New stores: 14 days of Protector freeFirst 25 stores.Uptime, daily scans, Slack and Telegram alerts. No card.See what’s included
Store Monitoring

Signs your store is losing sales — and where each one hides.

When revenue softens, the instinct is to blame demand: the market, the season, the ads. Sometimes that is right. Often something is broken in a way that produces no error message, and the store has been quietly converting worse for a fortnight. These are the signals worth taking seriously, ordered from fastest-acting to slowest, with the place each one actually hides.

Free plan, no credit cardRead-only connectionAlerts to email, Telegram, Slack or Discord

Traffic is steady but orders are not

This is the highest-signal pattern in ecommerce and the one most likely to mean something is broken rather than demand-driven. Demand changes usually move traffic and orders together. A gap between them points at the funnel.

The usual causes, in order of likelihood: a payment gateway declining or timing out, a shipping-rate configuration that returns no rates for a region, a discount code that errors, an app script failing on the cart page, a required field that is broken on mobile, or a tax or address validation rule rejecting valid input.

Orders stop entirely for a period that normally has them

Zero orders during a window that historically has orders is the strongest single signal there is, and it is unambiguous in a way that percentage drops are not. It is also the easiest to miss overnight, when it can run for eight hours before anyone opens a dashboard.

It rarely means demand vanished. It usually means checkout is broken, the store is unreachable for some customers, or a payment provider is rejecting everything.

Margin is falling faster than revenue

If revenue holds but margin slips, look at prices before you look at costs. An expired sale that never came down, a repricing app applying an old rule, a bulk edit on the wrong collection, or a currency rounding drift will all sell perfectly well — that is what makes them expensive.

Your best sellers are out of stock, and still being advertised

A sold-out product does not stop receiving traffic. Ads keep running, search results keep ranking, and the landing page keeps converting at zero. This is a signal that shows up in ad spend before it shows up in revenue reports.

Mobile conversion drifts away from desktop

Most storefront traffic is mobile and most owners browse their store on a laptop. A layout regression, a slow-loading hero image, a cumulative layout shift that moves the buy button, or an app banner covering the header will all hurt mobile only — and will be invisible in the way you personally look at your store.

Core Web Vitals are the measurable proxy: LCP for how quickly the page becomes useful, CLS for whether things move under a thumb, INP for whether taps feel responsive.

Organic traffic is sliding without a ranking crash

Gradual organic decline is usually structural rather than algorithmic: title tags overwritten by a theme update, canonicals pointing at the wrong URL, a noindex left on after a staging deploy, product schema removed by an app change, or a sitemap that stopped updating.

The tell is that it starts on a specific date. Demand-driven decline is gradual from the beginning; a technical cause has an edge you can find in a graph.

Customers mention problems you have never seen

One support message about a broken page usually represents many silent visitors who simply left. Treat “it wouldn’t let me check out” as a monitoring finding, not an anecdote — and note the device, browser and country, because it is likely to be specific to one of them.

What Cassian™ actually does

Specific checks, with the plan they run on.

Cassian™ compares order volume against a weighted historical baseline and raises a critical alert for a period that should have orders but has none.

Every plan, including free.

Significant-drop detection flags a period well below its expected volume, and gradual-decline detection catches a multi-hour slide that no single reading would reveal.

Significant drop on Analyst and above. Gradual decline on Protector and above.

Price changes are diffed daily against the previous catalog snapshot, with zero prices and inverted sale prices treated as critical.

Every plan with a connected store.

Core Web Vitals are measured daily for mobile and desktop separately, with 30-day trends — so a regression shows as a step change on a date.

Every plan, including free.

Missing titles, missing meta descriptions, broken heading hierarchy, missing canonicals and missing or incomplete product schema are all detected on every scan.

Every plan.

Broken internal links and missing product images are validated by HTTP request, not inferred — with blocked domains classified separately so they never count against you.

Every plan.

FAQ

Common questions

How do I know if my Shopify store is losing sales to a technical problem?
Look for a gap between traffic and orders. If sessions are steady and orders are down, the cause is almost always in the funnel rather than in demand — a payment gateway declining, shipping rates failing to return for a region, a discount code erroring, or an app script breaking the cart. Demand changes usually move traffic and orders together. The other strong signal is a period with zero orders that historically has them; that is rarely demand and almost always breakage.
What is the fastest signal that something is wrong?
Zero orders in a window that normally has them. It is unambiguous, it needs no interpretation, and it precedes every other symptom — support messages, review damage, ad waste. It is also the check Cassian™ runs on every plan including the free one, for exactly that reason.
My revenue is down but nothing looks broken. What should I check?
In this order: prices, for an expired sale or a bulk edit that changed more than you meant; stock, for a bestseller that sold out while still being advertised; mobile Core Web Vitals, for a layout or speed regression that only affects phones; and the date the decline started, because a technical cause has a sharp edge in the graph and a demand cause does not.

Imagine if your store could tell you when something's wrong.

Free scan, results in minutes. Zero-order alerts on every plan, including the free one.