15 Research and Analytics Questions to Answer Before Redesigning a Website

Before a redesign, answer two sets of questions. From analytics: who visits, where they come from, which journeys lead to enquiries or sales, where people drop out, what they search for, which pages and devices underperform, and what your baseline is. From users: what they came to do, what nearly stopped them, what words they use, what they compare you with, and what they need to trust you. The answers decide what the redesign should change, and give you a baseline to judge it.

Redesigns often start from opinions ("the site looks dated") and end with a new look but the same problems. This list is for marketing leads, product owners and founders briefing a redesign. Treat the answers as evidence for a brief, not a design. For whether you need a redesign or a rebuild, see website redesign vs rebuild.

1. Who visits, and from where? Split traffic by channel, country and device. A site that gets most enquiries from mobile search needs different priorities from one driven by desktop referrals.

2. Which pages do people land on? Landing pages, not the home page, are often the real front door. Search Console and analytics landing-page reports show which pages carry the most weight.

3. Which journeys lead to an enquiry or sale? Look at the paths of visitors who converted. Which pages did they see? How many visits did it take? Protect what works.

4. Where do people drop out? Find the steps with the largest drop: a service page with many exits, a form with many starts and few submissions, a checkout step. These are candidates for fixing first.

5. What do people search for on the site? Site search terms reveal missing content and confusing navigation in the visitor's own words.

6. Which pages and devices underperform? Compare conversion and engagement by template and device. A page that converts well on desktop and poorly on mobile points to a mobile layout problem, not a content problem.

7. What queries bring people from search? Search Console shows the queries you rank for and their click-through rates. Make sure the redesign keeps or improves the pages behind valuable queries; URL and content changes can lose rankings.

8. What is the baseline? Record conversion rates, enquiry volume and quality, key page speed measures and top landing pages before you change anything, or you will not know whether the redesign helped. See measuring conversions across channels.

9. What were they trying to do? Ask recent visitors, customers or leads what brought them to the site and whether they managed it. Their tasks should shape the navigation.

10. What nearly stopped them? Customers who converted anyway can tell you about doubts and friction: unclear pricing, missing information, a confusing form.

11. What words do they use? Note how people describe their problem and your service. Navigation and headings in customers' language are easier to scan than internal jargon.

12. What did they compare you with? Knowing the alternatives, including doing nothing or using a spreadsheet, tells you what the site must explain.

13. What did they need to trust you? Ask what reassured them or what they looked for: examples, process, pricing, people, reviews.

14. Where do they get stuck when they try? Watch five or so people attempt key tasks on the current site. Usability testing reveals problems analytics only hints at. See usability testing.

15. What do sales and support hear? Staff who talk to customers know the recurring questions and misunderstandings. Review support tickets and sales call notes for patterns.

A redesign is most likely to help when it starts from answered questions: what visitors are trying to do, where they fail, why, and what must not be broken. Spend a little time on analytics and a little with users, write a brief from the evidence, and measure against a baseline. For research methods in more depth, see user research methods.