Shopify vs custom store for artist merch

Shopify wins the store when artist merch needs a real catalog, checkout, tax, and shipping. A custom storefront wins when the merch is the brand film — the drop, the lookbook, the object — not a themed product grid wearing a wordmark. Most working artists should not pick a side: keep Shopify as the commerce engine and put a Shopify cart on a custom artist site. This spoke sits under Websites That Feel Like Films. The craft bar is the film. The cart is plumbing.

The short answer

• Use Shopify (Online Store or a cart behind your site) when you sell variants, run drops, collect tax, or ship physical goods.
• Build a custom storefront when merch is a destination — cinematic product pages, drop UX, photography that cannot live inside a default theme — and you can staff the engineering.
• Hybrid is the default: artist site owns identity; Shopify owns SKUs, checkout, tax, shipping, refunds.
• Headless Shopify (Hydrogen / Storefront API) customizes the storefront, not the payments stack. Checkout stays Shopify-hosted.
• A Stripe form on Framer is not a store. It is a payment link that will lie about sizes, tax, and sold-out.

What people mix up when they say “custom store”

Three different builds get called “custom.” They are not the same job, the same bill, or the same failure radius.

What they said	What they actually mean	Who owns checkout	Who owns inventory
“Custom Shopify store”	Paid or commissioned Online Store theme	Shopify Checkout	Shopify admin
“Headless / Hydrogen store”	Custom UI on the Storefront API	Still Shopify-hosted checkout	Shopify admin
“Custom store, no Shopify”	Stripe (or similar) + your database + your shipping math	You	You, until you forget

If the pitch is “we will make checkout ours,” ask which row they are in. Shopify’s own Hydrogen migration notes say customers are always directed back to a Shopify-hosted checkout whether the storefront is Liquid or Hydrogen. You can brand a checkout subdomain. You do not replace the checkout binary.

A custom theme is still a Shopify store. A custom site with Buy buttons is still a Shopify store. A custom cart you wrote is a new company you did not mean to start.

Name the row before you name the budget.

When Shopify should own catalog, checkout, tax, and shipping

Pick Shopify as the store engine when merch is commerce, not a poster with a button.

• More than a couple of SKUs, or any size/color matrix
• You need receipts, refunds, discount codes, or gift cards
• A manager must change stock on a Tuesday without a designer deploy
• You ship physical goods and need real rates or honest flats
• You collect sales tax in more than “we guessed 8%”
• Paid traffic will land on a product URL, not a mood film
• You will sell the same object from more than one door (site, email, ads, a merch table)

Shopify’s tax and shipping admin exist so those rules are not hardcoded in a Webflow CMS field. Inventory tracking exists so quantity is a number, not a paragraph.

Stay on a payment link or a tiny on-site cart only when all of this is true: one SKU, almost no sizes, almost no volume, nobody needs codes, and you can live with “email us for the last medium.”

Shopify setup that actually makes it a store — not a theme demo:

1. Create the product with the real size/color variants. Do not hide variants in the title string.
2. Turn inventory tracking on per variant before you publish a number (inventory setup).
3. Decide where you must collect tax, then configure Taxes and duties. Do not type a percentage into the description.
4. Add shipping rates you can honor — real carrier rates or an honest flat. Print-on-demand lead times belong on the merch chapter, not as “ships tomorrow.”
5. Place a test order to a real ship-to. Read the receipt. Refund it in admin.
6. Hand the manager Shopify login. If they still ping the designer to hide a size, the handoff failed.

Job	Shopify admin	Artist-site CMS
Price, variants, discounts	Yes	Photos and copy only
Quantity	Yes	Sold-out label at most — never a second count
Tax / shipping math	Yes	Plain-language policy, matching the admin
Refund	Yes	No

The day a drop needs a quantity lock at midnight, a marketing-site cart becomes the outage.

When the merch is the brand film

Custom storefront work earns its keep when the shop is the identity surface — not when you are embarrassed by Dawn’s defaults.

