Cash Flow Planner

ROLE: Product / Design Strategist & Design Lead

Concept, design and launch of the first mobile native, cash flow planner built upon QuickBooks data using AI & ML.

Problem

Cash flow is the beating heart of every small business. Its rhythm & regulation is critical to whether a business survives and thrives. No good tool exists to help small business’ forecast or regulate their cash flow.

We set out to fix this.

Approach

Partnering with research and participating in qualitative sessions at outset, I quickly moved to align executive sponsors, critical stakeholders and establish product vision. This presentation helped communicate project intent both within the project team and messaging out to wider org. The prioritized principles helped anchor design and engineering decisions throughout the course of project.

  1. Clarity
  2. Simplicity
  3. Play
  4. Transparency
  5. Flexibility

Set up for success

Aligning deep customer insight, conceptual underpinnings, clear mission and showing how these met the teams quarterly strategic objectives (OKRs) along with a very narrow and clearly articulated problem statement laid a powerful foundation for the team and project.

Where previous attempts had been technical or data science focused, our customer-centric process of rapid design experimentation at increasing levels of technical fidelity and early / often validating with customers, enabled us to quickly land on a strong design solution which customers got very excited by, and stakeholders had great confidence in.

Design foundations

Cash flow had been explored several times before both from a widget standpoint and also in context of QuickBooks Capital so we had a lot of internal knowledge, had talked to many customers and collected many examples of how people currently solved for this — lots of spreadsheets!.

While it was harder to control when money would come into a business (and we had QBCapital to solve for this), customers talked of having some ability to choose when money went out from a business, this manual load balancing was their real-world hack for keeping cash positivity. This insight inspired me to push more for dynamism and interactivity vs. precision in data or visualization within the experience.

I identified the core components for the experience – time, money in and money out and defined a core objective:

Design process

From simple paper sketches to increasing technical sophistication, I led or directly delivered:

  • Series of sketchs & simple clickthrus in invision quick illustrative UI pictures to validate information hierarchy and key components
  • React prototype with real data in partnership with design engineering. Our first real functional prototype, which we tested with customers using their real business data from QuickBooks.
  • Framer interaction explorations┬áin partnership with junior designer / prototypers
  • iOS native testflight application in partnership with product & engineering
  • Product management guidance to support requirements / eng build. Messaging out and up regarding progress & decision making. Consultation with wider teams including legal, partnerships & compliance. Guidance on data tracking and analytics ensuring we maximize quant / usage insight
  • Ongoing and often user testing at all stages, activating wider team participation in user testing

Early explorations tried to tie visual timeline data directly to data elements, this helped with understanding but clearly would not work well within the constraints of a mobile solution – which we had declared we were going after.

We concurrently began exploring a series of mobile point prototypes in Framer, experimenting with gestures and interactions. Some affordances were not clear and some interactions were not conventional but these were extremely useful for thinking through chart / data interactions, exploring concepts and partnering with our engineering group. We ultimately ended up relying more on established mobile interaction patterns but with more time we could have pushed these innovative interactions further.

Subsequent design iterations focussed on refining the graph, introducing bar toggle, simplifying data editing & input, calendar manipulation and time scaling 60/90 days. At this point our team was additionally tasked with taking over all cashflow work and defining for the whole of QuickBooks. This required subsuming some other teams and adjustments to incorporate their timelines and requirements.

First live in-product launch took place within 10 weeks.

 

Building trust & consensus

It is important to involve people in these type of projects as much as possible, as early as possible and being able to articulate how this project built upon and diverged from earlier efforts was very important for certain audiences. Messaging out to different forums and creating opportunities for informal engagement also helped bring wider teams along.

Maintaining momentum

As this project started very small, we were able to iterate quickly and informally, while tracking some key requirements which we knew would be more challenging to implement. These proved durable and once were were able to resource a dedicated PM, formed the backbone of traditional product scoping. I ensured that both immediate tactical design needs were met, assets to be consumed etc while continuing to push ahead and validate future feature development through customer research and inspirational design exploration.

Learnings

We used amplitude to track usage analytics, measuring appeal via access, repeat access, data manipulation. We bolstered this with ongoing qual testing, tracking both observed and self-reported feedback. As had been so successful with QuickBooks Capital, we also provided customers a mechanism to participate in a paid survey within the app experience, this gave us an additional feedback source and helped prioritize feature enhancements and roadmapping.

Next steps for the project include extensibility – ability to easily scrub backwards & forward in time, time scaling & data granularity as well as ongoing visual design and graph refinements.

Impact

Project has been a huge success, implemented both as stand alone widget and is central feature to core product innovation and product evolution. It’s being actively promoted as a benefit and RTB for QuickBooks and is available in-product to early adopters and new subscribers, indicating significant and durable business impact.

Currently available to all new subscribers. About to roll out to existing mobile & web customers.

Publicly announced at QuickBooks Connect, November 2019. Press release, Articles on zdnet, accountingweb.

View
Kirsten Disse. All Rights Reserved.