Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

Why most website redesigns fail and how to get yours right

Six strategies for success.

[Photo: Getty Images]

Profile photo of Gabriel Shaoolian

When business growth slows, a website redesign often feels like the obvious fix. Yet large digital initiatives tend to underdeliver. McKinsey research shows that about 70% of organizational transformations fail to meet their goals—not because of poor execution, but because strategy and alignment fall short.

In my experience working with mid- to enterprise-level organizations, redesigns fail when they’re not treated as business initiatives. The pattern repeats: Teams start with looks instead of purpose, rely on assumptions instead of data, and stop iterating once the new site goes live.

1. Begin with a business case, not a mood board

Most redesigns start with a creative direction rather than a business goal.

Before discussing layouts, define the job your site needs to do. For example:

  • Improve the qualified pipeline
  • Raise demo-to-close rates
  • Shorten time-to-value
  • Lift the average order value
  • Reduce the support load

When teams anchor projects to outcomes, it’s easier to choose what to build and what to skip.

Forrester’s 2024 U.S. Customer Experience Index found that overall CX quality across industries had fallen for a third straight year. Yet, customer-obsessed firms—those that consistently center on user needs—demonstrated 41% faster revenue growth, 49% faster profit growth, and 51% higher customer retention rates than their peers.

The takeaway for any redesign is clear: companies that anchor digital decisions to real customer outcomes outperform those chasing cosmetic change.

Before sketching a single layout, clarify the role your site plays in that larger picture. Who are the audiences that matter most? What actions signal business value? And how will you measure whether the redesign moved those numbers?

When those questions guide the brief, design becomes a way to express strategy and not a substitute for it.

2. Let real users show you what works

Internal feedback is valuable, but user behavior is decisive.

Make discovery concrete: Analyze funnel drop-offs, run session replays to find friction, test prototypes with users, and A/B the ambiguous decisions. You’ll learn faster and argue less when decisions are tied to evidence.

If you need a quick lens on where most revenue is lost, start at checkout. Baymard’s ongoing research shows that the global average cart abandonment rate hovers around 70%, with measurable gains available from simpler forms, clearer costs, and fewer required fields. That’s a redesign scope you can justify with data, not opinions.

To keep those insights from disappearing after launch, set up a lightweight research repository or knowledge base so findings are searchable and shared. NN/g’s guidance on research repositories is a pragmatic primer for teams that want UX evidence to remain useful and effective.

3. Treat performance as a product requirement

A site that drags won’t convert, regardless of how polished the visuals are.

In March 2024, Google officially made Interaction to Next Paint (INP) a Core Web Vital, replacing First Input Delay. That change put real-world responsiveness—for example, how quickly the interface reacts to taps, clicks, and key presses—at the center of performance work. If your redesign plan is incomplete if it doesn’t include an INP target and a performance budget.

Rather than rely on dated “X milliseconds = Y% conversion” claims, use your own baseline: Measure conversion by traffic segment before changes, set page-type performance targets, and track the delta after releases.

Google’s Web Vitals documentation offers clear thresholds and tooling. Pair it with focused fixes, such as critical-path CSS, image formats, script deferral, and caching strategy. You’ll see the impact in your numbers, not in someone else’s case study.

4. Make your structure do the heavy lifting

Strong redesigns flow from a content and IA strategy, not from comps.

Start with a lean architecture that reflects how people actually evaluate you. For B2B, map the evaluation journey (problem framing to approach to proof to pricing to procurement). For e-commerce, align categories, filters, and PDPs with the terms and attributes your buyers use—then validate with search logs and on-site behavior.

If your organization struggles to connect day-to-day work with customer journeys, move toward a journey-centric operating model. NN/g’s 2024 guidance explains how centering teams on journeys, rather than isolated pages or features, reduces UX debt and improves the handoffs that usually break experiences.

5. Give one owner the authority to say “no”

Redesigns often bloat because every stakeholder adds a requirement. That’s how performance budgets get blown and navigation turns into a junk drawer.

Appoint a single accountable owner—someone measured on business outcomes—who can trade off requests, hold the line on speed, and keep the site’s jobs clear.

If you need executive cover, share outcome-based proof points.

Forrester’s CX Index demonstrates why this discipline matters: Firms focused on customer outcomes outpace their peers in growth and retention. Use that to back a less-but-better stance when requests pile up.

6. Treat launch as your work’s first step

A redesign that ships without a follow-on measurement plan is a relaunch waiting to fail.

Build a simple quarterly loop. Pick a few KPIs aligned to outcomes, such as:

  • Qualified demo requests
  • PDP-to-checkout starts
  • Support-deflecting self-serve completions
  • UX evidence reviews
  • Small-fix ships

Then, repeat. The growing effect of a year’s six to eight thoughtful iterations consistently beats a “Big Bang” rebuild.

If you need a strategic frame for operating this way, Harvard Business Review’s analysis on meeting customer demand quickly makes the case: The winners move value closer to the moment of need. In digital, this means reducing friction, not increasing surface area. 

Strategy pays every day

Redesigns don’t fail because teams lack taste. They falter when there’s no clear job to do, no user evidence, slow pages, and no cadence after launch.

Define outcomes, validate with behavior, engineer for speed, and run a steady improvement loop. That’s how a redesign earns its keep quarter after quarter.

Gabriel Shaoolian is the CEO and founder of Digital Silk.