Commission custom storefront craft when most of these are true:

Signal	Theme + cleaned Shopify is enough	Custom storefront is in play
Job of /merch	Show 3–8 objects, send fans to buy	The page is the campaign
Motion / lookbook	Static cards, strong photography	Scroll-tied film, sequenced drops, object-as-world
Catalog size	Small, manager-edited	Large, filtered, or tied to a content system themes fight
Staffing	No one to own a React app	Someone on call for the storefront
Checkout complaint	“The theme looks rented”	“The browse experience cannot be a theme”

Oliver Malcolm– or Red Leather–shaped sites can feel like a world on the apex domain and still clear money through Shopify. The custom work belongs on the film: type, photography, motion, the merch chapter. The custom work does not belong on rewriting tax.

Procedure if you think you need custom:

1. Write the job of the merch surface in one sentence. If it is “sell the tour tee,” you need a lookbook plus a cart, not Hydrogen.
2. Screenshot the three interactions a theme cannot do. If you cannot name three, you are buying taste, not architecture.
3. Confirm checkout, tax, and shipping stay on Shopify even if the UI is custom.
4. Price the engineering like a five-figure custom site, not like a theme install.
5. Name the maintainer. If the maintainer is “the designer, sometimes,” stop.

What “merch is the film” looks like in practice — so you do not use the phrase as cover for a theme allergy:

• The object is the campaign. The tee is not a grid cell; it is a still from the world.
• Browse has a sequence: trailer, lookbook, size, buy. A theme collection page cannot play that sequence without looking bolted on.
• Product photography matches tour light, grit, and crop. Amazon-white packs on a film homepage are the cut fans feel.
• Sold-out is a designed beat (“join the list”), not a theme empty state.
• You would still build this page if checkout were a separate hop. If the only reason you want custom is to hide Shopify’s chrome, you want a better theme and a lookbook — not a new stack.

A film-grade homepage next to a lazy shop trains fans to trust the music and distrust the object. That is a brand cut. It is not a reason to throw away the engine.

Hybrid: Shopify cart on the artist site

Hybrid is the architecture I ship most: the artist site is the film; Shopify is the cart; the merch chapter is a lookbook with doors into that cart.

Bridge	What it is	Use when
Product URLs	Link from branded cards to Shopify product pages	Size and color need a real picker
Cart permalinks	URL that lands in cart with variants loaded	SMS, email, poster QR, “this exact tee”
Buy Button	Embeddable product / cart snippet	One featured drop on a campaign page
Headless cart UI	Storefront API cart on your domain, then Shopify checkout	You have eng budget and a reason the lookbook cannot deep-link

Most catalogs should start on deep links. Embeds get heavy next to motion. Headless is a later hire.

The artist site can live in Framer, Webflow, or custom — pick that stack for the film, not for the warehouse. Framer vs Webflow vs custom is the marketing-stack chooser. This post is the merch-engine chooser. Do not make Framer pretend it is Shopify.

Hybrid rules that keep you honest:

• One catalog in Shopify. The CMS holds photos and copy, not quantity.
• /merch is curated. “View all” may go to Shopify if the catalog is large.
• Apex domain stays the artist site. Shop may be shop. or checkout on Shopify.
• Sold-out is designed on the artist site, not a dead button into a generic empty cart.
• The manager’s Tuesday tool is Shopify admin, not a deploy.

Launch sequence for hybrid (the one I actually run):

1. Shopify product live, tracking on, test order refunded.
2. Artist-site /merch with three to eight cards. Price visible without opening Shopify.
3. Each card CTA is a product URL (size matters) or a permalink (exact variant).
4. Footer / nav still lead with music and tour. Merch is a peer, not the apex.
5. Email and bio use the same URL as the site CTA. No third “buy here.”
6. Ads, if any, land on the product or permalink — never the homepage film.
7. Sold-out copy is in the CMS as a state, not as a quantity clone.
8. After first live sale, confirm the order, tax, and shipping on the receipt match the merch-chapter copy.

If hybrid feels like a compromise, you are still thinking in platforms. It is a split of jobs.

