· 16 min read

How Many Sales Does a Shopify Store Have? Free Methods

You cannot read a Shopify store's orders from outside. Here is what five free public signals can rank for you, and why every revenue number is a model.

Analyst reading a printed report of numbers next to a calculator

You found a store worth studying. It ranks on the first page for a term you care about, it runs ads you keep seeing, and its product pages are better than yours. The next question is almost always the same one: how many sales does this thing actually do?

The honest answer takes about ten seconds and then thirty minutes of work. You cannot read a Shopify store’s order count from the outside, and neither can any tool, because order data never leaves the merchant’s admin. What you can read is behaviour: how often the catalog changes, how long ads stay live, how fast reviews accumulate, how much traffic a panel estimates. Each of those is a rank, not a number. Put five of them together and you can order ten stores by momentum with far more confidence than any single revenue estimate would give you.

This guide explains what is actually published, why the sales counters on a product page are not evidence, how to build the rank yourself for free, and where the whole approach stops working.

The Number You Want Is Not Published Anywhere

It is worth being precise about why, because it saves you from trusting the wrong sources later.

A Shopify storefront exposes a narrow set of public surfaces: the rendered HTML of its pages, products.json and collections.json for catalog data, sitemap.xml for a list of URLs, and robots.txt for what you may crawl. That is roughly the whole public API. None of those endpoints carries order data, and none of them could — a storefront is built to sell a product to one visitor, not to report the business’s ledger to strangers.

Orders live on the other side of the checkout. When a visitor buys, the cart moves to Shopify’s checkout, the payment is processed, and the resulting order is written to the merchant’s admin. The storefront’s own scripts can see a cart and a checkout initiation; they cannot see whether the payment cleared, how many other people bought, or what the refund rate looks like. Public storefront code does not receive the numbers you are looking for.

So when a page says “1,204 sold in the last 24 hours”, that sentence was not produced by Shopify. It was produced by a sales-popup or social-proof app that the merchant installed and configured, sometimes seeded with a starting number so the widget does not look empty on launch day. Some of those figures track real orders. Some are decorative. You have no way to tell which from the outside, which is exactly why the number should be filed under copy rather than evidence.

The same caution applies to “Only 3 left”, countdown timers, and “27 people are viewing this right now”. Inventory counters can be genuine or threshold-triggered, timers are almost always set to reset, and viewer counts are frequently a range picked in a settings panel. None of this makes the store dishonest; urgency widgets are a normal conversion tactic. It does mean you should never build a research conclusion on top of one.

Three Kinds of Numbers, and Who Produced Each

Before ranking anything, label every figure you collect by its origin. The label decides what the number can support.

What you seeWho produced itWhat it can support
”1,204 sold today”, “27 viewing”A conversion app the merchant configuredNothing on its own; treat as marketing copy
”Only 3 left”An inventory or urgency appA weak stock signal, sometimes real
Review count on a productA review app, fed by customers and importsOrder proxy with a large error bar
Monthly visits, traffic sourcesA panel-based estimate (SimilarWeb-style)Direction and rough magnitude, not orders
Active ads and their start datesMeta Ad Library, published by the platformAdvertising intensity and campaign age
A revenue figure in dollarsA third-party modelOnly as good as its disclosed method

The pattern in that table is the whole article in miniature. The numbers closest to the truth are the ones the platform publishes about advertising. The numbers furthest from the truth are the ones presented most confidently, usually in dollars.

What the Storefront Will Actually Tell You

Two public fields are underused, and both are dated.

products.json returns each product with created_at, published_at, and updated_at timestamps. That means you can tell when a product entered the catalog, not just that it exists. If a store added nine products in the last three weeks and none in the six months before that, you are watching a store in an active testing cycle, and the timestamps prove it rather than suggesting it. The same file also returns variant prices, SKUs, and compare-at prices, which is how you date a discount rather than guess at one.

sitemap.xml gives you lastmod values per URL. Use it to see which product and collection pages were touched recently. A collection page that changes every few days is being merchandised; one that has not moved since spring is not. If you want the counting method for catalog size, our guide to how many products a Shopify store has covers the pagination and multi-file quirks; here the timestamps matter more than the total.

Neither field says anything about orders. That is the point of listing them first: the visible data is rich enough to date a decision, and silent on volume.

Five Free Proxies, and What Each One Ranks

Each proxy below answers a different question. None of them produces a sales number, and all of them produce a comparable value you can sort.

1. Review velocity. Record a product’s review count today, then again in seven days. The difference is not a sales rate, because most buyers never review and review apps allow importing, incentivised, and migrated reviews. But across stores in the same category, the ordering usually holds: the store adding forty reviews a week is doing more volume than the one adding two. Compare within a category, never across categories, since review propensity varies wildly between a $12 accessory and a $400 piece of furniture.

2. Ad intensity and campaign age. Open the Meta Ad Library, filter by the store’s page, and note two things: how many ads are active, and how old the longest-running one is. Volume tells you how hard the store is pushing. Age tells you whether the push is paying for itself — nobody keeps funding the same creative for four months if it loses money. Our walkthrough of reading competitor ads for free covers the filters and the cards in detail.

