A Proven UX Research Process to Redesign Your Website

By Zach Watson

local maxima chart

Website redesigns are a huge risk. You can throw away years of incremental gains in UX and site performance—unless you have a battle-tested process.

But what does that process look like? And how do you know if your team—or the one you’ve hired—is focusing on the most important things? Many areas yield minor improvements, but to make a real difference, and protect past progress, you need a repeatable, evidence-based approach.

I’m going to walk you step-by-step through our company’s UX research process for site redesigns—the role of each aspect, how long each should take, and what, generally, each entails.

That knowledge will help you:

  • Create a foundation for an internal research process.
  • Improve the vendor evaluation process.
  • Hold your current vendor accountable.

But first, a note on “radical redesigns”

As noted at the outset, full-scale site redesigns have the potential to backfire. A high-level stakeholder can overrule research-based decisions, designers can clash with optimizers, and the bevy of simultaneous changes can make it impossible to identify which changes are to blame if a redesign performs worse than its predecessor.

A “radical redesign” contrasts with a process of ongoing experimentation. That incremental approach makes it easier to isolate the impact of individual changes and avoid the risk of a total redesign. It also avoids a years-long fallow period between redesigns when competitors may continue to move ahead.

That said, there are three strong arguments for a complete redesign:

  1. Major company changes. A significant shift in your brand or product may justify a new look and feel for your site.
  2. Hitting the local maxima. If experimentation begins to show diminishing returns, a radical redesign can raise the ceiling for site performance.
  3. Low traffic. Sites with limited traffic may be unable to test changes in a reasonable amount of time.

Hitting the local maxima can help justify a “radical redesign.” (Image source)

Other issues may also prompt a large-scale redesign, like a site that relies on old technology (e.g. Flash) or one whose design is clearly dated. (The latter scenario is common among startups whose rapid growth makes an initial “homemade” site appear amateurish.)

If a full-scale redesign is the right choice, the process starts with a comprehensive research plan that includes both generative and evaluative research.

Generative vs. evaluative research

While there are several categories of user research, the most important breakdown is between generative and evaluative research.

Generative research helps you understand your user’s world, learn more about their challenges, and generate ideas for your new website design. These techniques—such as stakeholder and user interviews, which I cover below—are mostly qualitative because the best audience insights surface during conversation and observation.

When you’re building your research plan, put generative techniques at the beginning. That way, the research will define the direction of your design.

Evaluative research validates how well your redesign solves the problems that inspired the redesign. Evaluative methods are quantitative—numbers measure the effectiveness the redesign. Examples of evaluative methods:

Don’t rush through the generative techniques. It’s natural to want to move quickly, get something designed, and test the …read more

Read more here:: B2CMarketingInsider

Leave a Reply

Your email address will not be published. Required fields are marked *

CommentLuv badge