webshunya icon logo

Seen

How to Test Your Sitemap and Navigation Before a Website Redesign

Check whether people can find key services and information before you approve your new sitemap.

Milan Gohil

·

9 min read

A tactile paper branching pathway with a tiny dark human figure at its entrance.
A tactile paper branching pathway with a tiny dark human figure at its entrance.

OVERVIEW

Here is a short version of what I write

Test a proposed sitemap by asking representative users to find information for realistic tasks. Tree testing helps reveal whether the hierarchy and labels make sense before visual design influences the result.

  • Choose important tasks and check the existing experience.

  • Use card sorting if the content grouping is unclear.

  • Tree-test the proposed structure and examine wrong turns.

  • Revise labels and paths, then test the navigation prototype.

A new navigation can look cleaner and still make the website harder to use. Before approving a new sitemap, test something much simpler: Can people find the things they came for? You do not need to become an information-architecture expert. Start with the important tasks on your website, see where people expect to find them, then test the proposed structure before the visual navigation is designed.

NN/g recommends a similar redesign sequence: test the existing experience, use card sorting when you need to understand how users group information, and use tree testing to evaluate a proposed hierarchy before investing heavily in design or development.

The process in one view


Test the structure before the design. 1. Define tasks: What must visitors find?: Test the current website. 2. Check grouping: Use card sorting if grouping is unclear: Prepare a proposed sitemap. 3. Test the tree: Can people find the right destination?: Improve labels and paths. 4. Test the menu: Build the navigation prototype: Run a usability test. Choose the test that answers the uncertainty you have.

You can use this process on an existing business website before redesigning it, or use the same logic when planning a larger new website from scratch.

Want a quick first check? Give your website to AI

If your AI tool can access public webpages, paste your website into this prompt. It will not tell you how real users behave. It will help identify what is worth testing.

Copy this prompt

Review this website before I redesign its sitemap and navigation:

WEBSITE:
[PASTE WEBSITE URL]

Act as an information architect and UX reviewer.

Do not redesign the website yet.

First inspect the publicly accessible website and give me:

1. A simplified map of the current website structure:
Homepage
→ primary navigation
→ main sections
→ important subpages

2. The main products, services, resources and actions a visitor may need to find.

3. 8-10 realistic user tasks.

Do not use the exact navigation label inside the task.

For example, instead of:
"Find Equipment Repair"

write:
"One of our machines has stopped working and we need to know whether this company can repair it. Where would you look?"

4. Possible findability problems, including:
- overlapping navigation categories
- unclear labels
- pages that appear too deeply buried
- important pages with weak internal linking
- similar information appearing in multiple sections
- categories that seem based on internal company language
- important information with no obvious location
- pages that could reasonably fit under multiple sections

5. For every issue, separate:
OBSERVED:
What you directly found on the site.

HYPOTHESIS:
What you think may confuse users.

NEEDS TESTING:
What cannot be confirmed without real users.

6. Tell me which research method would help:

CARD SORTING
when I do not know how users would naturally group the information.

TREE TESTING
when I already have a sitemap/hierarchy and need to test whether users can find things.

FIRST-CLICK / PROTOTYPE TESTING
when the structure may be correct but I need to test the actual menu/interface.

7. Finish with:
- what appears worth keeping
- what should be investigated
- what should be tested with users
- the 10 highest-priority navigation/findability issues

Use that output as a starting point. Then test the important assumptions yourself.

Step 1: Write down what people actually need to find

Do not begin with:

What pages should be in our navigation?

Begin with:

What does someone need to accomplish on this website?

Imagine an industrial equipment company. Important tasks might be:

  • find a replacement part for an existing machine

  • check whether the company provides on-site repair

  • find technical documentation

  • understand whether equipment is suitable for food manufacturing

  • contact someone about an urgent breakdown

These are much more useful than starting with:

Products

Services

Solutions

Resources

because those are your categories, not the visitor's problem.

Turn them into testable tasks

Do not tell people which menu item to use. Bad task:

Find Pump Repair.

Better task:

A pump on your production line has failed and you want to know whether this company can repair it. Where would you look?

Why? Because writing the answer into the task makes the navigation appear easier than it really is. Tree testing specifically uses realistic tasks to see whether people can find destinations in a hierarchy.

Step 2: Test the current website before replacing it

Give someone the website and one task at a time. Record only what is useful:

Task

Found?

First place checked

Wrong path

Notes

Find repair service

Yes

Services

—

Direct

Find product manual

Yes

Resources

Support

Hesitated

Find spare parts

No

Products

Service

Gave up

Find emergency contact

Yes

Contact

—

Direct

You do not need a complicated UX report. You want to know: Where did people expect the information to be? That tells you much more than whether your current menu looks old.

Step 3: Identify what type of problem you actually have

Different problems need different tests.


Which navigation test should you use?. Card sorting: How should information be grouped?. Tree testing: Can people find content in the hierarchy?. First-click testing: Is the first navigation choice obvious?. Usability testing: Does the actual menu interaction work?. You do not need every method on every project.

