Most brands choose a Shopify Plus agency the wrong way. They shortlist on portfolio polish and hourly rate, sign with whoever demos best, and find out six months in that the team who sold the work isn't the team building it.
At $5M to $50M in GMV, and especially as you scale past it, the wrong choice is expensive, and you live with it for years. A fragile build slows every launch that follows. An over-customized theme becomes the thing your next agency has to unwind. A migration scoped without a post-launch roadmap stalls right when revenue should be accelerating.
Choosing well comes down to a handful of things that are easy to verify and hard to fake.
That last part matters, so this guide is built around it. Five criteria, and each one ends with a single test you can run in a discovery call or a reference check. If a criterion cannot be verified, it is a claim. The tests are how you tell the two apart.
Before you shortlist: know what you're buying
Timing comes first, because most brands get it wrong. Bring the agency in 90 to 120 days ahead of your target launch date, ideally while the Shopify Plus deal is still being structured. Implementation scope and post-launch roadmap should inform your Plus contract, not arrive after it. An agency engaged too late is executing against constraints they had no input on, and you are paying to work around decisions that were already locked.
Before you talk to anyone, write down what you are actually commissioning. Not "a new site," but a one-page scope memo: what launches, what migrates, what integrates, the one number the build has to move, and the date it has to hit. If that memo does not fit on a page, you are not ready to shortlist yet. The more precisely you can state your scope, the more honestly an agency can respond to it, and the faster you can tell a real answer from a pitch.
That single number matters as much as the scope. Are you protecting subscription revenue through a platform migration? Unlocking B2B without rebuilding the consumer experience? Recovering organic traffic by fixing load time? The KPI determines which agency is the right fit, and it gives you something concrete to hold them to after launch.
1. Confirm who actually builds it
Ask early: are the people who scope the architecture the same people who ship it? Plenty of agencies sell with senior strategists and deliver with junior developers who inherit your account on day one. A senior-only team costs more per hour and less per outcome, because the decisions that shape the build get made by people who have done it before.
Verify the partner tier, then look past it
Check Shopify's official partner directory, not the agency's homepage. The directory is the authoritative source for an agency's current tier.
Shopify's partner program has five tiers: Registered, Select, Plus, Premier, and Platinum. Registered is the entry point; every tier above it is earned through demonstrated commercial impact, meaning the revenue an agency drives for merchants and the number of merchants it brings to Shopify. A higher tier reflects real volume and complexity of work, which is a genuine signal. But partner status is the floor of your evaluation, not the decision. Two agencies at the same tier can differ enormously in the depth of their team, the range of builds they have actually shipped, and who is on your account after kickoff.
The tiers also map to the kind of merchant an agency serves. Shopify defines Platinum around enterprise merchants and Premier around larger, more complex merchant environments. In practice, Platinum agencies are built for enterprise accounts, with the teams and overhead that scale requires. That is exactly right for an enterprise brand, and usually more agency than a $5M to $50M brand needs or wants to pay for. The useful question is not what the highest tier is, but whose everyday work looks like your build.
Ask for named results at your revenue tier
Logos are not evidence. Ask for named clients at your GMV tier, specific outcomes, and contact information for two of them. If the agency demurs, that tells you something.
Set the standard, then verify the agency meets it. Softlimit, a Shopify Premier Partner with a senior team that has built on Shopify since 2011, publishes its results because that is the standard this guide describes. For The Perfect Jean, Softlimit engineered a custom product detail page managing 23,000 variants in a single seamless experience, and order volume rose 200%. For Polyvinyl Records, Softlimit migrated a 15-year custom platform to a unified DTC and B2B Shopify store, and post-migration revenue increased 20%. Named clients, specific numbers, both verifiable. Shopify documented the Polyvinyl migration in its own case study.
How to verify: Ask for the names of the developers who will build your project, in the first call. Then ask the two reference clients whether those names match who actually showed up.
2. Native-first architecture over tech sprawl
The best Shopify Plus agencies are not the ones who offer the longest list of integrations. They are the ones who will talk you out of adding another app when a native Shopify capability does the job.