3. Catalog churn. Count products added and products removed over a month, using the published_at timestamps and the sitemap. A store that adds and retires SKUs steadily is running a portfolio. A store that added one product two years ago and nothing since is running a single bet that already resolved.

4. Estimated traffic. A panel-based traffic figure is modeled from a sample, not measured, so treat it as an order-of-magnitude signal. It still separates a store doing tens of thousands of monthly visits from one doing a few hundred, which is often the difference that matters for whether a store is worth studying at all. Our guide to checking store traffic for free explains what those panels can and cannot see.

5. Stock-state flicker. Watch the variant selector on three hero products for two weeks. Variants that go sold out and come back indicate real inventory movement. Variants that are permanently sold out indicate a product the store has stopped restocking, which is a different and equally useful signal. Be careful with pre-orders, made-to-order items, and stores that display stock only below a threshold.

Rank stores on each proxy from 1 to 10, sum the ranks, and you have a momentum score. It is crude, it is unglamorous, and it will beat a revenue estimate you cannot audit.

Why Converting Proxies Into Dollars Goes Wrong

Every revenue estimate in this space runs the same equation.

Revenue = visits × conversion rate × average order value.

Visits can be estimated. Average order value can be partly read from public prices, though bundles, discounts, and shipping thresholds distort it. Conversion rate cannot be observed at all from outside, and it is the input that decides the answer.

Take a store with 40,000 estimated monthly visits and a $45 average order. At a 1% conversion rate that is 400 orders and about $18,000 a month. At 3% it is 1,200 orders and about $54,000. Same store, same traffic, same catalog, three times the revenue — and the difference sits entirely in a number nobody outside the admin can see. Two stores with identical traffic routinely differ several-fold on conversion because of offer quality, price positioning, and how warm the traffic is.

That is why a tool reporting “$31,400/month” is not lying so much as hiding two assumptions behind one confident integer. The integer is a model output. Models are useful; undisclosed models presented as measurements are not.

Use three questions on any revenue figure you encounter, including ones you generate yourself:

  • What is the method, in one sentence I could reproduce?
  • What date does the underlying data come from?
  • Is the output a range, and how wide is it?

A source that answers all three is worth reading. A source that answers none of them is a number-shaped opinion.

A Worked Example (Hypothetical, Composite)

Three stores in the same home-organisation niche, read over one month. The details are invented for illustration; the ranking logic is the part worth copying.

Store A lists 61 products, nine added in the last 30 days. Three hero products gained roughly 180 reviews combined in four weeks. Meta Ad Library shows 34 active ads, the oldest launched 97 days ago. Estimated traffic is in the low tens of thousands of monthly visits. Two hero variants cycle in and out of stock every week.

Store B lists 240 products, none added this month. Review counts on the top five products moved by eleven in total. Ad Library shows 41 active ads, but the oldest is 12 days old, and the creative set turns over almost completely each fortnight. Traffic is similar to A.

Store C lists 14 products across four unrelated categories, six added this month. Review counts are in single digits and static. Ad Library shows four ads, oldest 21 days. Traffic is a fraction of the other two.

Ranked on the five proxies, A comes first, B second, C third — and the interesting part is that B has more ads and more products than A. The distinguishing read is campaign age and review velocity: A is compounding, B is paying for a plateau and refreshing creative to hide it, C is a test portfolio that has not found its winner yet.

Note what the ranking did not produce: a dollar figure for any of the three. That was deliberate. The decision this ranking supports is “study A’s funnel and offer, watch B for a price or creative reset, check C again in six weeks” — and that decision does not need revenue to be correct.

Four Patterns Worth Naming

Once you have ranks for a set of stores, the combinations carry meaning.

Rising reviews, rising ads, growing catalog. The store is scaling. Whatever it is doing, it is working, and it is the best candidate for a funnel teardown. Our guide to reading a product page is the next step.

Flat reviews, high ad count, short campaign age. The store is buying traffic and not converting it into repeat demand. Constant creative refresh with no accumulation is the signature of a plateau being propped up.

Low ads, steady reviews, stable catalog. Often an email- and retention-led business, or one carried by organic search. Benchmark against it carefully — its economics assume a much lower acquisition cost than yours.

Shrinking catalog, falling ad count. Either an exit or a deliberate reset. Do not read a reset as failure without checking whether a new collection appeared at the same time; our guide to tracking store changes over time covers how to tell those apart.

Where Free Proxies Break

Five limits are worth stating plainly.

Small stores fall below the traffic threshold. Panel estimates need enough sample to model, so stores doing a few hundred visits a month often show as zero or not at all. The proxies still work on their other four dimensions; the traffic column just goes blank.

New stores have no baseline. A store three weeks old has no review history and no campaign age. You can see what it is doing, not whether it is working. Wait a month.

