How do you build and develop a custom growth-driven design website?

Strategy comes first because the launch won't include everything. When the wishlist is longer than the team can build well in one pass, something has to decide what goes first. In growth-driven design that's research about your buyers, combined with a goal that has a number on it.

Lean Labs' growth-driven design program opens with a strategy stage of 10 to 14 days. The work in it is research: who the buyers are, what they come to the site to do, how they move toward a purchase, and which numbers will count as success. That's one provider's schedule for its own program, and yours depends on how much research you need to do.

What you want out of this stage is a ranked list. Each item is something the site could do, tied to a buyer need and to the goal. The top of the list becomes the core of the first build, and the rest waits in order. The first build also has to work as a whole site, so it carries what those top items depend on, such as navigation, forms and the pages they link to.

The ranking is a large part of what makes the site custom, because it comes from your buyers and your goals. The design and build choices that follow are made to serve it. There's more on this stage in how to create a growth-driven design strategy and in how jobs to be done and user research shape it.

The launchpad is the first version of the site. It has to improve on the site you have now, launch quickly, and be easy to change afterward. Lean Labs schedules 60 to 90 days for its launchpad stage and aims it at the few features expected to produce most of the impact. Read that as a prompt to rank hard and build the top of the list first.

Suppose a B2B software company finishes strategy with forty items on its list. The research says buyers arrive wanting to see pricing and to know whether the product fits their industry. The launchpad might then be a homepage, a pricing page, two industry pages and the demo request flow. The older content moves over with a light cleanup. The resource library redesign and the ROI calculator stay on the list for later sprints.

Two build decisions affect how the site develops afterward. The first is to assemble pages from shared sections where the platform supports it. WordPress's Site Editor documentation, for example, notes that a change to a header or footer template applies to every page that uses it, so one edit updates all of them. There's more on that choice in modular web design versus custom development.

The second is to set up measurement before launch. Analytics, form tracking and the connection to your CRM start recording when they're switched on. Tracking switched on after launch misses the earlier visits it was meant to record.

The platform matters for a related reason. A sprint depends on changing a page, so who can publish matters as much as what the platform can do, which is the question behind whether HubSpot is the best platform for growth-driven design.

Once the launchpad is live, development continues in a repeating cycle. In the method's description of continuous improvement, each sprint is planned around one focus metric, and the ideas most likely to move it go into a time-boxed build. The team then learns from the experiments and the data, and shares the lessons with marketing, sales and service.

Lean Labs' program shows how one team fills in that cycle. A sprint opens with a look at where buyers get stuck, drawing on page analytics, form results, CRM records and what sales is hearing. The fixes are then ranked by their likely effect on pipeline, by how confident the team is in them, and by the effort they take.

Not every change needs a split test. A broken form gets fixed. A test is for the case where two reasonable versions compete and you want to compare them on live traffic.

HubSpot's page A/B test, available on Content Hub Professional and Enterprise, shows two versions of a page at one URL to half the visitors each. HubSpot recommends changing one variable at a time, which makes a difference between the versions easier to interpret.

A test doesn't always finish inside a sprint. How long it takes depends on the page's traffic, on how far apart the two versions perform, and on how the test is set up. So a test can keep running into a later sprint while other work continues.

Analytics show where visitors leave a page and don't show why. Investigate the reason before you choose the change, with interviews, usability sessions or what the sales team is hearing. We go further into this in turning user insights into hypotheses and experiments.

Those timings are what one agency advertises for its own stages. The method calls for a time-boxed sprint, and thirty days is Lean Labs' choice. Choose a length that gives your team a workable rhythm for planning, building and reviewing. A test can run across more than one sprint, so the length doesn't have to fit every test. The launchpad's schedule moves with how many pages it covers.

The ongoing stage can also be a separate agreement from the build. Lean Labs' growth-driven design program, for example, runs on an agreement that renews each quarter. Its Design Blueprint is a different engagement, a four-week sprint that sets visual direction before a build starts.

The distinction matters when you compare quotes. A price for strategy and a launchpad covers the first two stages. Development continues in the third, so know before you sign whether it's included, who will run it, and for how long. What a growth-driven design contract and retainer should include covers those terms.

Through all three stages, the ranked list from strategy stays the working document. Each sprint's lessons move items up or down. Keep it where the whole team can see it, and note beside each item the evidence that moved it.