What is the difference between growth-driven design and conversion rate optimization?

Growth-driven design has three stages: strategy, a first release called a launchpad, and continuous improvement. It also covers decisions about the site as a whole, such as which site to build, what to launch first and how to budget for the work after launch. We cover the method in what growth-driven design is.

CRO starts from an outcome. A team picks an action it wants more visitors to take and works out what stands in their way. That work is broader than testing button colors. CXL's guide to conversion research describes technical analysis, a heuristic review of key pages, analytics, recordings of visitor interactions, surveys, interviews and task testing. The research exists to produce hypotheses a team can test.

So both use research, design, development and experiments. A CRO project can end up changing a page's structure, or the route visitors take through the site, when the research points there. A growth-driven design team spends part of its time on small changes to forms and calls to action. What separates them is the thing each is organized around. Growth-driven design is organized around the site and its stages, and CRO is organized around a measured conversion outcome.

Suppose a payroll software company has a demo request form, and its analytics show that many visitors start the form and few submit it.

The team starts with research, and session recordings show visitors stopping at the phone number field. A reviewer notices that nothing on the page says what happens after the form is sent. CXL's guide treats observations like these as areas to investigate, and the recordings alone don't establish a cause. So the team asks sales, and the reps say buyers often open a call by asking whether they'll be phoned repeatedly. The team now has a hypothesis: visitors hold back because they expect sales calls they didn't ask for.

Suppose sales agrees to make email its first contact with every new demo request. The team can then make the phone field optional and add a line under the form saying a specialist will email to arrange a time. The line is only worth adding if it describes what sales will do.

With enough traffic, the team can measure the change with a split test. HubSpot's page A/B test, for example, sends half of a page's visitors to each version and is available on Content Hub Professional and Enterprise. Jakob Nielsen's 2005 article on A/B testing describes the limit of that result. The test shows which version had a higher completion rate among visitors who started the form. It doesn't show why, and an immediate action can leave out longer-term effects on the business.

That limit matters here in two ways. Suppose submissions rise. The hypothesis about unwanted calls is still what the reps reported hearing from buyers, and the rise doesn't prove it was the reason. A completed form is also only the first number. The team still needs the CRM to show how many of the new leads were qualified, and later how many became customers. An optional phone field can bring in more people who were never going to buy. Lead quantity, lead quality and revenue are separate measures, and a change can raise the first while lowering the second.

All of that is CRO. A specialist could do the same work on a five-year-old site that has never been through a growth-driven design stage.

In growth-driven design, the same form work is one cycle of the improvement stage. The method's cycle has four steps: plan, build, learn and transfer. The team plans by choosing a focus metric and ranking the work it expects to move that metric. Here the metric is the share of visitors who start the demo form and go on to submit it. The team builds the change and studies the result. Then it transfers what it learned. Sales raised the worry about calls in the first place, so what goes back to sales and marketing is the outcome: whether a larger share of visitors who started the form submitted it, and how the new leads compared in quality.

Lean Labs' growth-driven design program shows the two fitting together. The program runs improvement in 30-day sprints. It draws on page behavior, form results, CRM lifecycle data and sales feedback to find problems, and its criteria for ranking ideas include expected pipeline impact and effort. The work it lists includes messaging, forms, calls to action, personalization and lead routing. Where that work aims at a measured conversion, it is CRO, done inside a structure that also covers strategy and the launchpad.

The surrounding structure changes what the form fix competes with. The team already has a ranked wishlist from the strategy stage, so the fix has to earn its place against larger ideas such as a new pricing page. Once the fix is chosen, its test can run across more than one planning sprint. We go further into prioritizing the work in turning user insights into hypotheses and experiments and into measurement in what metrics to track in growth-driven design.

When the site as a whole no longer fits the business, the job is a rebuild, and CRO research can inform it. Growth-driven design is one way to order that rebuild, and it keeps a budget for improving the site afterward. We work through when a rebuild is the right call in whether a website redesign will fix low conversion rates. A company with a sound site and one underperforming path can start with CRO on that path alone. A pricing page that few visitors leave for the demo form is a path of that kind.