Lovable puts a storefront in front of you fast, and it looks the part. Then a customer pays, and its documentation gets noticeably thinner. Below: what a build costs on its published rates, and the one sentence in its own docs that decides whether it suits your shop.
Start with the basis, because most pages ranking for this phrase do not have one. Everything below was read on lovable.dev and docs.lovable.dev on 10 August 2026. We did not hand Lovable a brief and score what came back, so nothing here describes code it wrote for us, and the page does not pretend otherwise. What it reviews is what the company publishes about its own product, which has the useful property that you can go and check it.
On its own ground the product is strong. The gallery listed 193 projects the day we looked, with Ecommerce sitting among twenty one categories, and a remix drops you inside something that already looks finished. Its built in backend gives a new project a database, authentication, storage, edge functions and AI with no setup on your part. For a landing page, an internal tool or a subscription product, that is a short road and a pleasant one.
Payments are further along than the sceptics allow, too. Lovable documents accepting subscriptions and one off payments inside any app it builds, with test cards before you switch to real ones. Read a little further down that same page and it tells you plainly what the feature is scoped to. Built in payments support digital products and software, in its own words, listing subscription tiers and downloads such as courses or music.
Then comes the sentence that decides this review. For physical products, its documentation points you at its Shopify integration, which it calls a full ecommerce suite, and says Shopify handles payments, inventory and logistics when you go live. Take the other path and use built in payments for physical goods instead, and it warns that inventory and logistics will need to be managed separately. Either way the shop half is not the part Lovable is offering to build.
So the verdict splits cleanly by what you sell. Files, courses, seats, a monthly plan: Lovable will carry that end to end and you can stop reading. Anything that sits on a shelf and has to be counted before it goes out the door: you are being routed to a platform you rent, on top of the builder you already pay for, and that is a different product than the one the marketing implies.
None of that is an argument against chat builders in general, and if the whole approach is new to you the honest walkthrough of one project from first prompt to a deployed site is what learning to vibe code actually looks like.
The other half of the due diligence is the meter, since a credit is not a message and a stopped request is not a refund. Their published rates are pulled apart line by line in what a Lovable credit actually buys.
MaShop starts where this review says Lovable stops.
Lovable bills in credits, and it publishes the rates rather than burying them, which it was under no obligation to do. Two modes, priced differently. Plan mode is a flat one credit for every message you send, however long it is. Build mode costs what the work costs, so a small edit is cheap and a whole page is not.
Its own worked examples put numbers on that. Making a button gray is quoted at half a credit. Removing a footer, nine tenths. Adding authentication, one and a fifth. Building a landing page with images, two. The docs are careful to call these illustrative rather than a price list, and that caveat is fair, but the shape is clear enough to plan against.
One line matters more than the rates, and it is the one people find out later. A Build mode request you stop partway through is charged for the work already done. Hitting stop is not a refund, and neither is a wrong answer. The meter measures how much computation happened, not whether the result was any good, so a bad afternoon costs the same as a good one.
Set that against the allowance. The first paid tier starts at $25 a month and carries 100 credits, with Business starting at $50. At two credits for a page with images, an allowance like that is a few dozen substantial requests before you are topping up, and topping up is itself something the free tier cannot do.
Describe what you sell and who you sell it to.
Get it built by AI from front to back.
Your shop goes live ready to take its first order today.
If what you sell arrives as a download or renews every month, stop here and go back to it. This whole review turns on a gap that does not exist for digital goods, and MaShop would hand you a warehouse model you have no use for. Lovable will get you to a working, deployed product faster, and that is not a small thing to be good at.
The design gap is worth saying out loud too. A gallery of 193 projects is a real asset, the defaults look finished, and something a designer drew by hand will out photograph a generated shop on its first day. MaShop makes the opposite trade on purpose, putting the working parts first, and you can see that choice being made the first afternoon you spend with it.
And on a simple brief they are quicker off the line. There is also a daily allowance of five build credits, which on the free tier is capped at thirty a month, so poking at a small idea can cost nothing at all. We have no equivalent to that and it would be dishonest to pretend the comparison is close.
If this review has you weighing a switch rather than a first purchase, the version of the argument about what actually gets generated on each side is on replacing Lovable for a commerce project.
The gallery is also worth opening before you count on it, because a starter that looks like a shop and a starter that runs one are different objects. The ecommerce ones are set against a checklist in what a Lovable commerce template leaves you to build.
And if the shortlist has a second name on it, the one it usually has, both vendors are read on their own documentation in the head to head between Lovable and Bolt.
One last thing, since this review ends up pointing there. If Lovable routes you to Shopify and you want to know what that actually costs you over a year before you agree to it, the arithmetic is worked through on the page about leaving Shopify.
Built around what AI costs not what your shop earns.
Try MaShop on a side project. No card required.
Includes:
For people running a commerce app full-time.
Includes:
Built for agencies and high-volume teams.
Includes:
No, and the page says so rather than implying otherwise. Every claim here comes from what Lovable publishes on lovable.dev and docs.lovable.dev, read on 10 August 2026, and quoted where the exact wording carries the point. A build we ran and scored ourselves would be worth very little to you anyway, since we sell a competing product and you would be right not to take our word for it. Their documentation you can go and read yourself.
For what it is scoped to, yes, and this review is not a hatchet job. It generates a polished front end quickly, and its built in backend covers a database, authentication, storage and edge functions with no setup on your part. Its payments feature genuinely handles subscriptions and one off purchases too. The reservation is narrow and specific: built in payments are scoped to digital products and software, and its own docs route physical goods elsewhere.
For digital goods, yes, end to end. For physical goods its documentation points you at the Shopify integration, which it describes as a full ecommerce suite, with Shopify handling payments, inventory and logistics when you go live. That means a separate paid Shopify plan on top of your Lovable one. Choosing built in payments for physical products instead is possible, but the docs are explicit that inventory and logistics then have to be managed separately.
Its published paid tiers start at $25 a month carrying 100 credits, with Business starting at $50, and both scale up from there. The number that matters more is the burn rate: Plan mode charges one credit per message, Build mode charges by the complexity of the work, and a request you stop partway through is still billed for the work completed. Its own illustrative examples run from half a credit for a small style change to two credits for a landing page with images.
Yes, with one condition worth knowing before you subscribe. Its Git sync is two way and the docs describe the repository as a synced copy that you own. Downloading the codebase directly, and editing code in the editor, are paid plan features. MaShop commits the application to a repository you connect from the very first build.
On a lovable.app subdomain by default, and connecting a custom domain requires a paid plan. Its built in backend runs in one of three regions, chosen when you enable it and not changeable afterwards. A shop MaShop generates runs on your own hosting account and your own database from the first build, so there is no migration to plan for later.
Describe the shop once, and MaShop builds the selling half too.