Is anyone actually looking after your store?
Most online stores don’t break on the day they’re built. They rot quietly — a year of skipped updates, a backup nobody ever restored, a checkout that got half a second slower every quarter. Answer 18 questions about your store and find out how much risk has piled up, what it’s plausibly costing you, and what to fix first.
- ⏱ About 2 minutes
- 🔒 Runs entirely in your browser
- 📋 18 checks · 5 areas
- 🧾 Printable fix list
- 💸 Free, no email required
Your store, as you describe it
—
What this plausibly costs, in money
Read this as an order of magnitude, not a forecast. It is arithmetic on the band you selected and published industry averages — it knows nothing about your margins, your customers or your market. The speed figure is anchored on the widely cited Deloitte and Google Milliseconds Make Millions study, which found that a 0.1 s mobile improvement was associated with roughly 8.4% more retail conversions. That was measured on already-slow sites and it emphatically does not continue in a straight line — you cannot save three seconds and multiply. So this uses a diminishing-returns curve that rises steeply for the first second and flattens towards a ceiling of 20%, and it shows a range rather than a single number because the honest answer has that much spread in it.
What to fix, in order
Highest consequence first. Each item says why it matters and what the actual fix is.
Want someone to look at the code instead of your answers?
This page can only score what you already know. A real audit reads the database, the plugin list, the server config and the error logs — and finds the things nobody knew to tell you about. SwiftForge Web, who built this tool, do that part: they take a copy of the store, go through it, and send a written report with fixed prices. It costs nothing and there is no obligation.
Running a smaller store? There is a cheaper way to stop this happening again.
Most of the findings above are not one-off repairs — they are things that need somebody to keep an eye on them: updates applied, backups taken and tested, the store checked after each change. Agency retainers are priced for stores turning over serious money, and if yours isn’t there yet, that maths never works.
GrannyHost is SwiftForge’s answer to exactly that. Hosting plus the daily care list for small WooCommerce and PrestaShop stores — monitoring, updates, backups, SSL, plugin maintenance — at a price built for a small shop rather than an agency budget. Same servers and same people as SwiftForge, run as a separate low-cost brand.
How the score works
Five areas, weighted by how much damage each one does when it fails. Within an area every question counts equally, and an “I don’t know” scores as poor rather than neutral — because an unknown you haven’t checked behaves exactly like a problem you haven’t found.
| Area | Weight | Why that weight |
|---|---|---|
| Software currency | 25% | Unsupported software is the root cause behind most “the store suddenly broke” and nearly all compromises. It also blocks every other fix. |
| Recoverability | 20% | Determines whether a bad day costs you an afternoon or your business. An untested backup counts for very little. |
| Security & access | 20% | Abandoned plugins and stale admin accounts are what automated scanners actually find. |
| Mobile speed | 20% | The only area here that costs you money every single day rather than on the day it fails. |
| Oversight | 15% | Decides how long any of the above runs unnoticed. Low weight, high multiplier on everything else. |
The bands
| Score | Band | What it usually means |
|---|---|---|
| 85–100 | Clean | Actively maintained. Keep doing what you’re doing, and test a restore. |
| 70–84 | Watching | Fundamentally sound with specific gaps. Usually a few hours of work. |
| 50–69 | Drifting | Nothing is on fire but nobody is steering. This is where most stores sit. |
| 30–49 | Neglected | Several compounding problems. Upgrades are now risky because nothing is tested. |
| 0–29 | Running on luck | Still trading, but one plugin update or one bot away from a bad week. |
What this is, and who pays for it
Full disclosure. Store Sheriff is built and paid for by SwiftForge Web LLC, an engineering team that rescues and maintains neglected WooCommerce and PrestaShop stores. They have a commercial interest in you discovering that your store needs work. That is exactly why this tool shows you its weightings, its formulas and its caps — so you can check its reasoning rather than take its word.
It exists because the first conversation with a store owner is almost always the same one. The store makes money. It was built years ago by someone who has moved on. Nobody can say what version it runs, when it was last backed up, or whether the backup works. The owner isn’t careless — nobody ever told them what to keep an eye on.
So this is the list. It is free, it needs no email address, and it has no upsell gate: the full result, every finding and every fix is on this page whether or not you ever talk to anyone.
What it does not do
- It does not inspect your store. It scores your answers. If an answer is wrong, the score is wrong.
- It is not a security audit or a certification, and it deliberately gives you no badge to display — a self-assessment is not something your customers should be asked to trust.
- It does not store anything. There is no database, no account, no analytics on this page and no form submission. The scoring is JavaScript running on your own device. Closing the tab erases it. The “copy a link” button encodes your answers into the link itself, so only whoever you send it to can read it.
- It does not ask for credentials, and you should be suspicious of any free tool that does.
If you only do three things
Most of the value in this list comes from a small number of fixes. In rough order of return:
- Restore a backup somewhere that isn’t your live store. Not “check that backups run” — actually restore one. This is free, takes an afternoon, and converts your biggest unknown into a known. Most owners who do this discover something broken.
- Get off end-of-life PHP. It is usually a dropdown in your hosting panel, but only after the store is on versions that can handle it — which is the part that needs care. See what end-of-life PHP means for a store.
- Delete every plugin and admin account nobody can justify. The cheapest security work there is. Deactivating is not enough — inactive plugin files are still reachable and still exploitable.
After that it stops being a list of tasks and becomes a habit. SwiftForge publish the schedule they actually use as a daily, weekly, monthly and quarterly checklist — worth copying whether you ever hire anyone or not.
Questions people ask
Is this really free, and what’s the catch?
Genuinely free, no email required, full result on the page. The catch, stated plainly: it is built by an agency that sells store repair work, and a low score is good for their business. Judge it on whether the reasoning holds up — the weightings, formulas and caps are all published above.
Does my data go anywhere?
No. There is no form submission, no database, no account and no analytics script on this page. Scoring happens in JavaScript on your device. The address you type for the PageSpeed button is used only to build a link to Google’s own tool and is never sent to us.
How accurate can a questionnaire be?
It is as accurate as your answers, which is its real limitation. It cannot find the things you don’t know about — an abandoned plugin with a known vulnerability, a database that has grown to 40 GB of expired session rows, a cron job that silently died. In practice a questionnaire like this catches the common, expensive problems and misses the specific ones. That’s what reading the actual code is for.
Why does “I don’t know” lower my score?
Because an unverified assumption and a real problem behave identically until the day you need them not to. If nobody can say whether backups work, you do not have working backups — you have a hope. Treating unknowns as neutral would make the score comfortable and useless.
Does it work for Shopify or BigCommerce?
Partly. The hosted platform handles your software versions, PHP and server patching, so those questions stop applying and the score adjusts. Speed, apps, admin access, oversight and the “can you leave” question all still very much apply.
My score is bad. How urgent is it?
Unsupported software plus no tested backup is the combination worth acting on this month, because the fix gets harder and more expensive the longer it waits — the further behind you fall, the more versions you have to jump at once. A slow store is costing you money daily but it is not an emergency. Everything else can be scheduled.
Can I use this with my clients?
Yes. Freelancers and agencies are welcome to run it with store owners, print the result and use it as the basis of a quote — including quotes that compete with SwiftForge’s. Link to it rather than copying it and you’ll always have the current version.