How an IA Overhaul Drove a 63% Increase in Lead Inquiries

*To respect NDA restrictions this case study has been anonymized

Jump Links:

Finding the Problem

Designing the Solution

Showing the Impact

The Essentials

Vector Span is a global infrastructure leader. Their site suffered from 3,000-page Information Architecture (IA) focused on internal silos rather than client acquisition and showing an integrated service offering.

Who were the users?

Primary Users: Potential public and private sector clients needing support with large infrastructure projects.

Secondary Users: Potential new employees seeking roles such as engineering and HR.

What was the solution?

A logical, client-centric IA restructure that merged fragmented service silos, resulting in:

+63% increase in qualified inquiries post-launch

-75% decrease in total page count

What was unique?

Internally sensitivity: having to contend with potential negative reactions of moving business unit pages around.

Performance-Led UX: Leveraged GA4 analytics and tree testing to move the needle on key business metrics, including a 63% increase in lead inquiries

Define & Research

📍 Where were they starting from?

A crawl with screaming frog showed over 3000 pages, many of which were hidden to users but still add bloat to the site.

Define & Research

What were the top level themes?

1

The existing IA functioned as an internal org chart, not a client acquisition tool - severely limiting digital maturity and lead generation

2

The primary strategic problem was to re-align a 3,000-page content hierarchy toward external client needs

Highlight: Customers shifting to digital experiences

67%

B2B buyers prefer to research online on their own vs traditional customer service methods like sales calls. (Link)

This is crucial as the current Vector site was not optimized for this independent research.

Define & Research

31 Survey Responses Later…

Discovering broad trends inside the organization in terms of pain points and business goals
Supplementing the moderated interviews

Brand awareness stood out as a clear winner for key business goals

Define & Research

8 Moderated Interviews

Remote interviews were conducted to generate detailed findings about business struggles and opportunities

Who did I speak to?

Finding #1: Prioritize Content that Proves Expertise Through Impact

How did this Vector Span project tangibly benefit a client? I recommended moving away from generic claims to impactful narratives.

“There's too much generalization… We do this, we're good at this… from the client's perspective, everybody says that.”

“There's too much generalization… We do this, we're good at this… from the client's perspective, everybody says that.”

“There's too much generalization… We do this, we're good at this… from the client's perspective, everybody says that.”

Finding #2: Prepare for Generational Changes Amongst Clients

Younger more digitally savvy clients are beginning to emerge with new browsing and purchasing habits - interests in data, sustainability, and innovation.

"Clients are leaning more on digital sources of information than personal relationships like they used to."

"Clients are leaning more on digital sources of information than personal relationships like they used to."

"Clients are leaning more on digital sources of information than personal relationships like they used to."

Design and Testing

Design & Testing

🎯 What were my design goals?

The discovery process outlined above identified the following to focus my design efforts.

1

Consolidating fragmented service and solutions offerings

2

Highlighting sustainability and innovation content

3

Focus on prioritizing high-value content (e.g., project case studies).

The Initial Redesigned IA

Documentation helped the client understand our rationale for the updates

To prevent surprises, removed pages were tracked

An example of how a redesigned section was presented to the client

Highlight: A Failed Design Direction

The business often slots solutions under services. So I tried to put solutions neatly under them. But…

Solutions that span across multiple industries would have led to too much duplication.

The business often slots solutions under services. So I tried to put solutions neatly under them. But…

Solutions that span across multiple industries would have led to too much duplication.

Design & Testing

Testing Breakdown

Highlight: Why did I pick tree testing?

Provides direct feedback on how users will navigate through the structure; resulting in targeted improvements

Keeping future conversations with the client centered on client/user needs

Who did we test with?

20 participants divided across job seeker and client personas were recruited through Respondent with a 75$ incentive.

Design & Testing

Theme #1: Users mirrored the need for example projects on the website.

This could mean specific technologies, expertise, and locales
Individual project case studies or dispersed across the site in callouts

"When I look at projects, it just makes it more clear to me what the company exactly does."

How did I change the designs based on this?

Placing orphaned central projects page in the main navigation

Established a rule that requires project proof for new pages

Design & Testing

Theme #2: Mixed reactions to such a broad service offering.

How would it be feasible for a company to perform at a high level in each category

"It looks like they're in basically every industry... I'm like, wow, they do it all, which makes me wonder... jack of all trades and master of none?"

How did I change the designs based on this?

How did I change the designs?

I facilitated workshops to identify pages to prioritize and which pages can be removed
Industries that were not actively competitive or lacked projects were removed
🚨 Without user testing data - the client likely would have backslid into their old habits of mirroring their internal structure over user needs

I facilitated workshops to identify pages to prioritize and which pages can be removed
Industries that were not actively competitive or lacked projects were removed
🚨 Without user testing data - the client likely would have backslid into their old habits of mirroring their internal structure over user needs

I facilitated workshops to identify pages to prioritize and which pages can be removed
Industries that were not actively competitive or lacked projects were removed
🚨 Without user testing data - the client likely would have backslid into their old habits of mirroring their internal structure over user needs

Miro board used for workshops

Highlight: a major reduction in unnecessary pages

Highlight: major reduction in pages

75%

Reduction in "Industry" pages

35%

Reduction in "Services" pages

Redesigning the navigation bar

I took the initiative to provide my team with some nav designed based on the IA updates. Here are some highlighted improvements:

Where they started from

Retitled their sections to be more user oriented and less like their internal organization. (Ex. Industries -> "Who We Serve"
Created category pages:
- allowed users to view all of Vector's offerings
- aid reserving the list in the navigation bar as their highest priority industries
Sustainability was moved to the primary nav as per our research findings

Outcomes and Reflections

Outcomes & Reflections

✅ Successes

By the numbers

+53% in Unique Users
+84% in page views per session
+63% in contact form submission

*three months after launch

Leadership Buy In

Presented the final designs to the Chief Client Officer and 14 senior staff. Was crucial for securing organizational support for the upcoming changes.

Setting up for future growth

Vector struggled with a continual stream of new pages
Established organizational that guide when a page can be created.

🗺️ Opportunities

Global Expansion

The current IA focused on the North American market, creating the opportunity to explore localization for global regions like Asia and Europe.

Global Expansion

Extending the new IA and visual design to other fragmented web properties, such as subsidiary company sites and the corporate intranet.