Karbon is a feature-rich global operating platform for accounting firms that needed better solutions to onboard new users.

The challenge

At the time, Karbon was a 200 person start-up that had historically relied on sales, implementation and customer success teams to drive growth. In 2023, the team shifted to a product-led approach where I was tasked with designing onboarding features to better engage, activate and convert customers in a free 14-day trial.

The Impact

While balancing my core feature work, I implemented a new home page and multi-step wizards which streamlined setup activities and reduced time-to-value for customers. At the time, we didn't yet have the tooling or infrastructure in place to correlate our experiments to retention metrics. That said, I am super proud that these experiments helped me influence our executive team to implement a more data-driven framework for measuring future growth experiments.

My Contribution

User Research
Product Management
UX & UI Design
Prototyping
User Testing
Measurement

An accounting firm manager who lacks adequate visibility over their team’s work has a hindered ability to ensure they will meet client expectations. Above all else, they're desperate for visibility and rightfully look to Karbon to address this.

I'd heard time and time again that customers choose Karbon for its solution to work management, only then to discover what our email integration feature can do for their collaboration and communication—our stickiest feature. But I didn't have a definitive answer to what the golden moment is in a free trial that convinces a firm that Karbon is right for them. Does that even exist? I observed and spoke to users in their 14 day free trial to better understand their motivations and expectations.

Digital interface of Karbon's contact import process

It became clear that Karbon's first-use experience poses challenges as it is a complex product that requires a lot of set up before a firm fully realises its value.

For example, to experience work visibility and reporting—a critical "aha!" moment—an operational leader would need to:

  • Build efficient workflows: Setup automated workflows in bulk, assign things like budgets, team roles, and deadlines, before inviting the wider team to experience workflow in Karbon.

  • Import client contact data: Client workflows are connected to hundreds if not thousands of client contacts. So detailed contact data needs to be imported in bulk before workflows can even be created.

  • Make individual contributors love and use Karbon: Only once client work and contacts are set up can individual contributors in the team use Karbon and find value in completing daily tasks and keeping their assigned projects up-to-date—this is required in order for the manager to benefit from that visibility and know work is being delivered on time.

Historically, these things all required hands-on assistance from Karbon’s implementation teams. We needed a better way.

How might we simplify the setup process so managers can experience visibility and reporting, with their team's real work, from the moment they land in the app?

The key to successfully converting a new trial user into a paid customer would be for the product to effectively guide them toward realising the value of Karbon during their initial session. While simplifying some non-negotiable set up activities. My goals were to:

  • Clarify that Karbon enhances workflow and team visibility.

  • Visualise the team's real projects on a Kanban board.

  • Invite the first team (3–5 members).

  • Guide them to reporting features to drill down on team progress.

User interface modal Karbon's user interface with a calendar to book a call

I chose to isolate the set up steps in a guided wizard so I could craft the right story and keep users hyper-focussed on their jobs to be done.

To achieve a populated dashboard of client projects, users need to jump through a series of steps that happen in different areas across the app. So I chose to isolate these steps in a guided wizard to keep users focussed on only the actions they need to take, and let the tech to do the heavy lifting in the background.

This early on in their buying journey, I was very conscious of the user’s mental load and eagerness to get into the app. My focus was on crafting and iterating the right copy and story that would demonstrate value and boost their psych to helping them feel a sense of progress and remain compelled to continue. Every little detail had to count.

Digital interface of Karbon's contact import process

After a few weeks, 87% of users had made it through the new onboarding steps—69% returned for a second session.

At the time we didn't have a reliable metric or mechanism for AB testing to compare conversion impact against. Our team was okay with this trade-off given the majority of users (and likely higher quality leads) successfully completed the set up steps and we had a new baseline to improve upon.

Digital interface of Karbon's contact import process

Regarding iteration, there's so much I would have loved to explore here! Like the impact more personalisation might have on conversion for our different personas. But given the feature-rich nature of Karbon, our small team size and our general lack of onboarding solutions, we had bigger fish to fry...

