replica rolex

How Redesigning Doubled Conversion Rates

March 21, 2017

By (Austin Knight)

At INBOUND, we launched a major iteration of Since then, we’ve been monitoring performance, running experiments, and collecting a substantial amount of performance data.

We can now confidently say that the results are in, and we have some big learnings to share with you. So let’s get to it.

Like most major projects at HubSpot, this one began with compelling data that pointed to a few big opportunities in our site. While we’re always iterating on, we tend to push one or two major iterations per year, resulting from the culmination of these incremental learnings. In this case, we were focused on:

User Experience was suffering from what I like to call Design Debt. This is what happens when you run a bunch of experiments and the site is performing well quantitatively, but the experience of using it just isn’t all that great.

Design patterns may be inconsistent, navigation flows may feel disjointed, and critical parts of the site may end up looking like they were an after-thought (that’s because they were). When this happens, you have to refactor your site and scale your learnings across the entire design, creating an updated and consistent experience.


By applying the learnings from our individual experiments across the site, we knew that we’d be able to increase conversion in a meaningful way and build a strong foundation for CRO efforts going forward. Experiment results shouldn’t live in isolation; they need to be distributed across teams and designs, and we were ready to do that in a big way with this project. But perhaps more importantly, our product was changing.

Multi-Product, Internationalization, and Accessibility

Prior to INBOUND 2016, we knew we would be launching the HubSpot Growth Stack, which includes free versions of all of our products. This meant our conversion flow would be fundamentally changing — not just for US customers, but for every HubSpot user around the globe. It was an incredible opportunity to reassess how we introduced our product, and brought users into it.

Ultimately, we would be rethinking our approach to altogether. An initial audit indicated that this would require a refactoring or redesign of around 70 critical pages on And it would all need to happen over the course of 3 months, in time for launch at INBOUND.

But we didn’t just begin with such an ambitious and resource-intensive project based on some data and a few hunches; we still needed to prove out the concept with a live MVP (minimally viable product), and ensure that it was the right thing to do.

Proof of Concept

Over the course of a few days, Matthew Barby and I silently launched an MVP with HubSpot Sales. We wanted to see what would happen if we took all of the data, hypotheses, and wild ideas that we had and put them into a single design without reservation.

I’m talking illustrations, big drop shadows with simulated depth of field, colors that broke our …read more

Source:: HubSpot Blog