A florist we work with in Doral had a website that took thirty-one seconds to load on a phone. Not thirty-one seconds to finish every animation. Thirty-one seconds before a customer could see anything at all. The shop was fine. The flowers were fine. The website, in every practical sense, was closed.

That's not an unusual number for a small retail business on a hosted platform, and it's the most expensive problem almost nobody measures.

The math

Google's own research puts the jump in bounce probability at over 100% when a mobile page goes from one second to six. Past ten seconds, most visitors leave before the page renders. At thirty seconds you're not competing with the shop down the street. You're competing with the back button, and you're losing.

Here's what made that Doral case worth writing down: nothing was wrong with the shop's design. The weight came from things nobody had looked at. A logo saved as a 2.6 MB PNG. A platform loading its own scripts. Images served at full resolution to a phone screen. After the work, the page weight went from 5,834 KB to 2,347 KB and load time from 31 seconds to 7.2. Still not fast. But the difference between "unusable" and "usable" is where the customers are.

The remaining seconds came from third-party scripts the platform loads and the shop can't remove. That's its own lesson about hosted storefronts.

Why local retail specifically

You're usually on someone else's platform. Florists, boutiques, bakeries, and studios mostly run on hosted systems that trade control for convenience. That's a reasonable trade, until you discover you can't delete the 280 KB of code the platform injects on every page load.

Your traffic is mobile and impatient. Someone ordering flowers is often doing it on a phone, in a hurry, for an occasion that's today. There's no second visit.

Your competition is a national site with an enormous budget. The order-by-6pm delivery giants load fast because a team works on it full time. You won't beat them on speed alone, but you don't need to. You need to be fast enough that the person who wanted the local shop stays with the local shop.

The playbook

1. Measure before you believe anything

Run your own site through Google PageSpeed Insights on mobile, not desktop. Desktop numbers flatter everyone. Write down the number. Most owners are genuinely shocked, because on their own phone, on their own wifi, with the site cached, it feels fine.

2. Find the heavy file. It's usually just one

On small-retail sites, the damage is rarely spread out. It's one enormous image: a logo exported at print resolution, a hero photograph straight from a camera, a banner nobody resized. Compress it, serve it in a modern format, and a big share of the problem disappears in an afternoon.

3. Fix the platform-shaped problems you can

You can't remove the platform's own scripts. You can control what you upload, how many apps and widgets you install, whether every product photograph is 4000 pixels wide, and whether three chat and popup tools are competing on load. Every widget has a cost, and most earn less than they cost.

4. Let the profile carry what the site can't

If the storefront is slow because of decisions you don't control, put more weight on the Google Business Profile: complete categories, product photographs, hours that survive the holidays, and reviews answered in the language they were written in. For a local shop, that profile is often a faster and better-converting front page than the website.

5. Then fix what people see first

Speed gets them to the page. The first screen decides the order, and it has to answer three questions: what you sell, whether you deliver where they are, and how fast. A florist whose homepage doesn't say the delivery cutoff is losing orders that were already won.

What good looks like

A mobile page under three seconds. Not perfect. Good enough that nobody leaves before seeing the product.

Images sized for phones. A product photograph doesn't need to be 4000 pixels wide to look right at 400.

A profile that outperforms the storefront. If the platform limits you, stop fighting it and win in the place you fully control.

Where to start

Measure. Then find your one heavy file. In our experience with small retail sites, those two steps recover most of what you can recover without touching the design at all.

If you want to know your own numbers, our free five-page audit measures them for you. 48 hours, no sales call.