You are unsure about…

Use

How information should be grouped

Card sorting

Whether people can find content in a proposed hierarchy

Tree testing

Whether the first navigation choice feels obvious

First-click testing

Whether the actual menu interaction works

Prototype/usability testing

NN/g describes card sorting as a way to understand how users naturally categorize information, while tree testing evaluates whether users can find key items inside an information architecture. You do not need every method on every project. Use the method that answers the thing you do not know.

Step 4: If the grouping is unclear, test how people group it

Imagine the website currently contains:

  • Pumps

  • Motors

  • Valves

  • Installation

  • Maintenance

  • Repair

  • Spare Parts

  • Engineering

  • Rental

  • Monitoring

Internally, the company may organize these by departments. Customers may think differently. They might see:

Equipment: Pumps Motors Valves Spare Parts

Services: Installation Maintenance Repair Engineering

Other: Rental Monitoring

Or they may organize everything by industry. That is exactly the kind of uncertainty card sorting helps investigate: it exposes how users mentally categorize information rather than asking the internal team to guess.

Step 5: Build the proposed sitemap

Now create the simplest hierarchy that reflects what you learned.


From overlapping labels to a testable structure. Before: Products: pumps, motors, valves: Solutions: systems, engineering: Services: repair, engineering, installation. Problem: Engineering appears twice: Products and Solutions overlap: Visitors must guess where to look. Proposed: Equipment: products and spare parts: Services: install, maintain, repair, support: Clear Resources, Support and Contact routes. Illustrative structure—not a proven result. Test it with real customer tasks.

Before

  • PRODUCTS

    • Pumps

    • Motors

    • Valves

  • SOLUTIONS

    • Pump Systems

    • Engineering

    • Industrial Systems

  • SERVICES

    • Repair

    • Engineering

    • Installation

There is immediate overlap. Where does engineering belong? What is the difference between Products, Solutions and Services?

Proposed structure

  • EQUIPMENT

    • Pumps

    • Motors

    • Valves

    • Spare Parts

  • SERVICES

    • Installation

    • Maintenance

    • Repair

    • Engineering Support

  • INDUSTRIES

  • RESOURCES

  • SUPPORT

  • CONTACT

This is not automatically the correct answer. It is now a structure you can test. That difference matters.

Step 6: Tree-test the proposed sitemap

Do not design the mega menu yet. Do not spend time on icons, hover states or dropdown animation. Give people the hierarchy without visual design and ask them to complete the important tasks. NN/g notes that tree testing is particularly useful because teams can evaluate menu labels and category structure before page layouts or navigation interfaces are designed. For each task, pay attention to four things.

Did they find it?

The basic success/failure.

Where did they look first?

This tells you what category felt most likely.

Did they go directly there?

Someone reaching the correct answer after three wrong paths is different from someone choosing it immediately.

Where did unsuccessful users finish?

The wrong destination often tells you what label or grouping caused the confusion. NN/g's guidance for interpreting tree tests specifically includes success, first click and directness as useful measures.

What the result might look like


Compare findability before and after. Set one task: Find maintenance documentation: Use the same task for both structures. Record the old structure: Correct destination: ____: Correct first section: ____: Wrong turns: ____. Record the proposed structure: Correct destination: ____: Correct first section: ____: Wrong turns: ____. Review the difference: Which labels caused confusion?: Did the new structure help?. Blank test worksheet. Add actual participant results; do not invent percentages.

Task

Find maintenance documentation for equipment you already own.

Illustrative measure

Old structure

Proposed structure

Found correct destination

55%

86%

Chose correct section first

32%

74%

Frequent backtracking

Yes

Low

Most common wrong choice

Resources

Support

These figures are illustrative, not results from a Webshunya project. Use your own test data when comparing structures. The point is not getting a perfect score. The point is seeing whether the proposed structure removes confusion.

Step 7: Don't stop at the main menu

People do not find everything through:

Homepage → Navigation → Page.

They also arrive through:

Google → Service/Product Page → Related Information

or:

Article → Relevant Service

or:

Product → Documentation

or:

Footer → Support

So once the hierarchy works, check the relationships between important pages. Google recommends a logical site structure, links to important pages from other relevant pages, and descriptive internal-link text that helps people and Google understand the destination. Ask:

If someone lands directly on this page from Google, where should they naturally go next?

That is part of findability too.

Step 8: Don't change URLs just because the sitemap changed

Suppose this URL already exists:

/company-training

After the redesign, the page moves visually under:

  • Services

  • → Corporate Training

That does not automatically mean the URL must become:

/services/corporate-training

Navigation hierarchy and URL structure are related, but they are not the same thing. If an existing URL still accurately represents the page and has value, changing it only to make the new sitemap look cleaner can create unnecessary migration work. Google recommends crawlable, understandable URL structures, but the navigation hierarchy itself does not require every URL to mirror every folder level.

Step 9: Then test the actual navigation design

Once the hierarchy works, now design the navigation. This is where you test things the sitemap cannot answer:

  • Can people see where the menu opens?

  • Does the mobile navigation work?

  • Is a mega menu overwhelming?

  • Are important items visually clear?

  • Do dropdowns behave predictably?

