About

There are tens of thousands of apps for Shopify stores and no honest, complete ranking of them. The official listing is ordered by a model nobody outside the company can inspect. The blog round-ups are affiliate placements with the affiliation buried. Everyone in the middle is selling something and pretending they are not.

EcomApps does the obvious thing instead. It ranks every app by its public review data, lets app makers pay to appear above that order, labels every one of those placements, and publishes the exact amount behind each on the board. You can disagree with the model, but you can see all of it.

Apps listed

28,153

Categories

19

Paid to the board

$8

Ranking inputs

2

How the order is decided

Two tiers, and nothing else. Listings that have paid sit on top, ordered by what they paid, and every one of them says so on its own row. Everything below that is ordered by the score described here. No payment can enter the lower tier – not even ours.

The score is a rating you can trust

A 5.0 from one reviewer is not better than a 4.8 from five hundred, so we do not treat it that way. Every app’s rating is pulled toward the catalogue average by however little evidence stands behind it. The more people who have actually reviewed an app, the more its own rating counts and the less the average does.

score = (rating × reviews + C × M) / (reviews + M)

C = 4.729   the catalogue's average rating, measured from our own data
M = 200     how many reviews it takes before an app's own rating is believed

M is the part worth explaining. Think of every app as starting with 200 imaginary reviews sitting exactly at the catalogue average. One real review is one voice against 200 – it barely moves the number. Three hundred real reviews outweigh those imaginary ones and the app’s true rating comes through.

Worked through, on real apps

Read the two ends. A 5.0 with 340 reviews beats a 4.9 with 499 – quality wins when both are well evidenced. A 5.0 with 5 reviews loses to a 4.8 with 486 – a perfect score nobody has tested is worth less than a very good one hundreds have.

The question we get asked: does a 5.0 with a single review outrank a 4.9 with 499? No. It scores 4.730, against 4.851. A perfect 5.0 needs 164 reviews before it passes that app.

What “stores in the top 100k” means

Beside every app’s rating we print a second one: the average from stores in the top 100k. It is the number the App Store cannot show you, because it needs to know who wrote each review, and it is worth being exact about what it counts.

1. Every review we hold arrives with the reviewing store attached – its name and its domains.
2. Each Shopify store carries a platform rank from Store Leads: its popularity against every other store on Shopify, driven mainly by their own traffic estimates, with page rank and link data from Common Crawl and minor signals like social following behind it. Lower is bigger: rank 1 is the most popular Shopify store in the world.
3. Top 100k means platform rank of 100,000 or better. Roughly, a store with real traffic rather than one that opened last month and has sold to friends.
4. A store with several domains is judged on its best rank, so a brand is not penalised for running a second storefront.
5. A store we cannot match to a rank is left out of that average rather than counted as small. On every review we print which it was – top 10k store, top 100k store, outside the top 100k, or unranked store.

The 100,000 line is a judgement call, the way every threshold is. We publish it so you can disagree with it, and we publish the rank band on each individual review so you can ignore our line and draw your own. It is not traffic you can audit yourself, which is the honest weakness of the number.

None of this touches the ranking. The order is the score above, computed from the App Store’s own rating and review count. Store size is shown beside it, never inside it – the same rule we apply to trending and to money.

Trending is shown, never ranked

The score answers is this good. Trending answers is this moving. They are different questions, so we keep them apart – folding momentum into the order would make the ranking swing on a single good week.

expected reviews this month = reviews in last 90 days / 3
trend = (reviews in last 30 days − expected) / expected

Each app is measured against its own past pace, never against other apps. That is why a 40-review app and a 4,000-review app can both trend – comparing them to each other would just re-rank by size, which the score already does. An app needs at least nine reviews in ninety days to qualify, or going from one review to three would read as +200%. Above +25% is rising, below −25% is cooling. We publish both: an app whose reviews have dried up is as useful to know about as one accelerating.

What the score deliberately leaves out

1. Money. It cannot enter the lower tier by construction, and the checkout refuses bids on our own apps.
2. Install counts. They are detected from code on a storefront, so an app with no visible front-end reports a number that is quietly wrong rather than missing. We would rather leave it out than rank on it.
3. Built for Shopify badges. Not in any data we can reach. We would use them if we could.
4. How recently an app was updated. Same reason – not published anywhere we can read.

We do not copy Shopify’s own order and do not claim to. Theirs blends signals they have never published. Ours is one formula, printed above, that you can check against any row on the site.

Who runs it

TouchPoints Inc. We build apps for Shopify merchants – VideoPoint (AI video for fashion brands) and TryPoint (AI virtual try-on). So we are not a neutral outsider, and saying otherwise would be the first lie on the page.

How we handle that

1. 1. Our apps are labelled From the operator everywhere they appear.
2. 2. Our apps cannot be bid on. Not “we choose not to” – the checkout rejects them, so the revenue figure above contains no money of ours.
3. 3. They rank where their review data puts them, which today is nowhere near the top, and any competitor who bids goes above them.
4. 4. Competitors are welcome to buy the spot directly above us. Several already do exactly that on boards like this one. That is the model working, not a bug in it.
5. 5. The bar at the foot of the page naming the operator and our two apps is a byline, the same one every board of this kind carries. It is not a listing, it never enters the ranking, and it is not counted anywhere in the figures above.

Where the data comes from

Public App Store listing data – name, description, category, review count, average rating, install counts where they are detectable, and how long the app has been listed. Refreshed regularly. Where a number is missing we say so rather than guessing.

Review text, the store behind each review, and the platform rank that store carries come from Store Leads. Two limits worth knowing. Reviews come back newest first with no way to filter by rating, so what we hold for an app is a recent window rather than its whole history – every review page prints how many we pulled and the dates they span. And install counts are fingerprinted from a storefront, so an app with no visible front end is undercounted; that is why we word it as seen on N stores and why installs never enter the score.

Not affiliated with Shopify

EcomApps is not affiliated with, endorsed by, or sponsored by Shopify Inc. “Shopify” is a trademark of Shopify Inc., used here only to say which platform these apps run on.