Should You Build or Buy Translation on Shopify?

Written by Deploi EditorialReviewed by Martin Dejnicki, Director of SEO & AI SearchUpdated August 2026Pricing verified July 2026 (research corpus; re-verify)

Translation on Shopify is a CUSTOMIZE call for mid-market stores: Shopify's free native Translate & Adapt covers basic manual plus auto translation and stores every string portably in Shopify, so the money goes to an estimated $10,000–$30,000 AI translation pipeline (Deploi estimate, illustrative) once you pass roughly three locales. Vendor risk now leads the scorecard: Weglot, the category's former default, was delisted from the App Store in early 2026.

Your profile — see how the verdict shifts

VerdictCUSTOMIZE on native Translate & Adapt · WAIT under three locales · never proxy translation
Buy score
4.2
Build score
7.4
Confidence
HighThe frontier already moved twice in the customize path's favor: Weglot's delisting made vendor risk concrete, and the free native baseline stores translations portably in Shopify
Reference scenario
$20M–$100M GMV · 3–5 languages · content-heavy catalog · agency dev bench
As of
August 2026

Decision at a Glance

Your profileVerdictWhy
Single-language storesWAITTranslation is a decision for the quarter you commit to a second market. Nothing to buy, nothing to build; Translate & Adapt is included and waiting when you are.
1–2 extra languagesWAITTranslate & Adapt covers basic manual plus auto translation at this footprint, and everything it stores is portable inside Shopify. A subscription here buys convenience, not capability.
3–5 languages, content-heavy catalogCUSTOMIZEPer-field manual editing stops scaling around here. An AI pipeline writing through the translation APIs keeps locales continuously fresh, and every string it writes stays yours.
6+ languages or high content velocityCUSTOMIZEWord- and language-tier app pricing compounds at this footprint while the pipeline's cost stays flat, and owned translations become a per-market SEO and AI-answer surface.

What Translation Actually Drives

OutcomeImpactHow it works
Revenue — directHighA storefront in the shopper's own language removes guesswork at the moment of payment, converting international traffic you already earn instead of leaving it to bounce.
Customer experienceHighSizing, ingredients, policies, and checkout copy read naturally in each locale, so international shoppers stop screenshotting pages into translation tools mid-purchase.
Revenue — indirectMediumNative-stored translations publish as crawlable localized pages on your own domains, opening organic search and AI-answer visibility in each market.
Operational efficiencyMediumA pipeline that re-translates only changed content replaces per-field manual editing, which is the task that quietly swallows a content team past three locales.
Data & insightMediumTranslations kept as structured content on your resources stay queryable and reusable, feeding the same glossary and tone rules into email, support, and AI surfaces.

Spend ceiling: Machine translation is a commodity; the scarce parts are workflow and ownership. Size spend to the glossary, review flow, and per-locale SEO fields, not to raw word counts. Storage is included with every Shopify plan, so a quote that's mostly storage and rendering is charging you for the part Shopify gives away.

What buying enables (top apps)

  • + Bulk machine translation of the whole catalog in days, managed per-language from one dashboard
  • + Visual in-context editors that let non-developers fix a translation without touching the theme
  • + Glossaries, SEO-field controls, and third-party translator hand-off workflows Translate & Adapt doesn't attempt
  • + Vendor-maintained coverage of new content types as Shopify adds surfaces

What building additionally unlocks

  • + AI translation prompted with your brand voice and full product context (metafields, FAQs, campaign copy), not string-by-string machine output
  • + Continuous localization: content-change webhooks re-translate only the diff, so locales never drift stale between campaigns
  • + Translations as owned, portable content on your Shopify resources, reusable across email, support, and AI answer surfaces with no export ceiling
  • + Zero storefront runtime dependency: no proxy hop, no injected script, nothing to strand if a vendor exits

Find Your Verdict in 3 Questions

  1. Are you selling in more than one language today, or committed to a second market this year?

    Yes: Go to question 2.

    No: Your verdict: WAIT — nothing to build or buy before a market commits; Translate & Adapt is included and ready when one does.

