Restaurants Come Second on Phones. Then They Make You Wait.

“It looks fine on my phone.”

The restaurant owner had said it twice by then, and both times he was right. I had his homepage open on my own phone while we talked, and it did look fine.

The text was big enough to read without pinching at it. The buttons sat far enough apart that his customer’s thumb would land on Reserve instead of catching Order Delivery on the way past, which is the whole of what a mobile check is asking about.

“So we’re fine.”

“You’re fine at that part. You’re better than fine, actually.”

85.3% of restaurant sites pass Google’s mobile checks, second of the 47 industries I graded, against 78% of everything I audited.

“Then what am I paying you to look at?”

“How long did it take to show up?”

The question he couldn’t answer

“It’s instant.”

“On your wifi. On the laptop that has opened it a thousand times.”

Silence, which I’ve learned to enjoy.

What his customers were looking at, during the gap he couldn’t put a number on, was a white rectangle with a logo floating in the middle of it. The menu button existed underneath a photograph that hadn’t finished arriving. Tapping it accomplished nothing, which a hungry stranger reads as a broken restaurant rather than a slow website.

Fitting a screen and arriving quickly are two entirely separate things, and owners confuse them constantly. Your industry sorted out the first one and never got told about the second.

A check can tell you the layout fits a hand. Whether the thing is any good to use is a separate question, and nobody has ever automated that one.

Restaurant sites average 55 out of 100 for mobile speed, 34th of the same 47, where every small business I tested averages 58 out of 100.

Second at fitting the telephone. Thirty-fourth at filling it.

“That seems harsh.”

I put 1,899 of these websites through a full speed test on a simulated phone, so it isn’t one bad afternoon on my laptop. Google doesn’t call a performance score good until it reaches 90, for whatever comfort that gives anybody.

What I read back to him

“Fine. What’s actually on it?”

I still had the page open, so I read him his own homepage.

A video of the kitchen playing quietly behind the logo. Eleven photographs of the dining room, each one still the size the camera saved it at, shrunk down to thumbnail dimensions by the browser only after it has downloaded every last pixel.

Three different fonts. A reservation widget, a chat bubble, a delivery pop-up, and a tracking pixel from an advertising campaign that finished in 2023.

“Someone told me to add every one of those.”

“I know. That’s why they’re there.”

Not one of them was a stupid decision on its own. Someone recommended each one, and each one made perfect sense on the afternoon it went on.

I’ve never seen a restaurant website turn slow from a single mistake. They turn slow from two years of reasonable choices stacked on top of one another.

“So I delete the video.”

“You can, and it will help a little. It won’t help nearly as much as you’re hoping, because the video is the thing you notice every time you visit, and the photographs are the thing you stopped noticing about eighteen months ago.”

Med spas buy the same trouble deliberately, and the expensive look they paid for is the last thing on the page to turn up.

”Nobody has complained”

Nobody will, and that’s what makes this one expensive.

A table of four deciding where to eat has three other tabs open and about ninety seconds to settle the argument. If your gorgeous homepage is still assembling itself while they glance down, they eat somewhere else.

No angry email. No complaint. You get a quiet Tuesday nobody can explain.

Your trade gets talked about more than almost any other. Talking back is the part your homepages can’t do, so the quiet stays quiet.

The distinction matters more than owners expect. A customer who decides your restaurant is expensive at least remembers your restaurant, and might argue about it with a friend later. A customer who decides your website is broken remembers nothing about you whatsoever.

“A slow page never stopped me buying anything.”

“You were buying a water heater. Nobody browses restaurants.”

Hunger is the least patient buying mood there is. A guy hunting for a lawyer will sit through a slow page, because he isn’t hiring five lawyers and he knows the decision matters. A family in a car with the engine running gives you the length of one traffic light.

That’s the trade you’re in. Your customers are always standing on a sidewalk, always a bit hungry, always holding a phone that’s doing four other things.

The thing he’d never done

“Have you ever watched it load on someone else’s phone?”

A long pause on the line.

“No.”

Almost nobody has. Your laptop has met that homepage a thousand times and keeps every picture from it saved and ready. The phone in a customer’s hand, two bars of signal, three blocks away, is meeting the whole thing cold.

So I sent him to his own parking lot.

Owners rarely believe this until they produce the number themselves, standing outside in the cold, watching a homepage behave in a way they had comfortably assumed was someone else’s department.

A whole city taught me the same thing. Calgary clears the mobile check at the national rate, and that turned out to be nowhere near the finish line.

Walk outside, off the restaurant wifi, onto your phone data. Open a private tab, because your everyday browser has been hoarding your own images for months and will lie to you about the wait.

Then count out loud until your opening hours appear on the screen, and be honest with yourself about the number you land on.

He got to nine. He rang me back that same afternoon, which no restaurant owner had ever done before and none has managed since, and his opening sentence was an apology to a customer he had never met.

Fix one thing

Compress the photographs, and have whoever built the site do it properly, at the size the page actually displays them.

That’s the sentence. Whatever the count was in your parking lot, the pictures are why, because with restaurants it’s nearly always the pictures.

If the pictures are already small and the page still crawls, the build underneath is the problem, and rebuilding restaurant sites takes up most of my week. Fast is a decision someone makes on day one, not a thing you bolt on afterward.

Then ignore the rest of this post. Ignore the fonts, the chat bubble, the pop-up, the tracking pixel from 2023, the score, the rank, the lot of it. I’ve watched owners try to repair nine things at once and finish none of them, and the other repairs are written up over here whenever you like, once the pictures are done.