Mirakle Research · September 14, 2026

Which app do merchants say they switched to?

Of the 1,095,175 Shopify App Store reviews we hold, 359 say that the writer moved from one named app to another app in the same category: about 3.3 in every 10,000. Most of them fall into three categories. In product reviews, 66 describe arriving at Judge.me. In subscriptions, 55 describe leaving Recharge. In email marketing, 57 describe leaving Mailchimp.

359

reviews that describe a completed move between two apps in one category

93.3%

of them were written on the listing of the app the merchant moved to

181

sit on the 16 routes that 5 or more reviews name

Every matched review is one strand of a ribbon, grouped by category. An app’s arc is as long as the reviews that mention it: brown where merchants say they left it, slate where they say they arrived. Routes that at least 5 reviews name are drawn between the two apps; every smaller route passes through the grey “other” arc of its category, so an app’s arc still counts all of its reviews. Widths count reviews, not merchants. On a narrow screen the app names are in the three category rings further down.

The first suspect

How a review becomes a route

Every number on this page is our reading of review text, so the reading comes first. A search for the words switched, migrated, moved, changed, came over and better than returns 17,254 reviews. Almost all of them are about something else: a theme, a plan, a move to Shopify from another platform, or “another app” with no name.

A review is kept only if it says in the first person that the move happened (“we switched from”, “I moved to”) and the other app’s name follows the preposition directly. Comparisons, intentions and recommendations are left out. The name has to be on a list we wrote by hand: 61 names that mean one listing wherever they appear, and 6 brand names that sell several products (Yotpo, Stamped, Smile, Bold, Loop, Appstle) and are read as a name only inside a category where they mean one. A review that tells of more than one move, or places the listing it was written on at both ends, is left out, because we cannot tell which move it is reviewing. Both apps have to share a category, and both have to have launched before the review was posted.

Each bar is what remains after one more rule, on one linear scale. The first step removes 16,858 reviews; the category rule 31 and the date rule 6.

To check the reading, we read 200 of the 359 in full, chosen by a fixed hash of their row numbers. The first reading found three wrong routes: two reviews narrated a chain of moves, and one named its own app’s brand. We changed the rules for each of those cases and read the sample again. 198 of 200 then described the move they were counted as; the other 2 described a move still being made. The reader was Claude, an AI model, and not an independent person, and the rules were corrected on the same sample, so 198 of 200 is the best case for this sample rather than a measured error rate. How many real switches the rules miss was not measured.

What we checked next

Which listing the review was written on

A move has two ends, and a review sits on one of them. 335 of the 359 (93.3%) were written on the listing of the app the merchant moved to. Only 24 were written on the app they left.

One dot per matched review. Filled slate dots were posted on the destination’s listing, outlined brown dots on the listing the merchant left.

That tilt decides which moves this study can see at all. The review names the other app, and the other app has to be on our list. With nine reviews in ten written at the destination, the other app is almost always the one that was left, so the apps shown losing merchants are, nearly by construction, the well-known names we listed. A merchant who leaves a small app for Klaviyo and writes about it on Klaviyo’s page is invisible to us unless that small app is on the list. The list also shapes the tilt: a move from a listed app to an unlisted one can only be seen from the unlisted app’s page, so some of the 93.3% comes from which names we wrote down rather than from where merchants choose to write.

Written at the destination

A review on Judge.me saying we switched from Yotpo. Counted when the app left (Yotpo) is on our list. The listing it was written on can be any app.

Written at the source

A review on Omnisend saying I switched to ActiveCampaign. Counted when the app moved to is on our list. Rare: 24 of 359.

The tilt could also be something our data made. Our guess was that a complaint written on the app a merchant left is the kind of review most likely to be deleted or archived, which would thin out the source side of the record. So we kept the 19 matched reviews that are no longer on the shelf (5.3% of the set). Without them the destination share is 94.4%; with them it is 93.3%. It is 90.5% for reviews posted up to 2022 and higher since. The review’s posting date is not the date of the move, and a review edited later keeps its first date.

The slate part of each bar is the share written on the destination’s listing, on a common 0–100% scale. Periods are by the year the review was posted; 2026 is incomplete.

