Performance
Website speed optimization and Core Web Vitals
Speed is a conversion problem before it is a ranking problem. A visitor deciding whether to wait for your site does not know or care that Google measures this. They just leave.
The numbers Google actually measures
Core Web Vitals are three metrics with published thresholds, so this is one of the few parts of SEO with an objective pass mark. Largest Contentful Paint should happen within 2.5 seconds. Interaction to Next Paint should be 200 milliseconds or less. Cumulative Layout Shift should stay at 0.1 or below. Interaction to Next Paint replaced First Input Delay and became a stable Core Web Vital in 2024.
Source: web.dev, Web Vitals. Those are the numbers we optimize against, and we report them from real measurements rather than from a score out of a hundred that flatters everyone.
Alaska makes this matter more, not less
Connection quality outside the Railbelt is genuinely worse than the averages most performance advice is written against. A page that feels quick on Anchorage fiber can be unusable on a rural connection or on a phone with two bars in the Valley.
That is not an edge case if you serve the state. It is a meaningful share of your customers, and they are often the ones with the fewest local alternatives to you. Building for the slow case costs nothing extra and is the difference between getting the call and not.
What usually causes it
In rough order of how often we find it: uncompressed images, which is almost always the biggest single win and the easiest fix. Then render-blocking scripts, usually a chat widget or a tag manager loading four tools nobody uses. Then fonts loading in a way that leaves text invisible for a beat. Then layout shift, where the page jumps as images and ads arrive and someone taps the wrong thing.
Underneath all of it, often, is the platform. A page builder that ships 800KB of JavaScript to render a page of text has a ceiling you cannot optimize past. When that is the situation we will say so rather than bill you for months of tuning that gets you from awful to merely slow. Sometimes the honest recommendation is a rebuild, and sometimes it genuinely is not.
How we approach a slow site
Measure, fix the biggest thing, measure again. Resist the urge to change ten things at once.
Measure real users
Field data from actual visitors on actual devices, not just a lab test from a fast machine on a fast connection. The two often disagree and the field data is the one that counts.
Rank the fixes
Ordered by seconds saved per hour of work. Images and scripts almost always sit at the top, and the exotic optimizations almost always sit at the bottom.
Fix and verify
One change at a time, measured after each, so we know which change did the work rather than guessing.
Report against thresholds
Pass or fail on the three published numbers, with the before and after. No vanity score.
Where we have done this
Alaska Table Makers
An image-heavy gallery site is the hardest performance case there is. A headless static build keeps it instant anyway.
View project
Attu AI
Custom interactions and a dark, detailed identity, on a build that still loads immediately.
View project
AK AI Guy
Static build with a custom CMS, so publishing new work does not cost speed.
View projectCommon questions
Speed optimization, answered
What are Core Web Vitals?
Three metrics Google publishes thresholds for. Largest Contentful Paint should occur within 2.5 seconds, Interaction to Next Paint should be 200 milliseconds or less, and Cumulative Layout Shift should be 0.1 or under. Interaction to Next Paint replaced First Input Delay and became stable in 2024. Source: web.dev.
Does site speed actually affect rankings?
It is one signal among many, and it is not the strongest. Its bigger effect is on conversion, because slow pages get abandoned before they can rank for anything. Optimize it for the customers first and treat the ranking benefit as a bonus.
Why is my page builder site slow?
Because it ships a large amount of JavaScript to render things that could be plain HTML, and then plugins add more. There is a ceiling you cannot optimize past. We will tell you where that ceiling is before you spend money trying to beat it.
Can you speed up a site you did not build?
Usually, yes. We audit it, tell you what is achievable and what it costs, and you decide. If the honest answer is that a rebuild costs less than the tuning, we will say that even though the tuning is the easier sale.
How fast should my site be?
Under the published thresholds on a mid-range phone on a middling connection, not on your laptop on office wifi. That second test is the one that flatters everyone and tells you nothing.