Auditing existing journeys to identify more quick-win opportunities.

Up until this point, we had the signup and some basic getting started implemented. I needed to level this up. I decided mapping what the user journey looks like for our target persona (Managers/Admin in a 1-3 seat firm) is fundamental and might uncover points of friction.

I ran a few workshops with customer implementation folks to understand the minimal steps a manager in a 1-3 seat firm needed to take to get setup and learn the fundamentals. Then narrowed in on non-negotiable steps such as importing client contact records and creating client projects from pre-loaded workflow templates.

Digital interface of Karbon's contact import process

One outcome of this was that it became clear we needed a far more intuitive, self-service flow to help users bulk import their client contact data.

We knew from past research that users who engage Karbon’s support team to manually import their client contact list convert at greater than 70%, and this is one of the stickiest activation activities a new user can take.

In a 3 week sprint, I worked with one product manager and two engineers to build a wizard aimed at delivering the shortest route for a new user to achieve a populated client list on their own. By reducing the required contact data fields down to the minimum, we were able to cut the import time of thousands of rows of contact data from 48 hours to mere seconds.

User interface modal Karbon's user interface with a calendar to book a call

Implementing scalable UI patterns to educate and onboard users at every stage of their journey.

Without a dedicated Growth team or resourcing, I needed to work with patterns that would allow me to be nimble and try things quickly without breaking the bank. I pitched some new patterns to my product and design team to implement in our design system, as a starting point for more experimentation.

Digital interface of Karbon's contact import process

Where to next? To truly engage and retain users, users need to develop rewarding habits in the product that align with Karbon's core value propositions.

At the time, Karbon lacked the necessary infrastructure to measure product-led growth. We knew a firm sticks at the 90 day mark but we didn't have a reliable indicator of retention (beyond daily or monthly active use) to focus our efforts. To truly engage and retain users, users need to develop rewarding habits in the product that align with real value. We hypothesised these to be:

  • Weekly inspecting work progress (Managers)

  • Daily completing tasks (Individual Contributors)

  • Daily team communication (Everyone)

Digital interface of Karbon's contact import process

I worked with our data team to understand actions users take that lead to a higher propensity for conversion. Then, combined these insights to develop a matrix of one-time achievements that would lead users to experience value and form habits that correlate to longer-term retention. These became my north star metrics for forming new hypotheses and experiments.

User interface modal Karbon's user interface with a calendar to book a call

Challenges faced, lessons learned, and next steps...

Feature-rich operating systems may remain Sales-led, so PLG initiatives need to work hand in hand with Marketing , Sales and Customer Support teams.

Product-led growth is fundamentally cross-functional. Marketing sets the stage for onboarding, which rolls into free trials, purchasing experiences, customer support resources and beyond. So it's important to think in journeys and asses how growth initiatives work hand in hand with wider company strategies or initiatives. Such as ways for users to book time for a call, access help centre content, reach out to support or provide product feedback.

All designers should factor PLG principles into their core feature work and product strategies. Especially in start-ups or scale-ups.

Working on Growth initiatives has fundamentally changed the way I think about building 'core' features or products. I now have a handful of new principles and questions in my design toolkit... How would we onboard new users to this? How might we help users quickly experience value ? How can we build network effects by enabling users to easily share experiences or invite others? How can we create features and interactions that make users foster a positive emotional connection with the product?

Timely access to user behaviour data is critical for effective experimentation. But qualitative evidence is still valuable and calculated risks can pay off.

Embracing a lean experimental approach is critical to learn, iterate and improve. So data requirements must be properly considered in the planning phase for effective experimentation. That said, this isn't always realistic in start-up environments with immature infrastructure or tooling, but it shouldn't get in the way of trying new things. Great outcomes can still be born out of leaning on best-practices and qualitative research.