On our checker, 85 and above reads "In good shape" and 70 to 84 "Some fixes needed". On any test, though, the number matters less than which findings sit behind it and whether it is going up. Some well-known shopping sites can score poorly on public speed tests and still take orders all day.
So before you panic about a 54, or relax about an 88, it helps to know what the test was looking at.
Why Big Websites Can Score Badly
Run a speed test on a household-name retailer and you may well see orange or red. There are ordinary reasons for it.
Big sites carry a lot of extra code. Advertising tags, analytics, chat widgets, personalised recommendations, video and pop-ups all have to load. Each has a business reason, and a phone has to download and run every one of them.
The test is a cold first visit on a slow phone. Google's speed test loads your page as a mid-range phone on a throttled mobile connection, with nothing saved from a previous visit. A regular customer of a big retailer may have the site's files saved on their phone already, and may shop through an app instead.
People already know them. A household name gets customers who type the name straight into the browser, come back week after week, and will put up with a slower page because they have already decided to buy there. The test has no way to measure that loyalty.
Automated accessibility tests only see the page code. A large site with thousands of pages and years of changes collects missing labels and low-contrast text faster than anyone can tidy them up, and the test reports every one it finds.
The score is still a fair measure of a first visit on a slow phone. The business is succeeding for reasons the test does not look at.
Why The Same Score Matters More For A Flower Shop
Your regular customers know you, but a first-time visitor still needs a clear route to order.
Think about who lands on your homepage. Someone has searched "florist near me" on their phone at lunchtime because their mum is in hospital, or they need a sympathy arrangement at a funeral on Friday. They do not know your name yet, and they may well have a few other shops open in other tabs. If your page takes a long time to show the bouquets, or the phone number is not tappable, they try the next shop in their search results.
An established retailer may have returning customers who already intend to buy there. A first-time customer choosing between florists has no habit to bring them back. For your shop, read the findings alongside your own online orders and customer feedback.
What A Website Score Measures
Most website tests, ours included, look at a handful of things a machine can check reliably:
- Speed on a slow phone. How long until the main image or heading appears, whether the page jumps about as it loads, and how long the page is too busy to respond to a tap.
- Accessibility. Whether someone using a screen reader, a keyboard or a zoomed-in screen can use the page. Automated tests catch common problems such as images with no description, buttons with no name and text too faint to read. They cannot catch everything, so a person still needs to try the site.
- Steps to order. How many pages, taps and boxes sit between a customer and a paid order, and whether they can see the delivery charge and total before they hand over card details. Our guide to what a good florist checkout looks like on a phone goes through those steps one by one.
- Security and search basics. Whether your site opens securely (an https:// address with no browser warning), whether Google is allowed to read the page, and whether link previews and local listings can pick up your shop details.
These are all things you or your website provider can change.
What A Website Score Does Not Measure
A website test only looks at your website, so most of what makes your shop busy is outside its view.
- Your brand. How well known you are in your town, and whether people ask for you by name.
- Footfall. Walk-in trade, the customer who orders at the counter after seeing your window, and the regulars who ring every Friday.
- Your reputation. Your Google reviews, word of mouth, and the funeral directors and venues who recommend you.
- Your flowers. The quality of your stems, your design style and how the bouquet looks when it arrives.
- How you answer the phone. Many shops still take sympathy and hospital orders by phone, and no website test hears those calls.
Your website score does not measure every factor in local search results. Google's own guide to local ranking names relevance, distance and prominence. If reviews are where you are weakest, our guide on how to get more Google reviews for your flower shop covers when to ask and how to reply. Digital Florists can send automated review requests to past customers.
How To Read Any Website Score
Whichever test you use, ask these five questions before you act on the number.
1. What Device and Connection Was It Tested On?
A desktop score is usually far higher than a phone score, because Google's desktop test runs on a faster simulated connection. Many customers find a florist on their phone, and Google's test of a phone visit is the harder one, so the phone result is the one to watch.
2. Was It One Lab Visit or Real Customers?
A lab test is a single simulated visit, so it can move a few points between runs with nothing changed on your site. Our guide to why speed scores change between runs explains lab scores, real-visitor figures and which one to trust.
3. Which Pages Were Checked?
A homepage test says nothing about your checkout. Your homepage might load quickly while the delivery date picker reloads the whole page three times. Check which pages a test covered before you treat it as a verdict on the whole site.
4. What Is the Score Made Of?
On many tests an overall score is an average of many smaller results, and averages hide things. A site can score well overall while one broken step stops orders, such as a delivery charge that only shows after the card details. (Our checker holds the overall score down when it finds a few of the worst problems, see below.) Read the individual findings and look for the ones that stop an order outright.
5. What Are You Comparing It With?
Compare your score with your own last result, not with a household name or the florist in the next town. Their customers and budget are different from yours. For example, a score that goes from 58 to 71 after you fix your images and add a tap-to-call link is a real improvement.