  • Is the first click obvious?

  • Can people reach important actions quickly?

A successful tree test proves only that the structure works reasonably well. It does not prove that your final header or mobile menu works. That is why information-architecture research and navigation-interface testing are separate stages. NN/g's IA guidance likewise separates structuring information from navigation-design choices and usability research.

If you have very little time, do this

You do not need a major UX research project for every small business website. Test five tasks:

Test

Ask someone to…

Main offering

Find one important product or service

Proof

Find evidence that the company can do the work

Buying information

Find pricing/process/specifications where relevant

Contact/support

Reach the correct next step

Deep content

Find one important page that is not in the main navigation

If several people struggle with the same task, investigate it before redesigning the menu around the existing structure.

When is the sitemap ready?

Before moving heavily into visual navigation design, you should be able to say:

  • ✓ We know the important user tasks.

  • ✓ We know which existing pages matter.

  • ✓ Important information has a logical home.

  • ✓ Major category overlaps have been resolved.

  • ✓ Navigation labels make sense outside the company.

  • ✓ Important tasks have been tested against the proposed structure.

  • ✓ Important SEO pages/URLs have been identified before migration decisions.

  • ✓ We know which pages need contextual/internal links.

  • ✓ The actual navigation UI will still be tested separately.

That is enough. You do not need to prove that every visitor will behave identically. You need enough evidence to stop making the entire structure from internal assumptions.

A sitemap should solve finding problems before it solves design problems

The easiest redesign mistake is keeping the same information problems inside a nicer header. If people cannot tell whether something belongs under Solutions, Services or Capabilities, redesigning those three buttons will not answer the question. If documentation is buried in a place customers never expect, giving the menu rounded corners will not make it easier to find. So before designing the navigation: List what people need to find.

Test the current routes. Change the grouping when necessary. Test the proposed structure. Then design the interface around something you already have reason to believe works. That is a much better place to start a redesign.

Sources and further reading

ARTICLE WRITTEN BY

Milan Gohil

I like looking beyond the obvious problem to understand what is really holding a business back. I write about the patterns, questions and practical lessons I discover while working through real challenges

ARTICLE WRITTEN BY

Milan Gohil

I like looking beyond the obvious problem to understand what is really holding a business back. I write about the patterns, questions and practical lessons I discover while working through real challenges

Book your strategy call

Book your call now

On this page
AI Summary

Book your strategy call

Book your call now

All Blogs

All Blogs & Articles

Two small paper browser windows connected by a sturdy orange bridge; a continuous thin black route crosses intact from old to new, with a small magnifying glass beside the destination..

Found

How to Redesign or Migrate a Website Without Losing SEO

Two understated cut-paper hands pass a single orange key above an open ivory folder holding a miniature website card and neatly arranged small tools.

Chosen

Website Handover Checklist: What Your Developer Should Give You

A single central paper website card branches into three distinct service cards carrying only simple object icons: a gear, a pencil and a wrench.

Found

How Many Service Pages Does Your Website Actually Need?

A small paper analytics panel with an unlabeled black line that dips; a large orange-rimmed magnifying glass examines a disconnected paper link below the chart.

Found

Why Your Website Traffic Dropped After a Redesign and What to Check First

A tactile paper branching pathway with a tiny dark human figure at its entrance.

Seen

How to Test Your Sitemap and Navigation Before a Website Redesign

A miniature paper shopping trolley carrying three simple product boxes crosses a broad orange bridge between two spare storefront cards.

Found

Ecommerce Migration Without Losing Product URLs, Feeds or Customer Journeys

Two small paper browser windows connected by a sturdy orange bridge; a continuous thin black route crosses intact from old to new, with a small magnifying glass beside the destination..

Found

How to Redesign or Migrate a Website Without Losing SEO

A single central paper website card branches into three distinct service cards carrying only simple object icons: a gear, a pencil and a wrench.

Found

How Many Service Pages Does Your Website Actually Need?

A tactile paper branching pathway with a tiny dark human figure at its entrance.

Seen

How to Test Your Sitemap and Navigation Before a Website Redesign

Two understated cut-paper hands pass a single orange key above an open ivory folder holding a miniature website card and neatly arranged small tools.

Chosen

Website Handover Checklist: What Your Developer Should Give You

A small paper analytics panel with an unlabeled black line that dips; a large orange-rimmed magnifying glass examines a disconnected paper link below the chart.

Found

Why Your Website Traffic Dropped After a Redesign and What to Check First

A miniature paper shopping trolley carrying three simple product boxes crosses a broad orange bridge between two spare storefront cards.

Found

Ecommerce Migration Without Losing Product URLs, Feeds or Customer Journeys

Before you invest in another fix, make sure it is the right one.

Your website may not need more pages, more content or another redesign. It may need clarity on what is actually holding it back.

Before you invest in another fix, make sure it is the right one.

Your website may not need more pages, more content or another redesign. It may need clarity on what is actually holding it back.

Something isn’t working as well as it should?

Your website may not need more pages, more content or another redesign. It may need clarity on what is actually holding it back.

© 2026 Webshunya Studio®. All rights reserved.