It seems like every design agency has their own process, framework, methodology: whatever you want to call it, it’s an overall plan. But it’s weird articulating these, you know? You have a plan, of course, but you always, always deviate from the plan. And then you have to deal with diminished expectations when the plan doesn’t go fully like you or the client wanted.
With all that in mind, here is the plan that Draft typically follows with its clients. We’re calling it the Draft Method, because that sounds nice. It’ll deviate for you, and that’s fine: we’ll come up with something slightly different that works best for us.
Here’s what we do:
1. Initial Optimization
We’ve never encountered a business that came in the door looking flawless – after all, they probably hired a consultancy for a reason. Here are some things we typically do before we touch anything else:
- Configuring funnels & goals in Google Analytics.
- Cleaning up messy data reporting.
- Running important quantitative research, such as heat & scroll maps.
- Coming to a consensus on key metrics.
- Assessing browser & platform compatibility.
- Fixing any grievously buggy issues in any parts of your website that facilitate the taking of money.
By the time we’re done with this phase, we’re probably already improving our business’s conversion rate. In a few cases, just this phase pays for Draft’s work into perpetuity.
Once we finish cleanup, we kick off the most essential part of growing your business: research.
Research can take many forms, from qualitative (customer interviews, free-text survey responses, usability tests) to quantitative (analytics, heat & scroll maps, click maps, survey scores). Both methods are essential for understanding customers’ motivations and existing behavior – so we can run tests that are more likely to win.
Research is a continuous process, and it’ll happen as we run tests as well.
Research is what gives us ground to stand on – and synthesis is how we turn our research into revenue-generating design decisions. During synthesis, we look at what we’ve gathered and plan as many changes to your business’s funnel as possible. That involves both one-off changes (this is broken, fix it) and test ideas (changing this might be a tradeoff, and we need to measure its economic impact).
These are pretty rigid categories – you’re either testing an idea or launching it straight away – and no, you can’t just pile test ideas into the one-off bucket. You aren’t hiring Draft to be that careless: you’re hiring Draft to unlearn the “MOVE FAST & BREAK THINGS” poster that you saw in Facebook’s headquarters (and which they unlearned years ago).
Once we get our test ideas together, we prioritize them based on potential impact and feasibility – and then we move to the next step.
A/B testing is a tool that lets you gauge the economic impact of a single design decision. Draft uses it often. At this stage, we build a testable design decision, configure some goals, launch it, and wait.
Once our tests are ready to call, we write up a report analyzing our findings, summarizing any new research that came out of the test, and providing next steps. If a variant won, we celebrate and launch it to all customers.
Repeating Steps 2 Through 5
While a test runs, we’re always researching for the next one. Ideally, we’ll be able to kick off the next test right when one ends; the goal is to maximize testing time, and always keep researching. You’re never done.
We are not a “growth hacking” strategy. We provide a comprehensive, no-nonsense methodology for improving conversion rates. That takes time, and it requires a culture and mindset shift.
Process matters more than tactics – and the ability to be flexible matters the most, because any process will always be set on fire by unforeseen circumstances within a week after kickoff.
Why do research-driven A/B tests improve success rates by 400%?
Research-driven A/B testing is the most effective, proven way to capture lost revenue for your business – and other testing approaches are fundamentally wrong. This free 6-lesson course teaches you what research-driven A/B testing is, how it differs from other approaches, and why it will help your business make more money – without snake oil or lame shortcuts.