When I started PageSpeed Matters fourteen years ago, I thought the hardest part of building a web performance company would be the technical work. Diagnosing slow load times, untangling plugin conflicts, coaxing better scores out of stubborn WordPress themes. And that part is genuinely hard.
But the real education came from the business side of doing over 1,500 speed optimizations since 2020. Here is what I wish someone had told me earlier.
Prioritization is everything, and most people are bad at it
Early on, I would look at a new client’s website and want to fix everything. Every image that was too large, every render-blocking script, every opportunity to compress and cache. The problem is that not every fix carries equal weight. Some changes move the needle dramatically. Others take the same amount of time and produce almost nothing measurable.
Learning to identify high-leverage work over comprehensive work changed how I lead projects and how I built my team. We ask one question before touching anything: what is the single change most likely to reduce Largest Contentful Paint for this specific site? That discipline has made our results more consistent and our clients happier than any other decision we have made.
Clients do not buy technical fixes. They buy outcomes.
A client does not care about Time to First Byte or Cumulative Layout Shift by default. They care that their bounce rate is high, that their ads are more expensive than a competitor’s, or that their checkout page keeps losing customers before they finish the purchase.
The shift in how we communicate changed everything. When we frame a performance problem as a lost revenue problem, and show the client what even a one-second improvement in load time has historically meant for conversion rates, the conversation becomes completely different. It stops being a technical project with a budget and becomes a strategic investment with an expected return.
Every business leader who works with a technical team should push their team toward this framing. Ask not what can be fixed, but what that fix is worth.
Scalability demands that you document decisions, not just outcomes
For the first few years, most of the knowledge about how we approached optimizations lived in the heads of the people doing the work. That was fine when we were small. It became a problem when we grew.
We started building detailed internal documentation for every type of site we work on: how we audit a WooCommerce store, what our standard approach is for a Shopify Plus merchant with a large product catalog, which hosting environments tend to introduce which kinds of problems. Not just what we do, but why we do it that way.
The result was that new team members could get productive faster, and our work became more consistent across clients. If you are building a service business and you want to grow without everything depending on you personally, documentation is not optional. It is the business.
Slow is expensive. Speed is undervalued.
One of the consistent surprises from working inside so many clients’ websites is how much slow performance costs relative to how little attention it gets. A company might spend thousands of dollars a month on paid traffic and not realize that their landing page is losing half of those visitors before the page even finishes loading.
Research published by Google and corroborated by multiple independent studies has shown that a one-second delay in mobile load times can reduce conversion rates by several percentage points. For an e-commerce site doing real volume, that is not a performance issue. That is a revenue issue. And yet it often does not make it onto anyone’s quarterly priority list until someone starts connecting the dots.
As a business leader, the lesson is to look for the invisible costs in your operations. They are rarely the ones being loudly complained about. They are the slow leaks that nobody has bothered to measure.
The right metric changes everything
We spent years helping clients improve their Google PageSpeed Insights scores. It is a useful proxy, but clients would sometimes get frustrated when their score improved and their business results did not change noticeably. Over time, we moved toward focusing on Core Web Vitals as measured in real user data rather than lab results alone, and toward tying performance improvements directly to conversion rate changes when we could.
That shift made our work more meaningful and our clients more satisfied. The lesson applies broadly: the metric you optimize for shapes everything your team does. Choose it carefully.
Fourteen years and 1,500 optimizations have taught me that performance work is really just business analysis with technical tools. The problems are always human before they are technical. The solutions always come back to the same disciplines: clarity, prioritization, and consistent execution.
Be the first to comment