How this choice affects conversion

I will not cite a conversion lift I did not measure, and I will not dress a vendor case study up as a Spurlock Studios result. Conversion here is mechanics.

Mechanic	Shopify theme store	Hybrid (site + Shopify cart)	Fully custom cart (no Shopify)
Extra hops	Browse and buy in one theme	Lookbook → product or permalink → checkout	Whatever you built, including the bugs
Brand trust before pay	Weak if the theme looks rented	Strong if the film is good and the door is obvious	Strong until tax, shipping, or sold-out lies
Size / variant path	Theme picker, usually fine	Must be obvious on the card or on Shopify	Easy to ship a pretty grid with a broken size state
Mobile	Theme quality varies; test it	Test the permalink on a phone, not the desktop customizer	You own every tap
Ads	Land on product URL	Land on product URL or permalink — not the homepage	Land on a URL that can complete a buy
Refunds / chargebacks	Shopify order object	Shopify order object	Your inbox

What actually kills merch conversion on artist sites, from hundreds of production sites — not from a fabricated GMV table:

1. The shop looks like a different company than the homepage.
2. The Buy path is a scavenger hunt through Music and Tour.
3. Size is unclear until after a second hop, then the size is gone.
4. Shipping copy says “tomorrow” on a print-on-demand tee.
5. Paid traffic lands on a cinematic homepage that cannot check out.

Fix those before you buy Hydrogen. A custom storefront on top of a lying size chart is a more expensive liar.

Diagnostic if merch “doesn’t convert” (run before you change platforms):

1. Click the live Buy path on a phone. Count taps to a size. More than two is a craft bug, not a Shopify tax.
2. Load the ad destination. If it is the homepage, you bought a film with a store buried in it.
3. Compare product photography to the homepage stills. If they do not share a world, fans feel a scam even when checkout is fine.
4. Attempt a refund as the manager. If that requires you, ops will leak into DMs.
5. Check whether two URLs still sell the same object. Duplicate doors with stale sizes will beat any storefront.

Measure the path, not the platform logo. If fans cannot go from “I want that hoodie” to a variant in one or two honest taps, the architecture is already losing — theme, hybrid, or custom.

Theme vs Hydrogen vs a cart you built

These are the three storefront builds. Pick with staffing, not with mood boards.

Build	Stack	Ships when	Breaks when
Online Store theme	Liquid, Shopify-hosted	You need a catalog live this month	You need browse UX a theme cannot express
Hydrogen / headless	React storefront, Storefront API, Oxygen or your host	You have a front-end team and a real UX ceiling	Nobody owns deploys, apps that assumed Liquid stop working, checkout still is not “yours”
Fully custom cart	Stripe + CMS + your tax table	You sell one object and can tolerate ops pain	Second SKU, first nexus question, first oversold drop

Shopify documents the Storefront API as the layer for products, collections, cart, and contextual pricing — commerce primitives, not a weekend plugin. Hydrogen is Shopify’s React path onto that API. Neither replaces admin, inventory, or hosted checkout.

App reality: a lot of Shopify apps assume the Online Store. Headless keeps the ones that speak APIs and drops the ones that inject Liquid. Budget that audit. It is not free.

Decision order:

1. Can a cleaned theme plus a branded /merch chapter do the job this quarter? If yes, stop.
2. Do you need custom browse on your domain, with Shopify still the engine? Then headless is a conversation, not a default.
3. Do you want to own tax, shipping, fraud, receipts, and refunds yourselves? Then you are starting a store company. Most artist teams should not.

Headless audit before you commit — if you cannot check these, you are not buying Hydrogen. You are buying a demo:

• Named engineer (or studio retainer) who ships storefront deploys
• List of Shopify apps you actually use; each one marked “works via API” or “Liquid-only, replace”
• Checkout subdomain plan per Shopify’s Hydrogen migration notes
• Cart → Shopify checkout handoff tested on mobile, including discount codes
• Product publishing still happens in Shopify admin, not in the React repo
• Analytics: you can still attribute orders to the headless channel
• A rollback: Online Store can take traffic if the custom storefront fails on drop night

Hydrogen is not “the expensive theme.” It is a second application. Staff it like one.

