What is a launchpad website, and how do you build one with growth-driven design?

The method's launchpad page sets the aim as a site that beats your current one on looks and performance without being the final product. It is the base for collecting data from real users and improving from there. The page also says speed shouldn't come at the cost of value to the user.

So the pages that ship are complete. What gets held back is work that ranks lower: page types fewer buyers need, custom components, ideas with less evidence behind them. Deferring that work keeps the first release small enough to finish, and live traffic then gives the team one more source of evidence for ranking it.

The launchpad is the second of the three stages of growth-driven design. Strategy comes before it and continuous improvement after it.

The scope comes out of the strategy stage. The method's strategy page describes that stage as understanding the audience, setting objectives and gathering a wishlist of ideas, which then becomes a plan for what to build. We cover that work in how to create a growth-driven design strategy.

The launchpad is where the wishlist gets cut. A useful test for each candidate page is whether a buyer needs it to reach a decision. Lean Labs' article on the three stages puts its launchpads at three to eight pages. For most B2B companies it names a homepage, two to four service pages, and one page built for conversion, such as a pricing overview or a booking page. Team bios, blog archives, resource centers and its most elaborate design components wait for the improvement cycles.

That range is one agency's scope. The number that fits your site follows from how many steps your buyer takes and how many services you sell.

Suppose a payroll software company has a 60-page site. Its research suggests that buyers compare plans, check which accounting tools the product connects to, and then ask for a demo. Sorting the site against that path gives a first release like the one below.

Pages that aren't rebuilt are part of the release too. The method's page says to give the rest of the site a "fresh coat of paint" while moving it into the new one. In the example, the blog and help articles arrive with the new header, footer and type, and a customer who logs in from the old site finds the same link on the new one.

The deferred items stay on the list. The calculator becomes a stronger candidate if the live data shows buyers stalling on the question it would answer, and a weaker one if they don't.

The method's page gives the build an order. Page plans come first, with an outline and flow for each key page. Copy is then written into those outlines to make a prototype, so design starts with real content in place. Design, development, testing and launch follow, and the remaining pages are restyled and moved across.

Each step depends on a decision from the one before it, which is why approvals shape the schedule. The method's launch process page recommends giving each piece of work a fixed time box and making one stakeholder accountable for each page. That person can collect opinions from colleagues, and the team still gets one answer per review.

Lean Labs' article describes the build as three sprints of about four weeks each: messaging, then design, then development. Published targets for the whole build vary, and we compare them in how long it takes to launch a growth-driven design site.

A launchpad exists so the team can learn from it, and learning needs something to compare against. Nielsen Norman Group's guide to benchmarking describes measuring an experience against a meaningful reference, and an earlier version of the same product is one of the references it names. For a launchpad, that earlier version is the site you're about to replace.

So take the baseline before the switch, on the pages you plan to rebuild, while they are still live. Bounce rate, exit rate and conversion rate are the measures Lean Labs' article watches on a launchpad. The method's page adds heatmaps and session recording from launch day.

A baseline tells you whether the numbers moved. It doesn't isolate the reason, because a launch changes copy, design, page structure and often URLs together, and traffic can shift over the same weeks. Where a decision rests on one change, a controlled comparison during the improvement stage is one way to estimate its effect. We list the measures worth keeping in what metrics to track in growth-driven design.

The deferred list becomes the starting backlog for the next stage, alongside whatever the live pages show needs fixing. The continuous improvement stage covers how that work is chosen.