Skip to content

Now booking Shopify builds for Q4 2026.

Check availability

How to Choose a Shopify Development Agency (12 Questions, 8 Red Flags)

A buyer's guide to hiring a Shopify development agency in 2026. The questions that expose real expertise, what good answers sound like, and when to walk away.

How to Choose a Shopify Development Agency (12 Questions, 8 Red Flags)

A Shopify build is one of those purchases where you cannot evaluate the work until you have already paid for most of it. By the time you know whether the theme code is maintainable, whether the site is fast, and whether the agency documented anything, the money is spent and switching costs are real.

The vetting conversation is where your leverage sits. This guide gives you twelve questions that expose how an agency actually works rather than what services they list, what a strong answer sounds like versus one that should end the call, eight red flags, and what a good build process looks like week by week.

Decide three things before any call

  • What problem you are actually solving. "Our site looks dated" and "our mobile conversion rate is half our desktop rate" lead to completely different projects. The second one might not need a rebuild at all.
  • Your template count. Count the unique page types you need: homepage, product page, collection page, cart, and then everything else. This is the single most useful number in a scoping conversation and most buyers arrive without it.
  • What your content situation is. Do you have copy and photography ready, or does that need producing? Content readiness is the most common cause of timeline slip, and it is on your side of the line.

The 12 questions

1. "What would you not build for us?"

The most revealing question available, and it inverts the usual sales dynamic.

Good answer: they name something. "You probably do not need a custom theme; a premium theme with proper customization gets you most of the way." "Headless would be over-engineered at your revenue." "That feature is an app, not a build."

Walk away: everything you mentioned is possible and recommended. An agency that agrees with every request is quoting the most expensive version of your description.

Why this works: most stores do not need a custom theme. A modern Online Store 2.0 theme with thoughtful customization delivers roughly 90% of a custom theme's value at a fraction of the cost, and for most stores under $2M revenue headless is significantly over-engineered. An agency that will tell you this is one that will scope honestly on everything else too.

2. "Custom theme or premium theme, and why?"

This choice alone can explain a 5x price difference between two quotes for the same store. If you do not ask, you may be comparing quotes for genuinely different projects.

Good answer: a recommendation with reasoning tied to your specific merchandising needs, brand position, and maintenance capacity. They mention that custom themes do not receive theme updates.

Walk away: "custom is always better," or a custom theme recommended before they have seen your catalogue.

3. "How many unique templates does this quote cover?"

Good answer: a named list. "Homepage, product page, two collection page variants, cart, about, contact, FAQ, blog index, blog post. Nine templates."

Walk away: "a full store" or "everything you need." Template count is the primary scope unit on a Shopify build. Without it there is no scope, and everything becomes a change order.

4. "Show me live store URLs, not screenshots."

Then check them yourself:

  • View source and confirm it is actually on Shopify
  • Run PageSpeed Insights on a product page and read the field data
  • Browse it on your phone, add something to cart
  • Check whether the site uses Online Store 2.0 sections or looks like a legacy build

Walk away: portfolio screenshots only, sites that turn out not to be on Shopify, or sites where the agency's contribution is unclear. "We did the design, someone else built it" is a legitimate answer but changes what you are buying.

5. "How do you handle site speed?"

Core Web Vitals are a confirmed ranking signal, and a slow store pays twice: lower organic rankings, and higher paid CPCs because Quality Score factors landing page experience.

Good answer: they name a target (LCP under 2.5 seconds on mobile), they mention field data from the Chrome UX Report rather than lab scores, and they talk about app weight and theme bloat as the primary causes on Shopify. Bonus if they mention that a speed baseline at launch is part of handover.

Walk away: "we'll install a speed app." Apps that inject JavaScript to boost speed often work against INP. The durable fixes are structural theme-level work.

6. "What happens to SEO during this project?"

Critical for any migration or rebuild, and the most commonly under-quoted line item.

Good answer: for a migration, a 301 redirect map built before the build starts, title and meta preservation, canonical handling, Search Console transfer, and post-launch monitoring. For a rebuild, they mention not simplifying heading structure or removing internal links, and benchmarking pre-launch metrics.

Walk away: no mention of SEO in a migration quote. Stores that launch without a complete redirect map routinely lose 30-60% of organic traffic, and that is not automatically recoverable.

