In 2018 I started working at OnlinePajak where I worked on a redesign of the tax invoice creation experience for eFaktur OnlinePajak. The primary focus of the redesign was to increase the user acquisition and help tax-payer simplify their tax management (preparation and report). Some of the ways we supported this in the redesigned the list and creation page to improving the accessibility for their data (sales, purchases, and tax report).


My role in this project was leading the early stage discovering, designing and testing. Today the new designs are used in OnlinePajak invoice.

Role: Product Designer

Duration: January to March 2018

Company: OnlinePajak

Analyse & running usability studies


I began my work by analyzing current designs and by conducting usability studies to test the current design. I focused on working with the product owner to understand the research objectives and working with the customer support to crafting the right research questions to ensure we could get valuable feedback.


analyze process

Studies insights


While I was analysing the findings from my study, I uncovered something interesting. In the study there was some screen in particular that elicited an incredibly negative reactions from participants.

review-old design

Will solving this problem help to increase the acquisition?


There are a lot of different issues in eFaktur (Invoice OnlinePajak) to be solved, so the first thing I did was make sure that this was a problem that was worth solving. To help communicate the problem to Product Owner, CEO and Developers and get buy-in. I used a combination of user feedback and internal metrics to show the depth and breath of the problem.

Who uses this?


Before start designing I wanted to understand what kind of user that uses this application, and what are their goals and motivations and context? It’s easy to think that eFaktur is just “taxable employers, businessman”. But I learned from research that when designing at this scale it can be helpful to break up users by thier use-case than their demographic.


What is the people problem?


I discussed with my engineers and PO and together we agreed on exacly what user problems we were trying to solve.

  1. Build user trust by matching eFaktur OnlinePajak to the real tax process
  2. Help their tax process by simplify the preparation and reporting

Design, test & iterate


After a lot of iteration on the sketching, we moved to more high fidelity mockups to work on the design details. At this point we’d made a lot of assumptions so I conducted usability studies where we got feedback on the new designs from real users.

Group 2
Group 2 Copy

Usability finding and recommendations


So I listed another usability finding that I got during the iteration and test. And I make three different severity rating on them.

Group 4
Group 5
Group 3

Final design that we shipped


After some enhancement from several usability we decided to have the final design.

Invoice homepage
Invoice homepage
popup csv
Import popup
popup csv
Import done and user option
Tax Invoice Creation
Tax Invoice Creation - Invoice number clicked
Tax Invoice Creation - Seller dropdown clicked
Tax Invoice detail page

Launch and evaluate


Finally after a long process to produce the new design and we completely confidence that our new design can increase the user acquisition we decided to launch the new design to the users and we start to analyse it for one month.

ua 2
ua 2

Reflection and next steps


Throughout the design process I worked closely with my engineers. This meant that once we started to build the product, it was really easy to communicate as we were already on the same page and we’d already worked out most of the technical constraints. I tried to be really proactive in this relationship throughout the development process, often this just meant grabbing coffee or hanging out at a desk together!


Even though eFaktur interface already redesigned and the user likes it, but it only impact the completion rate when user start to create invoice to approve their invoice, and it not increase user acquisition. So my next step is to analyse the step on the funnel to see which step need to be improved, if we see from the result that on the funnel our biggest drop is in the government certificate registration (enofa) so I might would try to redesign the user workflow for this application.