What a Magento and Hyvä Store Costs: Subscription, One Off, or Custom Build

There are three honest ways to pay for a Magento store on Hyvä, not two. You can rent one, as a monthly subscription where the build, the infrastructure, the updates and the support arrive together. You can buy an accelerator outright, paying a single fee for a prebuilt store that is yours from the day it is handed over. Or you can commission a build from scratch and own that. Most comparisons collapse this into cheap against expensive, which is the wrong axis, and most of them also leave the middle option out entirely. The real question is what you hold at the end, and what you are still paying for once you hold it. We have written separately on what Magento development costs by project shape. This piece looks at the same money through ownership instead.

Three ways to pay for the same store

The store itself can be identical in all three cases. Same Magento Open Source, same Hyvä Theme, same catalog, same checkout, same speed. What changes is the contract around it: whether you are renting continuous access, buying a finished thing outright, or commissioning one that does not exist yet. The contract is where the money starts behaving differently.

What a subscription actually rents

On a subscription you pay for continuous access rather than for a finished object. In our case, Liquid Framework starts at 799 EUR per month plus VAT on yearly billing, or 899 EUR per month plus VAT on monthly billing. The theme, the 70+ modules we maintain, the infrastructure, the updates and the support sit under that one line. PageSpeed is written into the contract at 95 to 100, which matters mostly because it turns the number into a commitment rather than a screenshot taken on a good day.

What you are renting, precisely:

• The accelerator itself, the theme and the module set, maintained by us and improved for everyone on it at once.
• The infrastructure and its upkeep, which stops being a line you have to think about.
• A support agreement with named response and resolution times, so a broken checkout has an answer and a deadline attached to it.
• Upgrades you do not schedule, because they happen underneath you.

What a subscription does not cover is work outside the standard setup. Migrations, large or messy catalogs, custom features and integrations such as Algolia, Connectif or an ERP are quoted separately, because they are genuinely separate. Any accelerator that implies otherwise is pricing a project it has not seen yet.

Exit terms, code ownership and any minimum commitment are contract questions rather than price questions, and they differ sharply between providers. They decide what a subscription is actually worth to you, so read them before you sign rather than in the month you want to leave.

What a one off accelerator buys

This is the option most comparisons skip, and it fits more merchants than either of the ones either side of it. A number of Hyvä partners sell their accelerator package as a single fee rather than a subscription: you pay once, the prebuilt store is installed and configured for you, and the code is yours from the day it is handed over. Hyvä's own write-up of these packages says plainly that pricing is monthly or one off, so this is the ordinary shape of the market and not an exception to hunt for.

Set against a subscription, what changes is that there is no monthly line, no minimum term, and nobody who can switch the store off. Set against a build from scratch, what changes is that you are not paying for somebody's first project. The package already exists, so both the price and the timeline are a fraction of what a build would be. You get ownership without buying it at build prices, which is the specific trade a lot of merchants are actually looking for when they say they want to own their store.

What to read carefully, because this is where one off packages differ from each other most: what happens after the sale. Ask whether updates to the accelerator are included, for how long, and what they cost once that period ends. Ask the same about support, and about who installs Magento security patches. A single fee with no update path means you own a store that stops moving on the day it is handed over, and two years later you are paying somebody to catch it back up. That can still be the right deal. It is simply a different deal from the number on the page.

What a build from scratch buys

Commissioning a build is the opposite trade. You pay for the work once, the code is delivered, and from that day the store is an asset on your side of the table rather than a service on somebody else's. That is genuinely the better deal when the store is the business: when you have a developer in house, when the roadmap is long and specific, when the catalog or the integrations are unusual enough that no standard setup would fit them anyway.

The cost does not stop at delivery, though, and that is the part most comparisons leave out. An owned store still needs:

• Hosting, bought and managed by you or by someone you pay to manage it.
• Security patches and Magento upgrades, which arrive on Adobe's schedule rather than on yours.
• Module updates, including every third party one you installed and then forgot about.
• Somebody available when checkout breaks at 11am on a Saturday.
• The next feature, and the one after that, quoted each time.

None of that is an argument against owning your store. It is an argument for counting it before you compare. What Magento maintenance actually costs goes through those numbers properly.

The licences sit in all three columns

Some costs do not care which route you take at all, and it is worth lifting them out of the comparison so they stop distorting it.