7. "Who actually writes the code?"

Good answer: named people, their roles, whether any work is subcontracted, and who your day-to-day contact is. If work is offshore, they say so upfront and explain the timezone overlap.

Walk away: evasiveness, or discovering mid-project that the senior developer in the pitch is not on your build.

8. "What is your change request process and rate?"

Scope always moves on a build. The question is what happens when it does.

Good answer: a defined process, a stated hourly or per-change rate, and a threshold below which small changes are absorbed. They mention that a signed specification is what change requests are measured against.

Walk away: "we're flexible" with no rate stated. That flexibility becomes an invoice.

9. "Do we own the code and the store?"

Good answer: unambiguously yes. Store ownership, theme code, domain, and any custom app built for you. Full admin access from day one, not at handover.

Walk away: any structure where the agency retains code ownership, hosts custom functionality on their own infrastructure, or holds the store owner account. This is a switching cost designed to trap you.

10. "What does handover include?"

Good answer: documentation of custom sections and how to use them, a recorded walkthrough, a list of apps installed and why, theme settings documentation, and a defined bug-fix window (14-30 days is standard).

Walk away: "we'll show you around on a call." Six months later, when you need to edit a custom section, that call is not something you can consult.

11. "What do you need from us, and by when?"

Weak agencies say they need very little. Good agencies give you a dated list.

Good answer: copy for every template, product photography at specified dimensions, brand assets, a named decision-maker, and a content deadline before build starts. They explain that development waits on content more often than the reverse.

Walk away: "we'll handle everything, you don't need to do anything." Then the project stalls in week four waiting for product copy nobody assigned.

12. "What happens after launch?"

Good answer: a defined bug-fix window, then a maintenance option with a stated scope and price. They mention that Shopify deprecates API versions and apps update, so 10-20% of build cost annually is a realistic maintenance budget.

Walk away: no post-launch plan, or a retainer pitched as mandatory without a defined scope.

The eight red flags

Red flag Why it matters
Quote before discovery, no template count There is no scope, so everything becomes a change order.
Portfolio screenshots instead of live URLs Screenshots prove nothing about build quality, speed, or authorship.
No SEO preservation in a migration quote Stores that skip redirect mapping lose 30-60% of organic traffic.
Agency retains code or store ownership A switching cost designed to trap you.
Custom theme recommended before seeing your catalogue The most expensive option, chosen before knowing whether it is needed.
Headless recommended below $2M revenue Significantly over-engineered, with ongoing engineering cost you cannot absorb.
Timeline that ignores content dependencies Every build that slips, slips here. An agency that does not name it has not managed many.
No stated change request rate "We're flexible" becomes an invoice.

On Shopify Partner and Plus Partner status

Worth understanding what these signals actually mean, because both are used heavily in agency marketing.

  • Shopify Partner. A baseline signal that the agency works on the platform and has access to development stores. The program is not difficult to join, so it is a weak differentiator on its own.
  • Shopify Plus Partner. Carries more weight, because it requires demonstrated Plus experience and Shopify vets it. If you are running a Plus build, this is a genuine filter.

Neither replaces the verification in question 4. Live store URLs at your complexity level tell you more than any badge does.

Agency, freelancer, or in-house

Best for Typical cost Watch out for
Freelancer Clear-scope work: theme customization, specific features, fixes $75-$200/hr, $500-$30,000/project Capacity limits, single point of failure, no QA layer
Agency Design and development integrated, migrations, integrations, brand-led builds $150-$300/hr, $3,000-$100,000+/project Junior execution behind a senior pitch; change-order creep
Offshore certified Partner Cost-sensitive builds with clear specifications 30-60% below US agency rates Timezone overlap, communication bandwidth, legal recourse
In-house developer Stores with continuous development needs at scale $70k-$140k/yr One person cannot cover design, development, and CRO well

The rough guide: freelancer for most stores doing $200k-$3M in revenue with clearly scoped work, agency when design and development need to be integrated or when brand experience is a material factor.

