A tenth of a second

Most advice given to small stores is unfalsifiable. Post more. Tell your story. Find your audience. You cannot test any of it, which is convenient for the people selling it.

Site speed is the exception. It has been measured properly, on real stores, at scale, and the finding is specific enough to be uncomfortable.

What the study actually found

Google commissioned Deloitte and the analytics firm 55 to answer one question. Does mobile site speed move revenue. They instrumented 37 European and American brand sites and collected data on over 30 million user sessions, monitoring load times hour by hour for thirty days, then compared what happened when the site was fractionally faster against when it was not.

The threshold they tested was 0.1 seconds. One tenth of one second, across every page in the journey. For retail sites, that improvement moved conversion rates up 8.4 percent and average order value up 9.2 percent. Shoppers did not just buy more often. They spent more when they did.

The funnel detail is the part worth sitting with. Progression from a product page to the basket rose 9.1 percent. Product listing to product detail rose 3.2 percent. Checkout to completed order rose 4.7 percent. The gain is not concentrated in one broken step. A tenth of a second lifted every stage of the journey, which tells you what speed actually is. Not a technical metric. A tax charged at every point where a person decides whether to continue.

Most stores are on the wrong side of it

This would be a footnote if the web were mostly fast. It is not. Using Chrome's own field data from real visits, DebugBear reports that only 49.7 percent of mobile websites pass Google's Core Web Vitals assessment. Desktop does better at 57.1 percent, and desktop is not where your customers are.

Roughly half of mobile sites fail a bar set at the seventy-fifth percentile of visits, which means even the passing sites are serving a slow experience to a quarter of their visitors. If you have never checked, the base rate says it is a coin flip whether your store is one of them.

How a fast store becomes a slow one

Nobody builds a slow store on purpose. It accumulates. The theme starts clean. Then a review widget goes on, because social proof matters. Then a chat bubble, a popup for the email list, a currency switcher, a countdown timer for a sale that ended in March, and three tracking pixels for ad platforms, two of which you no longer run. Every one of them was added for a reason that was good at the time.

Then the hero image, exported straight from a phone at full resolution because it looked sharp. Then the product photos, forty of them, same story. None of this feels like damage on a laptop with fast internet, which is where the owner always looks at it. On a phone with two bars in a car park, which is where a real customer opens it, the difference is the difference between a sale and a back button.

Speed is not a technical metric. It is a tax charged at every point where a person decides whether to continue.

Nobody audits after launch week

The reason slowness persists is the same reason most small-store problems persist. It is not hard to fix, it is just nobody's recurring job. There is no morning where checking load times is the thing you are supposed to do. It gets checked when the store is built, and then never again, while apps and images pile up for years.

Site health inside Vantage is meant to be that recurring job. The slow page, the broken link, the checkout error, caught while it is small and surfaced as something specific you can approve a fix for, rather than a score out of a hundred that means nothing on a Tuesday.

Deloitte's number is a tenth of a second. Almost no small store is arguing over a tenth of a second. They are carrying two or three whole ones they do not know about, and paying for them at every step of every visit.