Shopify Plus cost: the licence, the fees and the build
Shopify Plus costs $2,300 a month on a three-year term, or $2,500 a month on a one-year term. That is the licence, and it is the only part of the cost of Shopify Plus that anyone publishes. The rest is the fees on top of it, the apps you add, and the build, and the build is the one that varies by a factor of twenty.
Search for what a Shopify store costs to build and you get a range: ten thousand to five hundred thousand, followed by "it depends on your requirements" and a form. The range is not dishonest. It is useless, because it spans four different purchases and nobody tells you which one you are reading about.
So this is the version with the numbers in it. What each of the four costs is, which of them we charge for, what we charge, and what two engagements you can already read about on this site would cost at today's prices. Our prices are on the site anyway, so there is nothing here we are protecting.
Four different things get called "the cost"
When a founder asks what a Shopify Plus build costs, they are usually asking about one line and getting quoted on another. Separating them is most of the work, and it is the part a discovery call is supposed to do before anyone can say a number.
Shopify Plus pricing: the licence and the fees
Shopify publishes this one, which makes it the easiest to settle. Shopify Plus starts at $2,300 a month on a three-year term, or $2,500 a month on a one-year term, for standard setups and integrations. Above a certain volume it moves to a variable platform fee based on revenue and business model, negotiated with Shopify's sales team rather than listed on a page.
The licence is not the whole of what Shopify charges. Card payments cost money on top of it, and how much depends on whether they go through Shopify Payments or another provider.
Two things follow from the licence being Shopify's. The first is that nobody resells it, so if a proposal quotes you a "Plus package" with the platform folded into one number, ask for the split. The second is that the licence is a recurring cost and the build is not, and comparing them as though they are the same kind of money is how a cheap build on an expensive plan gets mistaken for a bargain.
The licence is also the least interesting number here, because it is the same for you as for everyone else at your volume. The lines below are where stores actually differ from each other.
The apps are the line that grows on its own
Every Plus store we have opened has an app bill nobody has audited in a year. Each app was a reasonable decision on the day it was installed, usually cheaper and faster than building the thing properly, and each one renews quietly.
Apps charge twice. There is the monthly fee, which is visible, and there is what they cost on the order path, which is not. On one checkout we audited, 2.4 seconds of a 3.9-second delay before the payment button became interactive came from three scripts, and two of them rendered nothing on the checkout at all. That second cost is paid in conversion rate, by you, every day, and it does not appear on any invoice.
Before you price a build, price the list you already have. Every app, its monthly fee, who asked for it, and whether a customer would notice its absence. It is an afternoon of work, it needs no developer, and it routinely finds two subscriptions nobody can justify.
The build, and why quotes for it vary tenfold
Two agencies look at the same store and quote $12,000 and $120,000. Both numbers can be arrived at honestly, because a fixed bid for an unscoped project is not really a price for the work. It is a price for the work plus the error bars, and the error bars are wide when nobody has read the codebase yet.
So the bid has to carry a discovery phase to narrow the unknowns, a project manager to coordinate the people doing the narrowing, an account manager so you have someone to call, and a contingency for what discovery still missed. None of that is padding in the fraudulent sense. It is the structure a large fixed bid requires. But you pay for it whether or not the risk it was priced against ever materialises, and you pay for it before a line of code exists.
The alternative is not a cheaper estimate. It is a smaller unit. If the thing you buy is one week of one engineer against a scope you both agreed, the estimate cannot be wrong by $80,000, because it was never $80,000. It can be wrong by a week, and you find that out in a week.
A fixed bid on an unscoped project is not a price for the work. It is a price for the work plus the error bars, paid up front, whether or not the risk shows up.
What we charge
Commerce Lab is $600 for a one-week sprint and $950 for two weeks. That is an engineer working full time on a scope the two of us wrote down before the sprint started, with a demo at the end of it and the code in your repository throughout. QA, DevOps and UI/UX are $250 each per sprint if the work needs them, and most Shopify work needs none.
There is no discovery phase to buy, because the scoping call is free and it is thirty minutes with the engineer who would do the work. There is no account manager, because the person writing the code is the person answering your messages. The first fourteen days are refundable, which is the part that carries the risk you would otherwise have paid a contingency line to cover.
What that price does not include: the Plus licence, your app subscriptions, and design direction for a brand that does not have one yet. We build against a design you have or a design you approve; we are not a creative studio and pretending otherwise would waste a sprint.
What real Shopify Plus work costs, in sprints
Here is the part nobody publishes. Two of these are engagements you can read in full, Pipers Farm and Habbal Jewellery, with the sprint counts those projects actually took. The other two are the shapes we are asked for most often.
the Pipers Farm engagement 2 MONTHS $3,800
the Habbal engagement, two labs 12 SPRINTS $11,400+
The spread between the first row and the last is the honest answer to "what does a Shopify Plus build cost", and it is about twenty to one. That is not because the day rate changes. It is because "a Shopify build" describes both a week of deleting scripts from a checkout and six months of building seven ways to sell a diamond on one backbone.
Two caveats that matter more than the numbers. The Habbal figure is a floor, not a quote: twelve sprints across Commerce Lab and Dev Lab Web lands somewhere between $11,400 and $16,800 depending on the mix, and that project genuinely needed both. And a sprint count from someone else's store predicts yours about as well as their electricity bill does.
The five questions that actually move your number
These are what we ask on the scoping call, in the order they change the estimate most. You can answer them before you speak to anyone, including us.
How many systems touch an order? An ERP, a 3PL, a tax connector and a subscription app are four integrations, and integrations are where estimates go wrong, because each one has an owner who is not you and a rate limit nobody documented.
Is there data to migrate, and does anyone own it? Customers, orders, subscriptions and loyalty balances each migrate differently. Loyalty balances are the ones that turn into a support incident if they arrive wrong, and the person who understands them has usually left.
Are there subscriptions? A store with recurring orders is a different piece of software from a store without them. Say so early; it is the single largest fork in a Shopify estimate.
One market, or several? Multiple currencies, tax regimes and shipping rules multiply the states a checkout can be in, and each one needs testing rather than reasoning about.
Who signs off, and how quickly? This is the question that decides timelines, and it has nothing to do with engineering. A sprint waiting three days for a decision is a sprint you paid for and did not use. It is also the one thing on this list we cannot fix from our side.
When the bigger agency is the right purchase
Sometimes it is, and it is worth saying which cases, because the sprint model is not a universal answer.
If you need brand and creative direction rather than engineering, buy that from people who do it well; we will build against what they produce. If you need someone accountable inside your organisation for a twelve-month roadmap, with a project manager who sits in your stand-ups and a procurement-friendly contract, that is a retainer and an agency, not a sprint. If you need a 24/7 support SLA with penalties attached, same answer.
What we are good at is the bounded piece of work with a definable end: the migration, the speed problem, the integration, the build that needs to exist by a date. If that is what you have, paying a discovery phase to discover it is a tax on already knowing.
What we would tell you on the call
Most stores that ask us for a quote need less than they think in the first month and more than they think in the sixth. The first sprint is usually an inventory rather than a build, because it costs one week and it reorders the priority list almost every time.
You do not need to talk to us to start that. The app list with owners and monthly fees, and a measurement of how long the payment button takes to become interactive on a mid-range Android phone, are both free, and between them they answer most of what a discovery phase would have charged you to find out.
If you would rather we did it, the price is on the page and the call is thirty minutes.