What survived

The routes, category by category

Read with that in mind, the routes are a record of arrivals, told by the merchants who arrived, about the well-known apps they came from. Within those limits they are not spread evenly. In product reviews, 66 reviews describe arriving at Judge.me and 15 describe leaving it. The named routes into it come from Yotpo Reviews (36), Stamped (16), and Loox (6). Yotpo to Judge.me is mostly an older route; 25 of its 36 reviews were posted by 2022.

In subscriptions the pattern runs the other way. One app is left and the arrivals are spread: 55 reviews describe leaving Recharge and 5 arriving at it. 45 of the departures name Appstle Subscriptions (14), Skio (13), Loop Subscriptions (11), and Stay AI (7). Skio’s listing now calls it a Recharge company, so some of those moves may end inside the same business. No route into Recharge reaches 5 reviews. Email marketing has both shapes: 57 reviews leave Mailchimp, and Klaviyo sits in the middle, reached by 22 and left by 25.

Product reviews · 119 reviews, 53 of them on routes fewer than 5 name

Yotpo 53 left, 3 arrived · Stamped 27 left, 1 arrived · Loox 10 left, 11 arrived · Junip 0 left, 12 arrived · Judge.me 15 left, 66 arrived

Email marketing · 97 reviews, 48 of them on routes fewer than 5 name

Mailchimp 57 left, 1 arrived · Klaviyo 25 left, 22 arrived · Spoks 0 left, 5 arrived · Seguno 1 left, 9 arrived · Omnisend 7 left, 26 arrived

Subscriptions · 86 reviews, 32 of them on routes fewer than 5 name

Recharge 55 left, 5 arrived · Bold 20 left, 1 arrived · Stay AI 0 left, 11 arrived · Loop 2 left, 15 arrived · Skio 3 left, 16 arrived · Appstle 3 left, 26 arrived

An app with a large review base collects more arrival reviews whatever the rate of arrival, and an app that asks new customers for a review during onboarding would collect more still, so the width of a ribbon mixes how many merchants moved with how many of them wrote about it. We cannot separate the two. For a developer, a route is a list of the competitors merchants name when they arrive. It is not a churn rate: we see no installs, no uninstalls, and no merchant who moved without writing. The reasons merchants give were not counted.

Every drawn route, with its count in each posting period where that count can be published
Category	Route	Reviews	Share of category	2012–22	2023–26
Product reviews	Yotpo Reviews → Judge.me	36	30.3%	25	11
Product reviews	Stamped → Judge.me	16	13.4%	6	10
Product reviews	Yotpo Reviews → Junip	8	6.7%	–	–
Product reviews	Loox → Judge.me	6	5.0%	–	6
Subscriptions	Recharge → Appstle Subscriptions	14	16.3%	–	–
Subscriptions	Recharge → Skio	13	15.1%	6	7
Subscriptions	Recharge → Loop Subscriptions	11	12.8%	–	–
Subscriptions	Bold Subscriptions → Appstle Subscriptions	9	10.5%	–	–
Subscriptions	Recharge → Stay AI	7	8.1%	–	–
Email marketing	Mailchimp → Omnisend	19	19.6%	–	–
Email marketing	Mailchimp → Klaviyo	14	14.4%	8	6
Email marketing	Mailchimp → Seguno	6	6.2%	6	–
Email marketing	Klaviyo → Omnisend	5	5.2%	–	–
Email marketing	Klaviyo → Spoks	5	5.2%	–	5
Chat	Gorgias → Commslayer	6	46.2%	–	6
Loyalty and rewards	Smile.io → BON Loyalty Program & Rewards	6	25.0%	–	6

A dash means fewer than 5 reviews in that period, or a count withheld because subtracting it from the total would reveal a smaller one. It is not a zero.

Method and limits

What the data can and cannot say

The 359 reviews name 95 listings and were posted between 2016 and 2026. They show which apps merchants say they left and joined, mostly in English and mostly at the app they joined. They cannot rank apps by customers gained or lost, and they say nothing about moves that nobody wrote down or that name an app outside our list.

Extraction, exclusions and the reading check

Snapshot 2026-09-14, taken read-only from the production database in one repeatable-read transaction. The corpus is every stored review, including 156,010 deleted or archived ones. A review contributes at most one route. There are no exact text duplicates in the retained set (0).

