Coconut saving for tax
Overview
One of the biggest anxieties around self-employment is how much tax is owed and whether enough has been saved to pay the bill at the end of the tax year. Coconut already offered a fairly solid estimate of how much tax would be owed, but customers had varying degrees of habit around actually saving towards that figure.
My contribution
Product strategy User research Product design
The team
1 × product designer 3 × engineers
Year
2022
Process
Hypothesis
Creating an automated saving tool would allow users to ‘set and forget’, knowing that money is being set aside in the background to cover their tax bill and cementing Coconut as an essential tool for a self-employed person.
This automated tool would set money aside in a segregated account, allowing users to feel safe that they weren’t spending money that they would owe to HMRC at the end of the tax year.
Research
I sent out a survey to our customer base to better understand their pain points around saving for tax, and how they save currently. I made use of both our community forum to contact users directly, as well as an in-app message to a subset of our more highly engaged users to deliver the survey.
The results were completely unanimous in the need for such a tool to help automate the saving, but it highlighted the need for a couple of different modes that would be required – one which is completely automated, and another which automates the figure, but waits for user confirmation before moving any money.
UX
With the results from the research I set about mapping out the experience of the feature. I worked closely with our engineering team to understand the limitations of the platform currently and what would need to be developed in order to support such a feature.
One of our main limitations was supporting external bank accounts, which would have required some form of third party integration. To speed up development, we made the call to only support customers that held a Coconut current account, allowing us to develop on tech we were familiar with.
Engineering remained closely involved in helping to define the UX to ensure that it was buildable and all edge cases were accounted for. I used a combination of Whimsical for flows and wires, Marvel for prototyping and Maze for customer validation.
Outcome
Due to the potential scale of the feature, I worked closely with the PM to ensure that we started small and released in sprint-sized chunks. To help gather data, we created a beta testing program so we could still release without a full public launch. This allowed us to validate our ideas and help define the next release.
The UI was kept understated, with a simple progress bar when viewing savings in the context of the tax estimate. The feature had a subsequent screen for transaction activity and a few flows coming off that for deposits, withdrawals and settings for the automation side.
To help with marketing the feature we named the automation element ‘VaultAgent’ to bring an element of personality to the feature, as well as an easy reference for the user when an action was required of them.