What native-first means in practice
Native-first means building on Shopify's own infrastructure (B2B, Markets, Checkout Extensibility, Shopify Functions) before reaching for third-party apps. It means consolidating your tech stack rather than expanding it. Every app you add is another monthly fee, another thing that can break during a Shopify update, and another piece of custom code your next developer has to understand.
The practical translation: the cleaner the build going in, the faster every change you want to make after launch. An over-customized theme does not just cost more to build. It costs more to maintain, more to update, and often requires a full rebuild sooner than it should.
That is why the sharpest question to ask a prospective agency is not "what apps do you recommend?" but "which apps would you push back on, and why?" An agency that cannot name any is not thinking about your build. It is filling it.
The largest version of this trap is headless. Headless commerce is the right call for a narrow set of brands, and it gets sold to far more than actually need it. An agency that leads with a headless pitch before it understands your catalog, your team, and your roadmap is selling complexity you pay to build and pay again to maintain. We regularly see brands come back to native Shopify after a headless build that never fit their business, once the rebuild cost and the day-to-day friction catch up with them. So when an agency proposes headless, ask them to name exactly what your brand needs that native Shopify cannot do. If the answer is vague, the answer is native.
The migration test
How an agency talks about data migration tells you whether they have actually done it at scale. A well-scoped migration covers order history, customer accounts, redirect planning for every URL that carries organic traffic, and, if you run subscriptions, continuity through the billing cutover. Migrating years of order history without breaking customer accounts is one of the places a Plus build quietly goes wrong.
If you run subscriptions, ask how they will move them. The billing mechanics usually live in the subscription platform itself, Recharge for example, and in whether your payment tokens transfer cleanly. So what you are actually testing is whether the agency knows that path cold: how active subscribers, billing dates, and saved payment methods carry over, and how the cutover is sequenced so no one gets double-charged or dropped. "The app handles it" is half the answer. The other half is the plan around it, and that is where these projects go wrong. Softlimit migrated Moon Juice to Recharge in 25 days against a hard contract deadline, because the migration was scoped as a revenue-protection problem from the start, not a technical task handed off at the end.
How to verify: Ask the agency to name one app they would remove from your current stack and one they would refuse to add. A real answer is specific and comes with a reason; a filler answer names nothing.
3. A real post-launch revenue roadmap
Most agencies treat post-launch as a support retainer: bug fixes, Shopify updates, an SLA that tells you how fast they will respond when something breaks. That is a reasonable service, but it is not a revenue plan, and the difference is where the build pays off or doesn't.
Ask the agency what the first 90 days after launch look like and what number they are accountable for. A strong answer involves CRO work planned before launch, a hypothesis about where conversion or AOV can move, and a named person who owns the outcome. A weak answer involves a monthly hours bucket.
The distinction is not abstract. Softlimit relaunched Verb in 8 weeks with a 52% faster storefront. Rebuy bundles and upsells, planned and integrated before launch, drove 69X ROI and a 14% AOV lift within the first 30 days. That is not a maintenance story. It is a revenue roadmap executed from day one.
If you are not sure what your current build is actually capable of, start with an audit before you commit to a full redesign or migration.
How to verify: Ask who owns your revenue number in the 90 days after launch, by name. If the answer is a role, a team, or an hours allotment instead of a person, there is no owner.
4. Whether they build for agentic commerce, not just human UX
Most agencies are not thinking about this one yet, and in 2026 that is starting to matter. Buyers no longer arrive only through a browser. They arrive through AI assistants that read your store, compare it, and increasingly transact on a shopper's behalf. An agency still designing purely for the human clicking through a product page is building for the traffic you have, not the traffic that is coming.
Agentic readiness is not a feature you bolt on. It is the same native-first thesis from the last section, pointed at a new reader. An AI agent reads structured data, clean product data, and Shopify's own agentic capabilities. A store built on sprawl and over-customization is as hard for an agent to parse as it is for your next developer to maintain. The clean build wins twice: once for the human, once for the machine.