What a good build process looks like

Phase What should happen
Discovery (week 1) Template count agreed. Written specification produced and signed by both sides. Content requirements listed with dates. Baseline metrics recorded (conversion rate, Core Web Vitals, organic traffic by page).
Design (weeks 2-4) Mobile-first designs for every named template. Reviewed and approved before any code is written. This is where changes are cheap; after build starts they are not.
Build (weeks 4-8) Development on a duplicated theme, never the live one. Regular preview links. Sections built to be reusable and editable by you afterwards, not hardcoded.
QA (week 8-9) Full purchase path on real devices. Every payment method. Every shipping region. Tax calculation. Every third-party integration. Core Web Vitals check against the stated target.
Launch Redirects live and spot-checked. Sitemap submitted. Tracking verified firing. Old site kept accessible for reference.
Post-launch (14-30 days) Bug fixes included. Search Console monitored daily for the first fortnight. Conversion rate compared to baseline by device. Handover documentation delivered.

Two things worth insisting on. Design approval before build starts, because changes after build begins cost multiples of what they cost in Figma. And baseline metrics recorded before anything changes, because otherwise nobody can prove the new site performs better than the old one.

A scorecard for comparing finalists

Score each agency 1-5 on the same criteria rather than going on gut feel or price:

  • Willingness to scope you down (question 1)
  • Theme approach reasoning (question 2)
  • Scope clarity and template count (question 3)
  • Verified live work at your complexity (question 4)
  • Speed competence (question 5)
  • SEO handling (question 6)
  • Named people and seniority split (question 7)
  • Change process transparency (question 8)
  • Ownership terms (question 9)
  • Handover quality (question 10)
  • Price-to-scope match

Highest total, not lowest price. A cheap build that loses 40% of your organic traffic or arrives undocumented and unmaintainable costs multiples of what you saved.

Questions to ask yourself before signing

  • Do I know exactly how many templates I am paying for?
  • Do I know whether this is a custom theme or a customised premium theme?
  • Is my content ready, or will the build wait on me?
  • Do I have baseline metrics recorded so I can judge the result?
  • Do I own the store, the code, and the domain?
  • Do I know what happens on day 31 after launch?
  • Am I comparing scope, or comparing sticker prices?

How we work

If you are running a shortlist, we are happy to be on it and happy to answer every question above on the record, including live store URLs you can verify yourself.

Our default: scope by template count with a signed specification before build starts, premium theme unless a custom build is genuinely justified, LCP under 2.5 seconds on mobile as a launch target, redirect mapping before the build for any migration, and full store and code ownership with you from day one.

See what our Shopify development and design services cover and what a typical engagement looks like week by week.

Frequently Asked Questions

What is the biggest red flag when hiring a Shopify development agency?

A quote produced before any discovery, with no defined template count. Scope is the whole ballgame on a Shopify build; an agency that prices before knowing how many unique page types you need is either guessing or planning to bill the difference as change orders. Close behind: portfolio screenshots instead of live store URLs, and no mention of SEO preservation in a migration quote.

Should I hire a Shopify Partner specifically?

Partner status is a useful baseline signal but a weak differentiator; the program is not hard to join. What matters more is verified live work on Shopify at your complexity level, whether they build on Online Store 2.0 conventions, and whether they can explain what they would not build for you. Plus Partner status carries more weight because it requires demonstrated Plus experience.

How do I verify a Shopify agency's portfolio?

Ask for live store URLs, then check them yourself. Confirm the site is actually on Shopify (view source and look for Shopify markers), run PageSpeed Insights on a product page, and browse the store on your phone. Screenshots prove nothing about build quality, speed, or whether the agency actually did the work.

Should a Shopify agency own my theme code?

No. You should own the store, the theme code, the domain, and any custom app built for you, with full admin access from day one. Some agencies retain code ownership or host custom functionality on their own infrastructure, which makes leaving expensive. Get ownership in writing before work starts.

How long should a Shopify build take?

A premium theme with customization takes 2-4 weeks. A custom theme build takes 6-10 weeks. A Plus enterprise project takes 8-16 weeks. Headless takes 10-20 weeks. An agency quoting a full custom build in three weeks is either using a template they will call custom, or the timeline will slip.

Want a real number for your store?

Send us the URL and the catalog size. We will come back with a fixed quote and the template list it covers, no call required unless you want one.

Get a build quote

Ritik Verma

Founder and Managing Director, 7SEA Marketing™

Ritik founded 7SEA in 2020 and still reviews every scope before it goes out. He writes here about the parts of running an ecommerce agency that nobody publishes.