Magento Open Source carries no licence fee. Adobe Commerce does, and where it appears it is usually the largest single line in the whole budget. Which of the two you actually need is a separate decision from how you pay for the build, and it is worth making first, because it moves more money than anything else on this page.

Hyvä Theme is now open source, which removed a licence that used to sit in this section of every quote. If your last proposal predates that change, what it means for merchants is worth ten minutes before you compare anything.

Hyvä Checkout is still paid. Hyvä lists it at a one time fee of 1,000 EUR, which includes one year of support and updates and a Hyvä Theme licence, valid for one Magento 2 installation with unlimited domains and store views. From the second year it is 250 EUR per year, or 1,000 EUR once to cover five. We have written on whether the paid parts of Hyvä are worth buying. The only point here is that this cost is identical whichever of the three you choose, so it is not a difference between them.

The three year view

Three years is the right window. It is long enough for a build's second and third year costs to show themselves, and short enough that the store has not been rebuilt again in the meantime.

On a subscription the arithmetic is deliberately boring. The monthly figure times thirty six, plus whatever sat outside the standard setup, plus Hyvä Checkout if you take it. You know the total in advance, it does not move, and it is the same number whether the year is quiet or not.

On a one off accelerator you pay once at the start, and after that the only question is what you keep paying for: hosting, updates if they were not included, support if you want it, and each change as you ask for it. For a store that stays close to the standard setup this is normally the lowest of the three totals over three years. It is also the one with the widest spread between providers, because everything after the initial fee is negotiable and most of it is invisible on the sales page.

On a build, the first year is the expensive one and the next two are the uncertain ones. You pay for the build, then hosting, then maintenance, then each change as it comes. The three year total lands below a subscription more often than agencies like to admit, as long as the store is stable and somebody internal absorbs the small things. It lands well above when nobody internal does, because then every small thing becomes a quote, a wait and an invoice.

So no shape is cheaper in general, and anyone who tells you otherwise has not asked about your store. A subscription is cheaper for a store that fits a standard setup and has nobody internal maintaining Magento, and it is the only one of the three where somebody else is contractually on the hook for speed and response times. A one off accelerator is usually the lowest three year total for a store that fits the standard setup and has somebody, internal or hired, to keep it current. A build from scratch earns its price when the store's shape is genuinely unusual. Any comparison that reaches a conclusion without asking which of those three you are is comparing something other than your store.

One thing is true in all three columns: three years is also roughly when the question comes back. Platform versions move on, a design starts to show its age, and the catalog has grown into a shape nobody planned for. A subscription absorbs that gradually, because the theme and the modules keep moving underneath you. A store you own absorbs it as a project, at a moment of your choosing, which is better when you choose it and considerably worse when you do not notice it has become due.

What the money buys in speed

Both routes can produce a fast store and neither one guarantees it. The HTTP Archive median Lighthouse score across Hyvä stores is about 47.5, which is a useful corrective to the idea that the theme does the work on its own. What moves the number is what gets removed, what gets tuned and what gets kept out.

Two measured examples from our own work. On Vitacom, largest contentful paint on phones went from 3.08 s to 1.35 s, and on desktop from 2.34 s to 0.67 s, measured at the CrUX 75th percentile between December 2025 and September 2026. On ToMyBaby, the mobile PageSpeed score went from 61 to 99 and total blocking time from 8,066 ms to 42 ms, measured on 2 and 10 September 2025. Those are field numbers on real stores, not a demo.

Which of the three you are

• Rent if the store fits a standard setup, nobody internal maintains Magento, and a predictable monthly cost is worth more to you than owning the code.
• Buy once if the store fits a standard setup but you want the code to be yours, and you have somebody who can keep it patched and current, whether that is a person on your payroll or a maintenance agreement you buy separately. Check the update and support terms before the price.
• Build from scratch if the store's shape is genuinely unusual, the catalog or the integrations would fight a standard setup, and you have or are about to hire a developer. This is the expensive answer and it is the right one less often than it gets sold.

If you sit between the two, the deciding question is usually not money at all. It is who gets the call when checkout breaks, and whether that person already works for you.

The next step

The two columns only separate once somebody has looked at your catalog, your integrations and who maintains the store today. Book 30 minutes and we will go through all three with you and say plainly which column your store belongs in, including when that answer is the one we do not sell.