So ask a prospective agency how they make a store legible to AI agents, not just to shoppers. A serious answer talks about structured product data, machine-readable content, and Shopify's agentic surfaces. A vague answer talks about "SEO" and moves on.
Softlimit builds for this on its own site, not just its clients'. We maintain a live agents.md file and an agentic-discovery sitemap, both public and verifiable, so AI systems can read who we are and what we do in a structured form. The space is early and we are not going to claim a finished playbook, because honest agencies say when something is still taking shape. But an agency that has not started building for agentic discovery on its own property has not started thinking about it for yours.
How to verify: Ask the agency to show you where an AI agent would read their own store or site today. If they can point to something structured and live, they are building for it. If they cannot, they are talking about it.
5. Whether they'll tell you no
This is the cheapest test of an agency's seniority, and almost nobody administers it.
Ask the agency who they are not a fit for. Ask what would make your project a bad engagement from their side. Ask whether they have ever pushed back on a client's scope or timeline, and what happened.
An agency with no answer to those questions has no boundary. That is not humility. It is a signal that they will agree to scope, timeline, and budget in the room and resolve the tension later on your dime.
Softlimit's own non-fit list: theme-tweak-only projects, sub-$3M early-stage brands, and vendor shopping without a genuine platform commitment. If that describes you right now, the better move is usually a strong off-the-shelf theme and a trusted freelancer for the tweaks, then a real agency engagement once your GMV and roadmap justify a custom build. Publishing that list costs leads. It also means the leads that remain are correctly qualified.
The same principle applies when an agency reviews your brief. If every estimate comes back right at your stated budget, and nobody flags technical risk or asks what happens if the ERP integration takes longer than scoped, that is a yellow flag. The agencies that have shipped at your complexity level know where the surprises live. They will tell you.
How to verify: Ask the agency to describe the last project they turned down and why. A senior team has a recent, specific answer. A team without a boundary will reach for a hypothetical.
How to check a reference (most buyers waste the call)
You asked for two named clients under criterion 1. What you ask them decides whether the call is worth anything. Three questions cut through the courtesy:
- What broke in the first month after launch, and how fast did they fix it? Every launch has something. The answer tells you how the agency behaves when the pressure is real, not in the pitch.
- Were the people in the pitch the same people in the project channel? This is criterion 1, confirmed by someone with nothing to sell you.
- Did you use the same agency for your next build? The most honest review a client can give is whether they went back.
What it should cost and how long it should take
On timeline: a mid-market Plus build or migration typically runs 12 to 16 weeks from kickoff to launch. Scope is the variable: a focused build with limited integrations sits nearer 12 weeks; a migration carrying B2B, subscriptions, and ERP connectivity runs to 16 or beyond. Faster happens under pressure, and rarely without tradeoffs. Softlimit relaunched Verb in 8 weeks and migrated Moon Juice to Recharge in 25 days, but both ran against hard deadlines on tightly constrained scope. Treat those as compressed exceptions, not the timeline you plan around.
On cost: a serious Shopify Plus build or migration commonly runs from the mid five figures into six figures. What moves a build up that range is complexity, not polish: the number of variants, whether you run B2B alongside DTC, subscription billing, ERP or 3PL integrations, and how much custom merchandising logic the store needs. A tightly scoped theme refresh sits near the floor, and the smallest engagements can start around $35K. A migration carrying subscriptions, B2B, and live integrations sits at the top of the range or past it. Ongoing retainers for strategy, CRO, and development support typically start at $5K per month on a 25-hour floor. Price the build by what it has to do, not by the number you hoped to spend.
The cheapest hourly rate is usually the most expensive decision. Saving on rate while losing organic traffic to missed redirects, or losing subscribers to a botched billing cutover, is the classic mid-market failure mode. The cost of the wrong agency is paid in the build that follows.
Questions to ask before you sign
Use this list in the discovery call. The answers will tell you more than the portfolio.
- Are the people who scope and architect the build the same people who build and ship it? What is the ratio of senior to junior developers on a project like mine?
- How many Shopify Plus builds have you done at my revenue tier? Can I speak to two of those clients directly?
