Site speed and conversions — how fast a service site needs to be.

Speed is the one website problem with no upside. A slow page never converted better than a fast one. It costs you traffic — Google has used page experience as a ranking input for years — and it costs you the leads who do arrive, because the person searching “emergency [your trade] near me” at 9pm on a phone with two bars has roughly the patience of a wasp. The good news: it’s also the most fixable problem on this list. You don’t need a new business model; you need someone who knows where the weight is.

What “fast enough” means — in plain English

Google measures real-world loading experience with three numbers it calls Core Web Vitals. Strip the jargon and they’re asking three sensible questions.

• LCP — Largest Contentful Paint. How long until the biggest thing on the screen — usually your hero image or headline — actually shows up? Target: under about 2.5 seconds. Past that, the prospect is staring at a blank rectangle wondering if they fat-fingered the URL.
• INP — Interaction to Next Paint. When someone taps a button or opens a menu, how long until the page responds? Target: under about 200 milliseconds. This is the one that makes a site feel “janky” — taps that don’t register, menus that lurch open a beat late.
• CLS — Cumulative Layout Shift. Does the page jump around as it loads — an ad slot popping in, a font swapping, a banner shoving everything down just as they go to tap “Call”? Target: under about 0.1. High CLS is how a prospect ends up tapping your cookie notice instead of your phone number.

A service site that meets all three on a real phone on a real connection is, by definition, fast enough. There’s no prize for being faster — chasing a perfect score past the point of diminishing returns is a different kind of vanity. But hitting the thresholds isn’t optional; it’s the floor.

Slow loses the mobile lead first

Most local-service searches happen on phones — someone whose AC just died, whose roof is leaking, who needs a lawyer by Monday. They’re not at a desk and they’re not “researching”; they’re in a hurry, with three tabs open on three of your competitors. A two-second difference in load time isn’t a footnote there — it’s the difference between the page that loaded and the two that didn’t. Mobile is also where the penalty compounds: phones have slower processors and flakier connections, so a heavy page that’s merely sluggish on desktop is genuinely broken on a mid-range Android over 4G. And because Google indexes the mobile version of your site, a slow mobile experience drags the rankings that bring you the traffic in the first place. The leak is at both ends of the funnel — which is why we treat mobile performance as a launch requirement, not a cleanup task. (More on that on does mobile design really matter.)

Nobody waits for a slow site to prove it’s worth waiting for. They just leave, and they don’t tell you why.

The usual culprits on a service site

It’s almost always the same five things, in roughly this order of damage:

• Enormous hero images. A 4MB photo of a truck, dropped in at full camera resolution and scaled down in the browser. The phone still downloads all 4MB. This one move alone can wreck your LCP. Properly sized and compressed, that image is 80–200KB and nobody can tell the difference.
• A page builder doing the heavy lifting. The big visual builders generate layers of nested markup and ship a pile of JavaScript and CSS to render a layout that plain HTML would handle for free. Convenient to drag around in; expensive on every single page load, forever.
• Plugin bloat. Twenty-six active plugins, each adding its own scripts and styles to every page whether that page uses the feature or not. The slider plugin loads on the contact page. The form plugin loads on the blog. It adds up to a tax on every visitor.
• No caching, no CDN. The server rebuilds the whole page from the database on every request instead of serving a saved copy, and every visitor — wherever they are — pulls assets from one origin server. Caching and a content-delivery network are basic hygiene; their absence is a self-inflicted wound.
• Render-blocking everything. Fonts, third-party widgets, chat bubbles, tracking scripts and analytics all loading before the page can paint — so the prospect waits on a live-chat tool they’ll never use before they can read your headline.

None of these are mysteries. They’re the predictable cost of building a site by accretion — a plugin here, a builder there, an image dropped in at full size because it looked fine on the desktop preview. A site built lean from the start doesn’t accumulate them; a heavy one can usually be put on a diet, but past a certain point the cheapest fix is a rebuild on a foundation that isn’t fighting you — which is the honest question on new website, or just fixes.

How fast we ship — and why it stays that way

Every site we build hits Lighthouse 95+ on performance, on every page, as a non-negotiable. That’s not a setting you flip at the end; it’s a consequence of how the thing is built — a hand-written custom theme instead of a page builder, images sized and compressed before they ever go live, only the scripts a page actually needs, caching and a CDN on by day one. Same standard behind the verified outcomes on the Tampa web-design case and every build on the web design service: a 14-day build, from $3,000, that’s fast because it was never allowed to get slow. The harder part is staying there — a fast site that grows by accretion gets slow again — so the patterns your team uses to add pages keep images right-sized and markup lean, and if you want someone watching performance over time, that’s what the care plan is for. Speed is a habit, not a one-off.

What to measure — and what to ignore

Measure: your Core Web Vitals on real mobile devices (Search Console’s Core Web Vitals report shows what real visitors actually experience — the only data point that matters). Watch LCP first; it’s usually the one that’s broken and the one with the biggest payoff. Re-check after any change that adds weight — a new plugin, an embedded video, a “quick” tweak from a contractor.

Ignore: chasing the last few Lighthouse points past 95 when nothing the visitor experiences improves; obsessing over a synthetic lab score while never testing on an actual phone; “page speed” tools that grade your homepage and nothing else (your service pages and contact page are where the lead converts — they need to be fast too). And ignore anyone who tells you speed alone fixes lead flow. It removes a ceiling. What you build under that ceiling is the part that converts — message-match copy (conversion copywriting) and proof that earns the call (trust signals) — and that’s where the rest of this cluster goes next.