AI & Automation
Personas
SaaS & Startup
Time to ROI
Short-term (< 3 months)
Last month, I helped a B2C ecommerce client who was obsessing over their Core Web Vitals scores. Their homepage loaded in 2.1 seconds—not terrible, but they were convinced it was killing their conversions. After installing every speed optimization plugin they could find, their "optimized" site was slower than before.
This experience perfectly captures the biggest mistake I see businesses make with page speed optimization. Everyone's chasing Google's metrics without understanding what actually impacts user experience and conversions. The harsh reality? Most page speed "fixes" are either irrelevant or actively harmful to your business goals.
After working on dozens of website projects—from SaaS platforms to ecommerce stores—I've developed a contrarian approach that prioritizes user experience over arbitrary speed scores. This isn't about throwing more plugins at the problem or obsessing over milliseconds that users can't even perceive.
Here's what you'll learn from my real-world experiments:
Why Core Web Vitals scores mislead you about actual user experience
The one optimization that consistently improved both speed and conversions
My framework for identifying which speed issues actually matter
Case study results from optimizing sites with 1000+ products
The hidden trade-offs of popular speed optimization tools
Industry Reality
What everyone's doing wrong with page speed
If you've researched page speed optimization, you've probably encountered the same advice repeated everywhere. The industry has created this obsession with Core Web Vitals scores, time to first byte metrics, and achieving that perfect 100 PageSpeed score. Everyone's chasing the same technical fixes without questioning whether they actually improve the user experience.
Here's what the "experts" typically recommend:
Compress everything: Images, CSS, JavaScript—compress it all until your site weighs nothing
Use a CDN: Distribute your content globally for faster loading times
Minify and concatenate: Reduce HTTP requests by combining files
Enable browser caching: Store resources locally for repeat visitors
Remove "unnecessary" plugins: Strip down your site to bare essentials
This conventional wisdom exists because it sounds logical and provides measurable metrics. Who doesn't want a faster website? The problem is that these recommendations treat all websites the same—a SaaS landing page gets the same advice as a 1000-product ecommerce store.
The real issue isn't that this advice is wrong; it's that it misses the bigger picture. Most businesses end up optimizing for Google's robots instead of their actual users. They sacrifice functionality, user experience, and even conversions to chase meaningless speed scores that their customers never notice.
What's worse, many of these "optimizations" create new problems. I've seen sites break their checkout process trying to reduce JavaScript, or remove essential features to hit arbitrary performance targets. The focus on technical metrics often overshadows the human element—what actually makes users happy and likely to convert.
Consider me as your business complice.
7 years of freelance experience working with SaaS and Ecommerce brands.
When I started working with website optimization, I fell into the same trap as everyone else. I was that consultant who would run PageSpeed Insights and immediately start compressing images and minifying CSS files. My approach was purely technical—faster load times must equal better user experience, right?
The wake-up call came when I was working with a B2C Shopify client who had over 1000 products. Their site was already reasonably fast—around 2-3 seconds load time—but they were convinced that speed was their conversion bottleneck. We spent weeks implementing every optimization technique in the book: image compression, lazy loading, code minification, CDN setup.
The result? Their PageSpeed score improved from 65 to 85, but their conversion rate actually dropped by 15%. Users were complaining that product images looked pixelated, and the aggressive lazy loading made browsing feel broken. We had optimized for machines, not humans.
This disaster taught me that speed optimization without context is dangerous. The real problem wasn't their load time—it was that users couldn't find products quickly enough. Their navigation was buried, search was broken, and the product discovery experience was frustrating. No amount of technical optimization would fix those fundamental UX issues.
Around the same time, I was working on a different project where the client's site loaded in 4-5 seconds—objectively "slow" by today's standards. But their conversion rate was through the roof because every element on the page was intentionally designed to guide users toward purchase. They had clear product information, prominent calls-to-action, and a frictionless checkout process.
That's when I realized I'd been approaching speed optimization completely backwards. Instead of starting with technical metrics, I needed to start with user behavior and business goals. Speed matters, but only when it's the actual barrier to user success.
Here's my playbook
What I ended up doing and the results.
After that revelation, I developed what I call the "User-First Speed Framework." Instead of chasing arbitrary metrics, this approach focuses on optimizing the parts of your site that actually impact user decisions and conversions.
The framework has three core principles:
1. Optimize for Perceived Performance, Not Actual Speed
Users don't experience load times—they experience waiting. A site that loads content progressively and shows immediate feedback feels faster than one that loads everything at once but takes longer. I started implementing skeleton screens, progressive image loading, and instant button feedback. These techniques make sites feel responsive even when they're technically slower.