What custom still cannot replace at checkout

Custom storefront is a browse decision. Checkout is a compliance and trust decision.

Surface	Theme store	Hydrogen / Storefront API	Fully custom (no Shopify)
Product grid / lookbook	Theme sections	You render it	You render it
Cart	Theme cart	Your cart UI via API	Your cart UI
Checkout	Shopify-hosted	Shopify-hosted (same handoff)	You
Tax engine	Shopify Taxes and duties	Same Shopify admin	You, your accountant, your bugs
Shipping rates	Shopify shipping	Same	You
Card processing	Shopify Payments or a gateway Shopify surcharges	Same	Stripe or whoever you wired
Checkout UI extras	Checkout extensibility — deeper targets generally need Shopify Plus	Same extension model, not Hydrogen components	Whatever you invented

Shopify’s checkout-technology docs (as of 2026) put most Checkout UI extensions on Shopify Plus, with thank-you / order-status extensions available more broadly except Starter. If a designer promises “custom checkout” on Basic, they are promising branding and maybe a thank-you tweak, not a rewritten payments flow. Confirm on the live checkout docs, not a 2023 tutorial about checkout.liquid.

You can set a branded checkout subdomain. You cannot Liquid-rewrite hosted checkout the old Plus way and call it done — Shopify moved that work onto extensions.

Checkout decision list (say no in this order):

1. Do we need custom browse? If no, stay on theme + hybrid. Stop talking about checkout.
2. Do we need a thank-you or order-status extra? Many of those exist below Plus — check the live matrix, then use an extension, not a new cart.
3. Do we need in-checkout UI (upsells, custom fields, validation)? That is generally a Plus conversation. Price Plus as rent, not as “the film.”
4. Do we need to replace hosted checkout? Then we are leaving Shopify. That is a company, not a merch chapter.

If the merch film needs custom, buy custom browse. If someone is selling you custom checkout to dodge a theme, they are selling a liability.

Sourced fee bands vs five-figure custom

I am not going to invent a GMV ladder or a “our artists make X after fees” table. Vendor numbers below are Shopify’s public US pricing table as of September 2026, labeled as a snapshot. Verify on Shopify’s pricing page and the pricing-plans overview the day you sign. Country, card type, and in-person rates differ.

Cost line	Shopify store (US public table, Sept 2026 snapshot)	Custom storefront on Shopify	Fully custom cart
Platform rent	Basic $39/mo monthly or $29/mo billed yearly; Grow $105 / $79; Advanced $399 / $299; Plus starts at $2,300/mo	Same Shopify plan, plus hosting for Hydrogen if not on Shopify’s host	Hosting + whatever you run
Card processing	Shopify Payments online standard cards on that table: 2.9% + 30¢ (Basic) down to 2.5% + 30¢ (Advanced); Plus listed as “most competitive”	Same	Stripe (or other) published rates — look them up; do not copy a tweet
Extra if you skip Shopify Payments	Third-party transaction fee on that table: 2% / 1% / 0.6% / 0.2% by plan (fees article)	Same	N/A — you already left
Theme / storefront labor	Free or paid theme; optional theme dev	Five-figure custom site territory for a real headless merch UI	Five figures and then some, because you also built ops
Apps / POD / shipping	Separate bills. Shopify is not free shipping.	Same, plus apps that do not run headless	You buy equivalents one by one

Hedges that matter:

• Those percentages are not Spurlock Studios results. They are vendor list prices from a public table that moves.
• Premium cards, international cards, and POS rates are separate lines on the same pricing page. I am not reprinting every adder here.
• Print-on-demand and carriers bill you whether the storefront is Dawn or Hydrogen.
• I am not your accountant. Nexus is a legal question. Shopify calculates tax after you turn collection on — it does not decide whether you must.

