Website Redesign: Tips & Considerations for Developing a Research Plan
Website Redesigns are a fact of every UXers life.
At UserZoom we work with hundreds of clients. Almost each one has at least one website redesign project they are working on. In the last two decades I have been a part of, even responsible for, several large redesigns and successful global roll outs.
In my experience with website redesigns I have come to realize the following, and wanted to share my tips with fellow practitioners.
Redesigning your website? Follow these tips to develop a successful research plan…
What do all redesigns have in common?
All redesign projects have two things: a goal and an executive champion. Generally speaking, redesign goals revolve around these categories:
- Making a site responsive (for mobile)
- Solving a particular customer pain point or improving a KPI (improved navigation, increased ease of use, making the site efficient, easy to find, better cross sells or an upsell, etc.)
- Brand or Visual refresh (site looks old and dated)
- Re-platforming or engineering infrastructure change (aging infrastructure)
It should be noted that some redesign projects have clearly stated and understood goals, some don’t. Some redesign projects have a clear definition of success and measurable KPIs, some don’t. You’ll just have to take this in stride. We’ll get more into the executive champion below.
Discovery & Definition Phase
As a UXer if you are involved in the discovery and definition phase then consider yourself lucky, because you are probably working for an organization with a high level of UX maturity. This is the phase where you discover the reasons why the redesign is happening, start to gather insights from users and define which KPIs you’re going to focus on and benchmark against.
User research methods that are helpful during the early phase of a redesign project:
- Ethnographic research: Visiting your customers / users for first hand observations.
- Surveys: If your redesign team has a good sense of the problems then a simple survey can suffice, otherwise feel free to get more granular.
- Competitive Evaluation: Business owners, product owners and executives are constantly looking at competitive sites and looking for (and making judgements on) what might or might not work. Gap analysis of features, functions, and ease of use is critical. UXers can conduct usability studies on competitor sites to bring end users’ perspectives into the larger organizational discussion.
- Baseline Study: Let’s call this Baseline study ONE. This is the best opportunity to conduct a Remote Unmoderated Large Sample Size study. It will allow you to capture KPIs that can then be compared with the new redesigned version. It will also show and clarify some hypothesis and assumptions built into the redesign goals.
In my experience, redesign projects don’t get funded without an executive champion making a business case for it. For a redesign to be successful, you have to understand the business case that the executive champion made. There has to be a core reason for kicking off a redesign project and all the investments that will follow. Your research plan should focus on a deeper understanding of the problem statement in the business case: “What customer problem(s) are we solving?” and “What exactly causes these problems?”
It’s also likely that someone’s bonus or future in the organization depends on the success of this redesign. Try to understand (if you can) how this executive champion’s bonus is calculated. This will help you connect design improvement KPI’s to something that the executive champion really cares about, such as improved NPS, increased upsells/cross sells, improved conversion/sales, increased average sale price, decreased call volume, etc.
Desk Research is a must. This might be controversial as it’s NOT considered a User Research method, but I still strongly recommend it. You can gain several insights from looking at data from Analytics teams, VOC surveys, Market research, and Data Analytics teams. Most organization have lots of data – it just might not be easily accessible or available. I always encourage UXers to dig deeper within their larger organization. Data exists, it’s just harder to find.
Do your due diligence to understand what User Experience Key Performance Indicators (KPIs) will be critical for your organization. Examples of what we typically use and see on a weekly basis are the following:
Behavioral KPIs (What they do):
- Success Rate
- Time on Task
- Page Views
Attitudinal KPI’s (what they say):
- Usability / Ease of Use
- Easy to Navigate
- Likelihood to Recommend (NPS)
- Content Clarity
- Brand Attributes
Design & Develop Phase
There are many questions that need researcher’s insights during design and development. Most UXers are involved during the design and development of any redesign project. While there are many research studies that are used during the design and development phase, these two categories in particular jump out:
1. Information Architecture:
Card Sort: Often redesign projects include some navigation or information architecture questions. Can users find the content? Do we have the right categories? Do these labels make sense? Card Sorting is the best way to map user’s mental models.
Tree Test: Often we follow up Card Sorts with a Tree test to ensure that the IA structure that is being proposed makes sense to the end users. Are they able to find something specific when presented with a tree of words? With no visual cues the taxonomy gets validated through a Tree test.
Screenshot Click Test & Screenshot Time-Out Test: Visual Designers, Art Directors, Creative Directors create several design variations and most visual designs are iterated for any redesign project. The number one goal is to ensure that the design supports user goals and business goals. Click tests are the best way to determine what areas of the screen users gravitate to when trying to achieve a certain goal.
2. Rapid iterative qualitative Think-out-loud usability studies:
If your teams use Agile development or Lean UX then the key is to provide rapid and actionable insights that the design & development teams can incorporate in their backlog. We often see multiple, but short, Think-Out-loud studies with a design prototype (Axure, Proto.io, InVision).
A lot has already been written about Agile Testing. The golden rule still applies: Be one or two sprints ahead. Plan for research ahead of time, especially when you don’t know what level of prototype will be available at the time of the study. Remote Unmoderated Think-Out-Loud studies with tools like UserZoom are rapidly gaining ground.
Launch & Measure Phase
As a consultant, a while ago, I was asked: “Do we need to get customer feedback on the design?”
My response was: “You don’t have to – you can choose to wait till the new design is live. Then you will get feedback. Can you risk waiting till then?” This executive then did everything he could to fund UX Research because he saw UX Research as a risk mitigation strategy.
Rapid usability testing during design and prototyping provides great insights into larger usability problems. But they don’t catch all problems. After working with many teams, year after year, its clear that certain types of issues are only evident once on a staging or in QA environment.
Research methods that are helpful during the latter phase of a redesign project:
- Usability Testing Think-Out-Loud: (Moderated or UnModerated) on Staging or QA environment.
- Baseline Study: Let’s call this Round TWO. You can now compare the redesigned new version, before release, with the live version baseline (Round ONE).
- Competitive Benchmark: Benchmark the new KPIs against those of your competitors.
- Voice of Customer: Collect user feedback from the live site – see what they think about the new design.
- Longitudinal Diary Studies: Track specific feedback from users over time.
A User Experience professional will most likely be involved in a website redesign project once every two years or so (I have no data, just basing this on anecdotal evidence). It helps to recognize that there are patterns most redesign projects go through; even if each redesign situation is unique, there is always anxiety, fear of the unknown, and fear that something will go wrong with the redesign and launch.
I have seen my fair share of failed redesigns and have learned a lot from those experiences which is why I wanted to share them with fellow practitioners. Talented Designers and Researchers are critical to almost every successful redesign I have seen and read about. So if you are involved with a redesign my absolute best and most fervent tip I can share with you is this – make the redesign data driven.
Don’t guess, measure. This will ensure there are no surprises.
Kuldeep Kelkar has more than 15 years of UX research, design and engineering experience and joins UserZoom from TATA Consultancy Services (TCS), where he was Head of User Experience. Prior to that, he led UX Design and User Research at PayPal for nearly a decade. He also has led UX efforts at KLA-Tencor and Icon Medialab.
Kuldeep is a renowned author and speaker, and he brings a BE in Industrial Engineering from Nagpur University and an Masters Degree in Human Computer Interaction from the Industrial Engineering department at Clemson University.