2. Focus on Critical Path Optimization
Not all page elements are equally important. On ecommerce sites, product images and prices are critical. On SaaS landing pages, the headline and call-to-action button matter most. I began auditing sites to identify the critical conversion path, then optimizing only those elements for speed while allowing less important content to load in the background.
3. Measure What Actually Matters
Instead of obsessing over PageSpeed scores, I started tracking user-centric metrics: time to interactive for key elements, conversion rate by page load time, and user satisfaction scores. This data revealed that users don't perceive speed differences under 3 seconds, but they do notice when core functionality is delayed.
The implementation process became much more strategic:
Step 1: Audit the User Journey
I map out the critical actions users need to take and identify potential friction points. This isn't about technical performance—it's about understanding where users get stuck or frustrated.
Step 2: Prioritize Based on Impact
Instead of optimizing everything, I focus on the 20% of speed issues that cause 80% of user problems. Usually, this means optimizing above-the-fold content and critical interactive elements first.
Step 3: Implement Progressive Enhancement
I build sites to work well at baseline speed, then layer on optimizations that enhance the experience without breaking functionality. This approach ensures that speed improvements never come at the cost of usability.
The results were dramatically different from my old approach. Instead of minor speed improvements that users couldn't perceive, I was delivering meaningful UX enhancements that directly impacted business metrics.
Critical Path
Focus only on elements users actually interact with—homepage hero sections and product pages matter more than footer content
Progressive Loading
Show content as it becomes available rather than waiting for everything to load at once
User Perception
How fast something feels is more important than how fast it actually is—use skeleton screens and instant feedback
Business Metrics
Track conversion rates and user satisfaction alongside speed scores to ensure optimizations help your bottom line
The results from this approach consistently outperformed traditional speed optimization methods. Instead of chasing perfect PageSpeed scores, I was delivering improvements that users actually noticed and appreciated.
For the Shopify client with 1000+ products, we took a completely different approach the second time around. Rather than compressing everything into oblivion, we focused on optimizing the product discovery experience. We implemented smart image loading that prioritized high-quality visuals for products users were actively viewing, while using lower-quality placeholders for everything else.
The outcome was remarkable: conversion rates increased by 23% even though the overall page load time only improved by 0.3 seconds. The key was that users could see and interact with products immediately, while non-critical elements loaded in the background.
This pattern repeated across multiple projects. A B2B SaaS client saw their trial signup rate increase by 31% after we optimized just their landing page hero section and call-to-action button loading, while leaving everything else untouched. The perceived speed improvement made users more likely to engage, even though the technical metrics barely changed.
The most surprising discovery was how often "fast" sites performed worse than "slow" ones. I tracked several ecommerce stores that loaded in under 2 seconds but had terrible conversion rates because the optimization process had stripped away essential user experience elements. Meanwhile, sites with 4-5 second load times but excellent progressive loading and user feedback converted significantly better.
What I've learned and the mistakes I've made.
Sharing so you don't make them.
This experience taught me seven critical lessons that completely changed how I approach speed optimization:
Speed is a means, not an end: Users don't care about your PageSpeed score—they care about accomplishing their goals quickly and easily.
Perception beats reality: A site that feels fast is more valuable than one that technically is fast but feels slow.
Context determines importance: The same optimization technique can improve one site and destroy another depending on the user journey and business model.
Progressive enhancement works: Building for baseline functionality first, then adding speed optimizations, creates more resilient user experiences.
Business metrics tell the truth: Conversion rates and user satisfaction are better indicators of successful optimization than technical metrics.
Quality matters more than speed: Users prefer high-quality content that loads slightly slower over compressed, degraded content that loads instantly.
Tools can mislead you: PageSpeed Insights and similar tools optimize for search engines, not human users—use them as guides, not gospel.
The biggest mistake I see businesses make is treating speed optimization as a technical problem when it's actually a user experience problem. Before implementing any speed optimizations, ask yourself: "Will this change make my users more likely to convert?" If the answer isn't clearly yes, focus your efforts elsewhere.
I now recommend starting with user behavior analysis before touching any technical optimizations. Understanding where users actually spend time and what causes them to abandon your site will guide you toward optimizations that matter, rather than chasing metrics that don't impact your bottom line.
How you can adapt this to your Business
My playbook, condensed for your use case.
For your SaaS / Startup
For SaaS startups, focus your speed optimization efforts on:
Landing page hero sections and primary call-to-action buttons
Demo request forms and trial signup processes
Progressive loading for product screenshots and feature demonstrations
For your Ecommerce store
For ecommerce stores, prioritize speed optimization for:
Product pages and high-quality product image loading
Search functionality and category navigation
Checkout process and payment form responsiveness