103 of the retained reviews were read through a brand name resolved by category, and 28 through the form “switched to this app from X”. Names resolve to current listings; renames and past category changes are not reconstructed. Two listings that sell reviews and loyalty under one name (Stamped, Okendo) count in both categories. Where a brand has two products in one category (Appstle Subscriptions and Appstle Memberships), the name resolves to the larger one unless the review was written on the other.

A route is published only when at least 5 reviews name it. Reviews on smaller routes are counted in every denominator and drawn only through the grey “other” arc of their category, so no count below that is drawn between two named apps. Category shares divide by all matched reviews in the category and period.

The reading check selects the first 200 retained reviews ordered by SHA-256 of a fixed seed and the row number; its checksum is verified at extraction, so a changed sample fails the run. The uncertain readings stay in the totals. No review text, reviewer or store name is published.

The name list

ActiveCampaign (Email marketing): activecampaign · active campaign
Ali Reviews (Product reviews): ali reviews
Appstle Subscriptions (Subscriptions): appstle subscriptions · appstle subscription
Bold Subscriptions (Subscriptions): bold subscriptions · bold subscription
CJdropshipping (Dropshipping): cjdropshipping · cj dropshipping
Consentmo (Cookie consent): consentmo
DSers (Dropshipping): dsers
EComposer (Page builder): ecomposer
Fera (Product reviews): fera
Foxify (Page builder): foxify
GemPages (Page builder): gempages · gem pages
Gorgias (Chat): gorgias
Growave (Loyalty and rewards): growave
Judge.me (Product reviews): judge.me · judge me · judgeme
Junip (Product reviews): junip
Klaviyo (Email marketing): klaviyo
LAI Reviews (Product reviews): lai reviews
LayoutHub (Page builder): layouthub · layout hub
Loop Subscriptions (Subscriptions): loop subscriptions
Loox (Product reviews): loox
LoyaltyLion (Loyalty and rewards): loyaltylion · loyalty lion
Mailchimp (Email marketing): mailchimp · mail chimp
Matrixify (Bulk editor): matrixify
Okendo (Product reviews · Loyalty and rewards): okendo
Omnisend (Email marketing): omnisend
Opinew (Product reviews): opinew
PageFly (Page builder): pagefly · page fly
Pagetify (Page builder): pagetify
Pandectes (Cookie consent): pandectes
PayWhirl (Subscriptions): paywhirl
Printful (Dropshipping): printful
Printify (Dropshipping): printify
Privy (Email marketing): privy
Rapid Reviews (Product reviews): rapid reviews
Recharge (Subscriptions): recharge · re charge
Recurpay (Subscriptions): recurpay
Replo (Page builder): replo
REVIEWS.io (Product reviews): reviews.io
Rivo Loyalty (Loyalty and rewards): rivo loyalty
Rivyo (Product reviews): rivyo
Ryviu (Product reviews): ryviu
Seal Subscriptions (Loyalty and rewards): seal subscriptions · seal subscription
Seguno (Email marketing): seguno
Sendvio (Email marketing): sendvio
Shogun (Page builder): shogun
Skio (Subscriptions): skio
SmartrMail (Email marketing): smartrmail
Smartrr (Subscriptions): smartrr
Smile.io (Loyalty and rewards): smile.io · smile io
Spocket (Dropshipping): spocket
Stamped (Loyalty and rewards · Product reviews): stamped.io · stamped io
Stay AI (Subscriptions): stay ai · stayai
Tawk.to (Chat): tawk.to
Tidio (Chat): tidio
tinyEmail (Email marketing): tinyemail
Trendsi (Dropshipping): trendsi
Weglot (Currency and translation): weglot
Yotpo Loyalty (Loyalty and rewards): yotpo loyalty
Yotpo Reviews (Product reviews): yotpo reviews · yotpo product reviews
Zendrop (Dropshipping): zendrop
Zipify Pages (Page builder): zipify pages · zipifypages

Brand names read only inside a category:

Yotpo: Yotpo Reviews in Product reviews and Yotpo Loyalty in Loyalty and rewards
Stamped: Stamped in Product reviews and Stamped in Loyalty and rewards
Smile: Smile.io in Loyalty and rewards
Bold: Bold Subscriptions in Subscriptions
Loop: Loop Subscriptions in Subscriptions
Appstle: Appstle Subscriptions in Subscriptions and Appstle Loyalty Program Reward in Loyalty and rewards

Download the frozen JSON Does paying more buy a better app? Which reviews stayed off the shelf