A Worked Example
These figures are illustrative, for two fictional shops.
Shop A scores 62 overall. Its homepage carries several large uncompressed photos, so the phone score is weak. But its prices are on every product, the phone number is tappable, the same-day order-by time sits next to the order button, and a test add to basket reaches checkout in three steps.
Shop B scores 74 overall. Its homepage is light and quick, but there are no prices on the product pages, the order-by time is buried in the delivery terms, and the delivery charge only appears after the customer has typed in their card details.
Shop B has the higher number. Shop A is probably the easier site to order from, and its fix is shrinking a few photos (our guide to website speed scores for florists covers exporting web-sized copies of large photos). Shop B's problems take longer to fix and are more likely to lose an order.
How Our Website Checker Scores A Flower Shop Site
We built the Florist Website Checker to score the things that lose a florist orders, and to say plainly where each result came from.
It scores seven areas: Florist Essentials, Ordering & Checkout, Speed & Mobile, Search & Local, Trust & Security, Design & Content and Accessibility. The overall score is weighted most towards Florist Essentials and Ordering & Checkout, because they decide whether a customer orders. How we check shows the weight of each area.
The quick check scores most of these from your homepage. Ordering & Checkout needs the full report, because one page cannot show it.
Most results are measured. The checker loads your pages in an ordinary browser on a computer screen and a phone screen, reads your page code and server replies, and uses Google's speed test. Google's mobile performance score is shown in the report using Google's own bands (90 and up is good, under 50 is poor), but it is not counted again, because its main parts (how soon the main content appears, whether the page jumps, how soon it answers a tap) are already scored as separate checks.
The AI review is labelled. On a full report, an AI reads your pages and screenshots and adds a second opinion on things like your photos and wording. Each of those findings is marked as an AI opinion, and counts for less than a measured result.
Some problems hold the whole score down. A few serious measured problems keep the overall score down until they are fixed, however well the rest does. Examples are a main address that is not secure, a site Google is blocked from reading, or a test order that finds no way to buy a bouquet at all.
Scores come in four bands. 85 and up reads "In good shape", 70 to 84 "Some fixes needed", 50 to 69 "Needs work", and under 50 "Needs urgent work".
The free quick check looks at your homepage and needs no account. Sign in and confirm the site is yours, and the full report covers more pages, marks problems on screenshots of your own site, and follows a test order towards your checkout. It may add one product to a basket, and it stops before any real personal or payment details.
When Part Of Your Site Could Not Be Tested
Read the note beside your score. Sometimes we can't test everything, and the report says so plainly. Anything we couldn't test doesn't count for or against you.
"Ordering not tested" means our test order couldn't get through your checkout, so your score doesn't tell you whether customers can. The report gives the reason, such as the product we picked being sold out. Your other areas still count.
"Checkout partly tested" means our test order got past the product page but stopped before the page where customers type their details. The report shows where it stopped. The steps it reached still count.
If we couldn't test a big part of your site, you still see the number, but as a rough guide without a verdict such as "In good shape". If we could only score a few areas, we hold back the overall number and show the areas we could score.
"Couldn't reach your site, try again later" means we couldn't read enough of your site to score it at all. That check gets no score and doesn't appear in your score history, so it can't drag your trend down.
What To Fix First
Work from the problems that stop an order towards the ones that slow it down:
- Anything that blocks ordering. A basket that will not accept a product, a checkout that errors, a page Google cannot read.
- Missing florist essentials. Prices, a tap-to-call number, your delivery area and the same-day order-by time.
- Steps and surprises at checkout. Forced account creation, and a delivery charge that only appears at the end. Our phone checkout guide for florists shows what to look for.
- Speed on a phone. Usually large images first, then extra scripts and widgets you no longer use.
- Accessibility gaps. Missing image descriptions, unlabelled form boxes and faint text.
Then run the check again and compare with your last result.
Common Questions
What is a good website score for a florist?
On our checker, 85 and above reads "In good shape" and 70 to 84 "Some fixes needed". On any test, the findings behind the number matter more, and the score to aim for is one that rises over time on the phone results and on the parts that affect ordering.
Why does a well-known retailer score lower than my small site?
Large sites tend to load more advertising, tracking and personalisation code, and the test measures a first visit on a slow phone. Their customers often know them already and come back regardless, which the test cannot measure.
Does a low Google speed test score hurt my Google ranking?
Page experience is one of many things Google looks at, and it is unlikely to outweigh relevance or, for map results, your Google Business Profile, reviews and distance. A slow phone page is still worth fixing because it can lose customers who are ready to order.
Why does my score change when I run the test twice?
A lab speed test simulates one visit, and conditions vary from run to run. Our speed scores guide explains why and which number to watch.
Does a clean accessibility result mean my site is accessible?
No. Automated tests find common problems, but some issues only show up when a person uses the site with a screen reader or keyboard. To find what the tests miss, ask someone who uses assistive technology, or an accessibility specialist, to try ordering from your site.