Imported reviews distort velocity. Stores migrating from another platform, or running an app that imports marketplace reviews, can show a large count that never moves. Watch the delta, not the total.

Wholesale and B2B stores are invisible. A store doing forty orders a month at $4,000 each looks tiny on every proxy and is not. If the pricing is tiered or the site asks for a business account, the consumer proxies do not apply.

One brand can own several storefronts. Reading each domain separately double-counts the brand and can make a single operation look like three competitors. Check the policy pages and the footer entity before you add stores to a list.

There is also a structural ceiling: all of this is manual, per store, and has no memory. Ten stores is a Saturday. A hundred is a spreadsheet nobody maintains.

If You Are the Store Being Read

The same five proxies describe your store to everyone else. That is worth sitting with for a minute.

A competitor can date your catalog additions, count your active ads, see how old your longest-running campaign is, watch your variants go out of stock, and estimate your traffic. They cannot see your conversion rate or your orders — but they can rank you against nine other stores in the niche, and they will read a plateau correctly more often than you would like.

Your own admin is the only place the real numbers exist. That is also why the more useful comparison is not “what does this store make” but “what does this store do differently from me, in the parts we can both see”. Our breakdown of what store analytics actually measure is the counterpart to this article: same discipline, applied inward.

A 20-Minute Monthly Routine

  • Pick ten stores from your niche, chosen for comparability rather than size.
  • Record five columns per store: review count on three hero products, active ad count, oldest active ad date, product count, estimated monthly visits.
  • Add two dated fields: products added in the last 30 days, and any hero variant currently sold out.
  • Wait seven days and repeat the review counts. The delta is the only column that measures movement rather than state.
  • Rank, do not total. Convert each column to a rank across your ten stores and sum. Resist the urge to weight the columns into a score with decimals; the precision is fake.
  • Write one sentence per store about what changed, in plain language. That sentence is what you will actually use next month.

Twenty minutes a month, ten stores, one sheet. It will not tell you what any of them earn. It will tell you which one to study, which one to ignore, and which one just changed something — and those are the three questions that move research forward.

If you want the same read across a much larger set, the manual method is the bottleneck, not the logic. TrendTrack indexes 1.5M+ Shopify stores alongside 10M+ Meta ads and 300M+ total ads, so the filtering you just did by hand can run across a category instead of a shortlist. TrendTrack promo code is WINNING - 20% off all plans for the first 3 months if you want to test whether the scale changes your output before committing further.

Apply WINNING for 20% off TrendTrack →

Verified 2026 · 20% off Starter, Pro and Business for your first 3 months

If you would rather stay entirely on free tools, start with the storefront itself. The free extension surfaces a store’s stack, catalog size, and ad activity in a side panel while you browse, which removes the tab-switching from the routine above without adding a number you cannot audit. Run the ten-store sheet once, then decide whether the manual ceiling is the thing actually limiting your research.

FAQ

Does TrendTrack have a promo code?

TrendTrack promo code is WINNING - 20% off all plans for the first 3 months. It applies to Starter, Pro and Business, works with monthly and annual billing, and stacks on top of the annual discount — an annual subscription with the code saves about 33% in year one.

Can I see exactly how many sales a Shopify store has?

No, and no external tool can either. Order data lives in the merchant’s Shopify admin and behind the checkout, and it is never exposed through the storefront, products.json, the sitemap, or any public endpoint. Anything presented as an exact order count for someone else’s store is an estimate built from proxies, whether or not the source says so.

What does “1,204 sold today” on a product page mean?

It is generated by a sales-popup or social-proof app that the merchant installed and configured, and some of those apps let the merchant set a starting number. The figure may track real orders or may be decorative, and you cannot tell which from outside. Treat it as conversion copy rather than as a measurement of volume.

Are Shopify revenue estimate tools accurate?

They are models, and their accuracy depends entirely on the conversion rate they assume, which is unobservable from outside. A store with 40,000 monthly visits can plausibly produce three times another store’s revenue on identical traffic. A tool that discloses its method, its data date, and a range is worth reading; one that shows a single confident dollar figure is not.

What is the most reliable free signal of a store’s sales volume?

Campaign age in the Meta Ad Library. An ad that has been running for months is being funded by revenue, since almost nobody keeps paying for a losing creative that long. Pair it with review velocity measured as a seven-day delta, and you have two independent signals that are hard to fake and easy to re-check.

Can I see how much a store spends on ads?

Partly. The Meta Ad Library shows every active ad and its start date for free, and for ads targeting the EU it also publishes reach and spend ranges under the Digital Services Act. Outside those cases you get creative and timing, not budget. Ad count and campaign age are still enough to rank advertising intensity across stores.

Is review count a good proxy for sales?

It is a proxy for orders with a wide error bar, not for revenue. Most buyers never leave a review, review apps allow imports and incentivised reviews, and a single order can generate more than one review. Used as a seven-day delta within one category it ranks stores usefully; used as an absolute number across categories it misleads.

    Share:
    Back to Blog

    Related Posts

    View All Posts »