Hack Growth with Data-Informed Design

Zhaochang He
Dec 30, 2019 · 7 min read

In the B2C world, we hear a lot about the buzz word “Growth”. While growth is measured in various ways, many of these companies have growth teams dedicated to driving organic growth by focusing on product strategies. While in B2B companies, sometimes growth is all about how many big sales deals are closed. Product-led growth is overlooked by sales-led growth. As a result, the impact of newly launched R&D projects on business growth is unclear. And there is no clear visibility of what product design and strategy have contributed to the business. At VMware, we aspire to drive more product-led growth with great user experience, utilizing the data-informed design methodology.

Data is a powerful tool the designers don’t always fully utilize. At Wavefront by VMware, a SaaS platform that helps site reliability engineers(SRE) and devOps to monitor their infrastructure and application, we used data to help us increase the free trial activation rate by 30%, and increase the dashboard creation rate from 5% to 25%. This case study will demonstrate how to utilize data-informed design to drive the organic growth of the enterprise business, from tracking the user data, testing out hypotheses through experiments and see the results, and to ultimately accomplishing the goals like expanding the user base and increasing the conversion rate.

Product-led Growth v.s. Sales-led Growth

So, what is Product-led growth? It relies on product usage and customer experience to acquire new users, retain its existing ones, and expand its user base. It doesn’t involve much sales pitch and the process is more data-driven. So, how do we enforce a design-informed design process to drive product-led growth in enterprise companies?

The first challenge is to understand what product-led growth means to us.

We used a framework to brainstorm product metrics to help get clarity about what growth means: how is growth reflected as tangible goals, what user metrics represent these goals, and what are the supporting metrics for the representing metrics.

Brainstorming product metrics with a framework

For example:

  • What is the business goal: Acquire new customers to generate more revenue.
  • What’s the user metrics that reflect the goal: Improve the conversion rate from new users to paying customers.
  • What’s the supporting Metrics: Active user engagement, higher product usage such as query usage, dashboard usage, alert usage.

After we defined the metrics we care about, we start to track it with a tool called Pendo. Since Pendo doesn’t require code instrumentation to track the user metrics, it’s easier for product managers and UX designers to set it up by ourselves.

Understand the product holistically before dive into individual data points

To drive the product-led growth, we need to have a holistic end-to-end story of our product so we don’t go sidetrack into single data points. We are informed by the data instead of purely driven by the data.

Wavefront by VMware is a SaaS product, it has a 30-day free trial for SRE and Developers to try and explore how our products help them monitoring their applications. Here is the end-to-end process of converting from trial users to paying customers:

  • Aware: Users will first become aware of Wavefront’s solution.
  • Sign up: Users decided to sign up for a free trial to explore the product.
  • Try: Users became more interested in our product.
  • POC: Users will talk to our salespeople for a proof-of-concept.
  • Buy: If they decided to purchase they will be entering in a contract.

In the first three steps “Aware, Sign up, Try”, the growth is more likely to be driven by the product itself, there is not much sales pitch or hand-holding. We want more organic growth in this process.

We had a heuristic evaluation of the user going through this process and found many opportunities that we can improve. We start from the low hanging fruits.

Aware — Signup — Try — Proof of Concept — Buy

Do experiments and track the results and continue to learn

We have a lot of hypotheses, and test out them in a few experiments and see from a data perspective if it validates.

Experiment 1: Improve sign-up form and remove email activation

The first low hanging fruit we found is after users sign up they have to go through an email activation, this out of context action and friction caused a huge drop-out rate. And the old free trial sign up form has usability and accessibility problems.

Old sign up form

As a collaborative effort with our marketing department, we decided to remove this email activation step and improved the sign-up form design by a better-designed form with compelling product benefits next to it. The sign-up activation rate increased by 20%.

New sign up form

Experiment 2: Improve password setup experience

Then, we discovered that when users are trying to set up a password, the system throws a huge error message that it’s not clear to the users what they should do.

Before: Huge Error Message

To help users easily set up a password, we made the password requirement clear upfront. We display the password requirements on top of the password field so users are aware beforehand, for example, “minimum 8 characters”, “at least 1 upper case”, “1 special character”, etc. And if the user entered a password that didn’t meet the requirements, we highlight which requirements are not met. We also allow users to toggle to see the password they entered. After launching this design, the sign-up activation rate increased by 10%.

After: Make password clear upfront

Because the activation rate increased, our user bases increased, we expanded the top of our sales funnel. And we are able to demonstrate that good user experience can actually drive business growth, with quantitative data to prove it.

Another benefit of product-led growth is continuous iterations with data. The more we test, the more we learn, and we continue to build upon it.

Experiment 3: Improve the onboarding flow with clear call-to-action

Now users entered the free trial, they will be dropped into an onboarding flow. Again from the heuristic evaluation and Fullstory records, there is too much text on each page of the onboarding flow that nobody reads. Also, the onboarding just serves as an introduction, it’s not very actionable, and users don’t know what to do in the onboarding flow. And another problem is that since the flow has 7 steps, we see many users dropped out after each step.

Before: 7 steps onboarding, a lot of text

So, we made a quick change. We reduce 7 steps onboarding flow into 5, and get rid of all the unnecessary text, and replace it with images that are easy to understand. More users are able to go through the onboarding flow.

After: reduce to 5 steps with less text

Another hypothesis we have is that active and engaged users will more likely to turn into paying customers. The more actions that users take within the free trial, the more features they used, they are more engaged. This is what we did: we build a better connection between the onboarding flow with the product itself. We put actionable buttons on the onboarding flow, for example, “Create a dashboard” will take users directly into the dashboard editor, there are about 15% of users clicking on that button. Users are more engaged by taking more actions in the product.

It’s worth mentioning that each experiment above only costs a few days to design and less than 2 weeks of engineering effort to build. While the investment is small, the return is significant.

Experiment 4: Improve Dashboard Editing Experience

Since the new dashboard and chart editor is redesigned, it has a much better user experience. With a shortened and simplified workflow, the users are able to create a dashboard with only a few clicks. Since we removed the technical barriers, the new users don’t need to learn the query language to create a chart or dashboard. The new users are able to visualize their raw data in seconds! After the launch, we saw the dashboard creation rate increased from 5% to 25%. This is a huge increase.

Only take a few clicks to create a dashboard, no need to write the query anymore.

The benefits of data-informed design and product-led growth

As designers, we wanted to explain what business value we bring to the table. Now utilizing the data-informed design to drive the product-led growth, we are able to explain what the impact design has made for the business with evidence.

We’re hiring!

The VMware Design team is looking for talented designers to help us continue transforming enterprise design. Check out our open positions!

Special thanks to Lior Matkovitch and Frank Hattler for partnering in driving product-led growth, and Grace Noh and Bonnie Zhang for reviewing the article.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade