A prospect finds you on their phone, in a parking lot, with two bars of signal. Your homepage starts to load. The logo appears. The hero image is still coming. They wait.
Then they don’t.
That’s the whole story of most website performance problems, and it happens thousands of times a year on sites whose owners have no idea it’s happening. The site looks fine on the office wifi. It looks fine on the designer’s monitor. It falls apart on a phone in a parking lot, which is where a meaningful share of local searches actually happen.
Speed is a design decision, not a technical one
The instinct is to treat this as something for a developer to fix after the fact. Compress a few images, add a caching plugin, call it done.
But most of the weight on a slow site was put there on purpose. The full-bleed video header. The four web fonts. The animated counters. The chat widget, the popup, the tracking pixels for platforms you stopped advertising on in 2022. Every one of those was a decision somebody made, and most of them were made without anyone asking what it would cost.
So the first question isn’t how do we make this faster. It’s what is all of this for.
We have watched clients cut a page’s load time nearly in half by removing things nobody could justify keeping. No engineering. Just a willingness to look at a homepage honestly and admit that the auto-playing video was there because a competitor had one.
What actually matters
Google measures three things, and they map cleanly onto what a visitor experiences:
How fast does the main content appear? Largest Contentful Paint. Usually your hero image or headline. Target is under 2.5 seconds. This is almost always an image problem — an uncompressed JPEG two thousand pixels wider than it needs to be.
How fast does the page respond when tapped? Interaction to Next Paint. Target is under 200 milliseconds. This is almost always a script problem — too much JavaScript competing to run at once.
Does the layout jump around while loading? Cumulative Layout Shift. Target is under 0.1. This is what makes someone tap the wrong button because an ad loaded and pushed everything down. It’s the most annoying of the three and the easiest to fix.
Run your site through Google’s PageSpeed Insights. Use the mobile score, not desktop. The desktop score flatters you and nobody is looking at your site on a desktop in a parking lot.
The fixes, in the order we’d do them
Images first. Serve WebP or AVIF instead of JPEG and PNG. Size them to their actual display dimensions. Lazy-load anything below the fold. On a typical small business site this is more than half the problem and it requires no code.
Fonts second. Every weight of every font family is a separate download. If you’re loading six weights and using three, delete three. Self-host or preload the ones you keep so text doesn’t flash invisible while they arrive.
Scripts third. Audit every tracking pixel, chat widget, and plugin. Anything you can’t name a current business reason for, remove. This is tedious and it is worth it.
Then caching and a CDN. Page caching plus a content delivery network so a visitor in Raleigh isn’t waiting on a server somewhere else. This is the step most people do first, and it’s genuinely the least important of the four — it makes a fast site faster, but it can’t rescue a page carrying eight megabytes of images.
The part that isn’t about speed
Here’s what gets lost in the metrics conversation: a fast page that doesn’t tell someone what you do or how to reach you is still a failure. It just fails faster.
When we rebuild a site, load time is one input. The others are whether the headline says something specific, whether the phone number is visible without scrolling, whether the contact form asks for three fields instead of nine, and whether someone landing on a service page can tell within a few seconds that they’re in the right place.
Speed buys you the chance to make that impression. It doesn’t make it for you.
Where to start
Pull up your site on your own phone, on cellular, away from your office. Count the seconds. Then try to complete the thing you most want a customer to do — call, submit a form, find your hours.
Whatever frustrated you in that exercise is your list. Most of the time you don’t need an audit to find the problem. You need to stop testing your website under conditions your customers never experience.
219 Group builds websites for clients across North Carolina — hospitality, professional services, higher education, consumer products, and non-profits. If your site feels slower than it should, let’s talk about it.
Our lifelong relationship begins here.
Tell us what you're trying to grow. Send us a note and someone from our team will get back to you within 24 hours.
Start the Conversation →Or call 1.910.483.4749 · Existing client? Submit a support ticket