Find the leaks before they cost you a quarter.
Most stores lose money in the same quiet way: something breaks on mobile, nobody notices for a month, and the only record of it is a flat quarter. Scout watches for it, ranks what is worth your next afternoon, and afterwards tells you what that work actually earned — including when the answer is nothing.
14 days free, then from $29 a month. One click to connect — no tag to paste and no developer. Reads your store, writes nothing to it.
Three things, and all of them are only worth anything if the number can be trusted.
Lose less to the things nobody noticed
A mobile checkout that breaks on a Tuesday costs a month of mobile orders before a monthly report shows it. Scout compares each day against the same weekday before it, so a real drop surfaces in days — and a quiet week stays quiet rather than generating noise you learn to ignore.
Spend the afternoon you have on the right thing
Most stores have more ideas than afternoons. Scout ranks what is worth doing by money at stake against the effort it takes, so the settings-page fix worth a few thousand a month outranks the redesign worth more and costing a quarter.
Know which of last quarter's work paid
Afterwards, Scout goes back and says what the change earned — and refuses to claim a number when something else shipped in the same window. That refusal is the point: it is what makes the figures it does report worth taking to whoever asked.
Which is why Scout says “this cannot be measured” as often as it says anything else. A tool that always has an answer is one you stop believing by the third month — and the months you believed it were the expensive ones.
Four steps, and analytics only does the first one.
Detect
Scout watches conversion, revenue, orders, AOV, add-to-cart, checkout, products and traffic — and tells you when one of them moves in a way it shouldn't.
Diagnose
Then it splits the change by device, browser, product, source, country and customer type, and lines it up against everything that changed on the store that day.
Act
Everything lands in one ranked backlog — by impact × confidence ÷ effort — so the next thing to do is the top row, not a judgement call.
Measure
Then it reports what the work actually earned, judged on revenue per visitor. Including the recommendations that did not work.
A number, or a sentence.
Conversion rate: 2.44% (−18.2%)
Mobile conversion is down 27% since Tuesday's theme update, on Safari, on the three products with a variant selector.
Average order value: $90.41 (−6.4%)
AOV fell because AUTUMN20 has no minimum order value and single-item baskets are up 41%.
Checkout completion: 62.1% (−3.8%)
Apple Pay's share of completed checkouts fell 21% four hours after a checkout app updated.
Both columns are reading the same store on the same day. The left one is true and the right one is actionable, and that gap is the whole product.
Diagnostic rather than descriptive.
Know whether the store is actually fine
Not another chart wall. Revenue per visitor decides, conversion rate and order value explain it, and the funnel says which step broke. Five tiers, and only the top one is allowed to call a win.
Spend your afternoon on the thing that pays most
One ranked list, by impact × confidence ÷ effort. A $2,600 fix you can do from a settings page outranks an $8,200 one that needs a developer — which is the right advice for anyone with one afternoon.
Find the half of your traffic that is failing
Mobile against desktop, Safari against Chrome, this landing page against that one. An average hides the fault; the split finds it — and the share column stops a collapse in 2% of traffic looking like a crisis.
Know what caused it, not just that it happened
Theme updates, price changes, discounts, app installs, campaigns. A drop next to the thing that caused it is a cause, not a number.
Trust the number enough to act on it
Scout separates what it measured from what it inferred, and says which is which. A guess labelled as a guess is useful; one dressed as a finding is not.
Never stare at a chart wondering what to do
Every issue ends in a short list of next steps — and 'keep watching' is on it when that is honestly the right answer.
Stop checking a dashboard every morning
You should not have to check. Scout tells you when something matters, and you set how much it takes to count as mattering — per metric, per channel.
Settle an argument with evidence, not opinion
When the fix is a judgement call rather than a bug, Scout drafts the test — hypothesis, variants, one primary metric — and hands it to Zinx Signal to run.
Prove the work paid for itself
Proposed, shipped, measured, banked. Only A/B tests and clean before/after windows count toward the total — and the things that lost money are on the same page, because a ledger of nothing but wins is a brochure.
Not what we recommended — what it earned.
Every CRO tool tells you what to change. Scout keeps the receipt: what shipped, when, what it moved, and whether that move can honestly be credited to the change.
- A/B tests count. Before/after mostly does not.Stores are seasonal and several things ship a fortnight. Measuring revenue before a change and after it produces a number that is mostly everything else that happened — presented as your work. Scout counts a before/after window only when its own change timeline shows nothing else moved in it.
- Judged on revenue per visitor, always.One of our own recommendations lifted average order value 7% and lost money, because conversion fell further. On AOV it files as a win. It is in the ledger as a loss.
- Monthly, never annualised.Multiplying a fortnight’s uplift by twelve assumes the effect persists, does not decay, and survives a change of season. None of that has been measured, so we do not do the multiplication.
Six changes, three counted
The total Scout claims is the smallest defensible reading, not the largest. The entries it will not take credit for stay on the page with the reason — because a ledger you cannot audit is a number you have to take on faith, and that is the opposite of the point.
What Scout can see, and what it cannot.
ShopifyRequired
Everything on every page of the product comes from here:
- Orders, sessions, products and checkout steps
- Theme publishes, app installs and price changes
- Discounts, collections and on-site search
The blind spotsOptional
Scout names what it cannot see and what it is inferring instead, so you know which findings are weaker and why. Each source closes a specific one:
- Meta and Google Ads — tells a creative change apart from a conversion fault
- Clarity and Hotjar — shows the page Scout currently has to infer
- Replo and theme feeds — dates a change to the minute
- Zinx Signal — stops Scout diagnosing your own running test as a fault
Four steps, and the last is stop checking.
- 01
Connect the store
One click on Shopify. No script tag, no tag manager, nothing to add to your theme.
- 02
It learns your baseline
Scout reads your last 90 days to learn what normal looks like for this store — not for stores in general.
- 03
Set what matters
Choose what it watches and how loudly. Conversion and checkout are on by default; the rest is yours.
- 04
Then stop checking
From then on it finds the problems, works out why, and tells you what to do. You open it when it asks you to.
Less than one problem it finds.
Every plan has every feature, including the ledger that tells you when a change earned nothing. The only thing that changes is how many visitors a month Scout is watching — so if the pixel ever stops reporting, your bill goes down rather than up.
Starter
Up to 25,000 visitors$29/ mo
Growth
Most storesUp to 150,000 visitors$79/ mo
Scale
Up to 600,000 visitors$199/ mo
Pro
Up to 2,000,000 visitors$449/ mo
Enterprise
Above 2,000,000 visitorsLet's talk
14 days free, then billed through Shopify on your usual invoice. 10 things are in the first plan and in every other one; there is no tier that withholds the part that makes Scout worth having.
The ones people ask first
Find problems before they cost you.
Scout watches the store so you don't have to. When something moves, it works out why and tells you what to do next.
- Connects in one click — no tag to paste, no developer
- Reads 90 days of order history the moment you connect
- Tells you what to do, not just what happened