Artist napkin (illustrative, not a quote):

1. Write the retail price of the tee.
2. Subtract blanks, print, and pack from your vendor — their number, not a blog average.
3. Subtract shipping you eat, or pass it through honestly.
4. Subtract the live Shopify Payments rate for your plan and country.
5. Subtract subscription divided by expected monthly units.
6. What remains is the merch line, or a loss you chose for the photo.

If that leftover is embarrassing, raise the price or kill the SKU. Do not “make it up on volume” with a three-city club tour.

Staff cost is the line people skip while arguing about 2.9% + 30¢:

Role	Theme + hybrid	Headless on Shopify	Fully custom cart
Manager	Shopify admin	Shopify admin + “is the storefront deploy green?”	Admin you built, plus you
Designer	Artist site + product photos	Same, plus storefront UI tickets	Same, plus every cart state
Engineer	Rare (theme tweaks)	Required	Required, forever
Accountant	Shopify tax reports after setup	Same	Spreadsheets until you rebuild reports

A theme subscription argument cannot beat an engineer who is not on the roster. If the band cannot retain storefront engineering, hybrid wins even when the theme is imperfect.

Custom browse is a craft cost. Shopify fees are a commerce cost. Do not mash them into one “Shopify is expensive so we should build a store” sentence. That sentence is how you fund a second tax engine.

What to ask a designer about this

A designer who only talks themes is not answering the merch question. A designer who only talks “fully custom checkout” is not answering it either. Take this list to the call.

• Which of the three builds are we actually buying — theme, headless on Shopify, or a cart outside Shopify?
• Where does inventory live on a Tuesday, and who edits it without a deploy?
• Where does tax get calculated, and who configured the regions?
• What does sold-out look like on the artist site and in the cart?
• Do ads land on a product URL or permalink, or on the homepage?
• If we use Shopify, is checkout still Shopify-hosted even if the lookbook is custom?
• Which Shopify plan do checkout extensions actually require for the extras in the deck?
• What is the exit — theme files, a Hydrogen repo, or a Stripe dashboard we cannot leave?
• Who is on call when the drop sells out at 12:04 a.m.?

Score answers, not vibes:

Answer you hear	What it usually means
“We’ll just use Stripe on the site”	No inventory brain unless they also built one
“We’ll make Shopify not look like Shopify”	Theme craft or hybrid. Good. Ask to see sold-out and mobile checkout.
“We’ll go headless so checkout is ours”	They have not read Shopify’s checkout handoff. Push back.
“Framer has a shop”	Fine for a poster. Fatal for a size run.
“You need Plus for the film”	Maybe for checkout extensions. Not for a lookbook. Separate the bills.

If they cannot draw engine vs surface on a napkin, they will ship two catalogs. You will refund sizes that do not exist.

Failure mode: a beautiful cart that cannot tax or ship

What breaks: the team hates a generic Shopify theme, so they build a Framer (or custom) merch grid with Stripe Checkout because it “matches the brand.” Quantity lives in a CMS field. Tax is a sentence in the footer. Shipping is “2–3 weeks” copied from last year’s drop. Midnight hits. Two fans buy the last medium. There is no inventory lock. The manager refunds from Venmo. The next drop repeats the story with better photography.

What it costs: chargebacks, a merch table that feels amateur next to the music, and a week of DMs that should have been a Shopify order object.

What you do instead:

1. Put the catalog in Shopify. Turn inventory tracking on before you announce a number.
2. Keep the film on the artist site. Point Buy at a product URL or cart permalink.
3. Configure tax for the regions you actually ship to, after you know you must collect — not as a theme decoration.
4. Set shipping rates you can honor. If it is print-on-demand, say so on the merch chapter.
5. Design sold-out on the site before the drop, not after the 404.
6. Refund inside Shopify. Stop mixing Venmo into a store.

Cleanup order if you already shipped the pretty cart:

1. Freeze new SKUs on the fake store.
2. Recreate each live object as a Shopify product with variants and tracking.
3. Replace every Buy URL (site, bio, email, ads) with the Shopify product or permalink.
4. Delete CMS quantity fields.
5. Run a test order in a real tax region and a real ship-to. Read the receipt.
6. Only then talk about Hydrogen.