- Who is my named project lead, and what else are they working on while my project is in flight?
- Do you subcontract any development work? If so, which parts, and how is that managed?
- What do I own at handover: theme code, app configurations, documentation, access credentials?
- Which apps would you push back on, and which would you refuse to add? Walk me through how you handle data migration, redirect planning, and, if I run subscriptions, billing continuity through the cutover.
- What does the first 90 days after launch look like, and what performance metric are you accountable for, by name?
- How do you make a store legible to AI agents, not just to shoppers?
- What would make this project a bad engagement for you? What kind of client are you not the right fit for?
That last question is the wedge. An agency that cannot answer it has no boundary. An agency that answers it honestly is worth talking to further.
Red flags
Watch for these in early conversations and in any proposal.
A proposal that arrives before a discovery call. If they priced it before they understood it, they guessed.
Every case study shows a screenshot and no number. Proof is a result, not a render.
They never mention what could go wrong. Every complex build has risk. The agencies that have done it know where it is. The ones that don't will find out on your project.
Frequently asked questions
What is a Shopify Premier Partner?
Shopify Premier Partner is the second-highest of Shopify's five partner tiers, which run Registered, Select, Plus, Premier, and Platinum. Every tier above Registered is earned through demonstrated commercial impact: referred and managed merchant revenue and multiple Shopify Plus or Enterprise engagements. It is a meaningful signal of an agency's track record and the scale of work it handles, though not the only thing to weigh. Verify any agency's current tier in Shopify's official partner directory before your first call.
How much does a Shopify Plus agency cost in 2026?
A serious mid-market Plus build or migration commonly runs from the mid five figures into six figures depending on scope complexity: variant architecture, B2B requirements, subscription infrastructure, and integration depth. Tightly scoped engagements can start around $35K. Ongoing development and strategy retainers typically start at $5K per month. Expect the higher end of that range for a build that includes migration, subscriptions, and B2B.
How long does a Shopify Plus build or migration take?
The typical range for a mid-market build or full migration is 12 to 16 weeks from kickoff to launch. Scope drives the timeline: a focused build is faster; a migration involving B2B, subscriptions, and ERP integrations takes longer. Very short timelines happen only under hard deadline pressure with tightly constrained scope. Softlimit migrated Moon Juice to Recharge in 25 days against a contract deadline, but that reflects a narrow, constrained scope, not a typical full-platform migration.
When should I bring the agency in: before or after signing with Shopify Plus?
Before. Ideally 90 to 120 days before your target launch, and while the Plus contract is still being negotiated. The implementation scope and your post-launch roadmap should inform the Plus deal, not be built around a contract you already signed. Agencies engaged early can flag what the build actually requires and help you structure the Plus engagement accordingly.
What does it mean for a Shopify store to be ready for agentic commerce?
It means the store is legible to AI agents, not just to human shoppers. That comes down to structured product data, machine-readable content, and building on Shopify's native agentic capabilities rather than on app sprawl that an agent cannot parse. The same clean, native-first build that is fast for a human to use and cheap for a developer to maintain is also the one an AI agent can read and act on.
Working with Softlimit
Softlimit is a Shopify Premier Partner, the second-highest of Shopify's five partner tiers, a level Shopify awards for multi-million-dollar merchant impact and repeated Shopify Plus and Enterprise engagements. We are based in New York, NY, and have built on Shopify for 15 years, working with mid-market DTC brands in lifestyle, consumer goods, pet, and music. Our client relationships often run past ten years. Our senior team works with roughly 16 retained clients and carries 4 to 5 concurrent Plus-level builds at any time.
The results in this guide (The Perfect Jean's 200% order volume increase, Verb's 69X ROI and 14% AOV lift in 30 days, Moon Juice's 25-day migration, Polyvinyl's 20% post-migration revenue increase) are from those engagements. Not unnamed clients. Not rounded figures. And every criterion above comes with a test you can run before you sign, which is the standard we hold ourselves to.
If the criteria in this guide describe what you are looking for, let's talk.
Let's Talk Shop(ify).