of dollars?
We achieved this by making sure that
This project is only possible because -
Existing team rapport is great
A great team to work with is not to be taken for granted. We had been through thick and thin together and we understand each others' strength and weakness.
Rely a lot of past experiences
Agencies we worked with trust our recommendation because of the good will and great work delivered in the past. Hence, stakeholders push back was manageable. We also made a lot of product decision based on our past developing other 2 payment products, bus booking and parking.
Little to no time for research and testing
One week of development and one week of testing it repeatedly with people around us. Not enough testing with main personas.
Making a lot of assumptions
We assume the baseline flow and just implemented that. I missed some edge cases and had to do product iterations, fixes and ops recovery support on those. (e.g. Employer bulk pay for staff, company fire walls and email clients link shortening)
Now that we have launched a payment service... Can we automate the manual portions? Think bigger!
How might we use this as a foundation for a government payment system in the long run to solve the problem mentioned in the research previously?
I facilitated a 2 day design thinking workshop with the team to unpack our learnings and our long term vision. And this is what we wanted to achieve in the long run -
I came up with some illustrations for the product by modifying royalty free illustrations from Freepik and standardising them with brand colors and styles
I designed the Pay Logo with a letter "P" and a checkmark inside to create the sense of assurance. The P also looks like a paper note fold-over.
As of Oct 2020, 4 months of operations
It is one of the most time compressed project I have done and I learnt that we can do so much if we focus. I left the the organisation before I could finish this project, so I shall leave this in good hands of the team to continue. The version of the website you see in the link at the top of the page is built by the team that took over.