The failure is not “we used Shopify.” The failure is “we used taste as a substitute for a store.”

Drop night: which architecture actually holds

Drops are the stress test. Evergreen tees hide bad architecture. Midnight does not.

Clock	Theme store	Hybrid	Headless	Custom cart (no Shopify)
T-24h	Inventory capped; theme product URL live	Same SKU; lookbook + permalink QA on phones	Storefront deploy frozen; checkout subdomain confirmed	You are the freeze. Hope the tax table is right.
T-0 announce	One URL in email, bio, stories	Same URL as /merch CTA	Same, plus “is Oxygen/Vercel green?”	Same, plus “is our server green?”
T+2 min	Quantity is Shopify law	Site sold-out still showing Buy is the bug to watch	API lag vs admin — know who checks	Race conditions you wrote
T+15 min	Refunds in Shopify	Refunds in Shopify	Refunds in Shopify	Inbox
Sold out	Theme empty + you should have designed it	Artist-site state must flip	Custom empty state must read admin, not a cached page	Whatever you remembered to code

Drop-night procedure (architecture-agnostic, engine = Shopify):

1. Cap inventory. The number in admin is the promise.
2. One buy URL everywhere. Print it. If two URLs exist, one is stale.
3. Two people: one watches Shopify orders, one watches the artist-site sold-out state.
4. Do not deploy the film site during the window unless sold-out is a one-line CMS toggle.
5. Do not deploy a headless storefront during the window unless you like incident theater.
6. After sell-through, confirm Spotify / ads / email still pointing at a product that can no longer sell — or kill those doors.

If you cannot staff step 3, you cannot staff Hydrogen. Hybrid with a permalink is the grown-up version of “we will just Stripe it.”

How to measure whether it is working

You are not measuring “Shopify vs custom” as a brand preference. You are measuring whether fans can buy the object without getting lied to.

Signal	Healthy	Sick
Path to variant	One or two taps from the merch chapter	Homepage scavenger hunt
Inventory truth	Shopify quantity is law; site sold-out matches	CMS says available, checkout says no
Refund volume	Occasional size swaps in Shopify	Weekend Venmo threads
Manager Tuesday	Admin only, no deploy	Designer ticket to hide a SKU
Ad landing	Product or permalink completes	Film homepage, then a second hunt
Brand cut	Shop photography matches the tour language	Theme stock models next to a film-grade home

Checklist you can run after the first real week of sales — not after a hypothetical:

• Place a test order on mobile with a real variant.
• Refund that order in Shopify. Confirm the manager can do it without you.
• Sell one unit to zero on a size. Confirm the artist-site CTA flipped.
• Click every live Buy URL in the bio, latest email, and ads account. They should hit the same SKU.
• Read the receipt: tax and shipping look like the policy, not like a guess.
• Time the path: announce → merch chapter → variant → checkout start. If you need a map, fans will bounce.

I will not give you a fake “good” conversion rate. Catalogs, price points, and fanbases are not comparable enough to steal a number from a SaaS benchmark and call it ours. Watch refunds, sold-out mismatches, and whether a Tuesday stock change needs a deploy. Those are the tells.

If those tells are clean and the shop still looks rented, you have a craft problem. Hire craft on the lookbook and the photography. Do not replace the engine to treat a theme.

Fourteen-day review after a real drop or a real week of evergreen sales:

1. Export orders. Count refunds and the stated reason (size, ship time, “looked different”).
2. List every live Buy URL. Confirm they resolve to the same Shopify product.
3. Re-run the mobile path. Did a deploy last week quietly break the permalink?
4. Ask the manager: did any stock change require a designer?
5. Compare merch-chapter copy on shipping and tax to an actual receipt. Mismatch is a trust bug.
6. Decide: craft pass (photos, theme chrome, lookbook) vs architecture pass (still on a fake cart). Do not buy a new stack to treat a photo problem.

If you only have a week — and when to wait

A week is enough to pick the engine and ship one honest door. It is not enough to build Hydrogen.

Do this week:

1. Open Shopify. One store. One product with the real sizes. Inventory tracking on.
2. Put a merch chapter on the artist site with brand photos and a price you do not hide.
3. Point the CTA at the Shopify product URL or a cart permalink. Test it on a phone.
4. Write sold-out copy. Do not wait for the 404.
5. Set shipping copy you can honor. If POD, say POD.
6. If you already know you must collect tax in a region, turn it on in Shopify. If you do not know, call an accountant — do not invent a rate in the product description.
7. Kill the old Gumroad / Venmo / CMS quantity field for that SKU.

Skip this week:

• Hydrogen, Storefront API, or a custom cart
• A full theme rebuild “so it feels like the album”
• Plus, checkout extensions, and a branded checkout subdomain
• Migrating every dead SKU from 2019
• Promising Spotify merch, ads scale, or a drop calendar you cannot staff

Wait on a custom storefront when:

You are here	Custom store is not worth it yet
SKU count	One poster, or a tip jar
Staffing	No one to maintain a React storefront
Money	You have not priced the tee after blanks, print, and live processing
Legal	You have not decided where you must collect tax
Craft	Photography is still vendor mockups. Fix photos before architecture.
Timeline	Release week is seven days out. Ship hybrid. Argue Hydrogen in the off-cycle.

Custom sites cost five figures because they are systems. A custom store is that bill plus commerce ops. If you cannot name the maintainer and the Tuesday tool, you are not ready. Use Shopify for catalog, checkout, tax, and shipping. Put the film on the artist site. Revisit custom browse when the merch chapter is already a destination and a theme is the actual ceiling — not when you are allergic to Dawn.

Shop engines are boring on purpose. The film should not be.

FAQ

Shopify vs custom store for artist merch?

Shopify should own catalog, checkout, tax, and shipping. A custom storefront is worth it when the merch experience is the brand film and you can staff the engineering. Most artists should run a hybrid: custom (or Framer/Webflow) artist site for identity, Shopify cart for money. Headless Shopify still checks out on Shopify.

How do I measure whether Shopify vs a custom store is working?

Watch the buy path, inventory truth, and Tuesday ops — not a borrowed conversion benchmark. A working split lets a fan reach a variant in one or two taps, keeps quantity in Shopify, matches sold-out on the artist site, and lets a manager refund without Venmo. If stock changes need a deploy, the architecture failed regardless of how cinematic the grid looks.

What usually fails first when teams try this?

Taste-led carts fail first: Stripe on a marketing site with quantity in the CMS, then an oversold size. The other early fail is a film-grade homepage next to an unmodified theme, so fans distrust the object. Headless teams fail third when they thought checkout would become “theirs” and discover Shopify still hosts it.

How long does this take to show results?

Hybrid can show an honest path in a week: one tracked SKU, one merch chapter, one permalink, one test order. Theme polish is days to a few weeks. A real custom storefront is a project — think in the same shape as other five-figure custom work, not a weekend. Results mean fewer refunds and a path fans can finish, not a fabricated GMV spike.

What should I skip if I only have a week?

Skip Hydrogen, a full theme rewrite, Plus checkout theater, and migrating every old SKU. Do one tracked Shopify product, a branded merch chapter, a phone-tested permalink, sold-out copy, and honest shipping language, then kill the second inventory. Custom browse waits for an off-cycle and a named maintainer.

When is this not worth doing yet?

A custom storefront is not worth it when you have one object, no tax decision, vendor-mockup photography, or nobody to own a React app. Shopify is not worth fighting if you truly sell a handful of units a year through DMs — but the moment you need variants, a drop lock, or receipts, the payment-link era is over. Wait for a real catalog or a real merch destination, then pick hybrid before headless.

CTA

Need a merch store that still feels like the film while checkout, tax, and shipping stay real? Start on /websites or book a sprint at /contact?intent=websites-sprint.