  2. Is Translate & Adapt keeping up: a few locales, manageable content volume, basic auto translation good enough?

    Yes: Your verdict: WAIT — stay native and free; if it strains later, the pipeline path is waiting and nothing you've translated is lost.

    No: Go to question 3.

  3. Do you have dev capacity, agency or in-house, for a 3–6 week pipeline build?

    Yes: Your verdict: CUSTOMIZE — build the AI pipeline on the translation APIs; every string it writes stays on your resources.

    No: Your verdict: BUY — shortlist only apps that write to native translation storage; the Weglot delisting is what the other architecture costs.

The TCC Scorecard — 12 Dimensions

TCC — Total Cost of Capability: what it actually costs to have this capability over three years, whichever way you get it. Each dimension is scored 0–5 for both paths. How we score →

DimensionBuyBuildWhy
Cost
Acquisition & implementationAn app bulk-translates the catalog in days; Translate & Adapt is same-day for the floor, but the pipeline that completes the customize path runs an estimated 3–6 weeks (Deploi estimate, illustrative).
Recurring feesTranslation apps bill monthly and tier by words and languages; Translate & Adapt is included with your plan, and the pipeline's recurring line is AI usage plus upkeep, both small.
Maintenance & upgradesThe vendor tracks theme and checkout changes for you; your pipeline carries roughly 15–20% of build cost per year (Deploi estimate) plus glossary and prompt upkeep.
Switching & exitDecisive after Weglot: proxy-held translations leave with the vendor, and even native-writing apps vary on cleanup. Pipeline output lives on your resources and survives any tooling change.
Risk
Vendor riskThe live case: Weglot, the category's biggest name, was delisted from the App Store in early 2026 (July 2026 research). Native storage plus your own pipeline leaves no vendor to lose.
Security & compliance surfaceProxy translation routes storefront traffic through vendor infrastructure; the pipeline keeps content inside Shopify but does hand it to an AI provider you choose and can swap.
Platform-deprecation exposureThe translation APIs and Markets are first-class surfaces; script- and proxy-based apps sit outside them. Your pipeline's exposure is routine API version bumps, roughly every six months (July 2026 research).
Value
Fit to requirementApps ship visual editors and translator hand-off workflows Translate & Adapt doesn't attempt; the pipeline matches your glossary, tone, and review policy exactly, but you build that workflow yourself.
Time to marketDays for an app to machine-translate everything; same-day for Translate & Adapt basics; weeks before the pipeline runs unattended.
Performance & scaleProxy- and script-rendered translation adds a runtime hop to every localized page view, and the app-bloat speed tax is a documented recurring pattern; native-stored translations render with the theme.
Data ownership & AI-readinessThe decisive dimension: owned translations are structured, reusable content for SEO, email, support, and AI surfaces; rented ones are an export request away, or gone at delisting.
Focus & opportunity costMachine translation is a commodity now, so the pipeline is mostly glue; Translate & Adapt carries the baseline while you build, which keeps the roadmap cost honest.

The App Landscape

AppStatusPricingBest for
Translate & AdaptNativeShopify's own translation app; the post-Weglot default, with translations stored portably in ShopifyFree (included with Shopify)The floor for every store: basic manual plus auto translation with nothing to unwind later
LangifyLiveLong-running independent app in the native-storage laneMonthly, flat-tieredA managed translation workflow without proxy architecture; confirm storage and export terms at signup
LangShopLiveBroader localization suite: translation plus currency and locale switchingTiered by plan and languagesTeams that want auto translation, glossaries, and switchers from one dashboard
WeglotDelistedDELISTED from the Shopify App Store, early 2026 (July 2026 research) The former category default; its proxy-served translations left with itn/a since delisting (July 2026 research)No one today. Listed as the category's vendor-risk case study, not a shortlist option

The Build Path

  • Translate & Adapt as the floor: Turn on the native app first: manual translation for money pages, auto translation for the long tail. Everything it stores sits portably in Shopify, so nothing done here is throwaway.
  • AI translation pipeline on the translation APIs: A small private app reads translatable resources, runs AI translation with your glossary and tone rules, and writes results back to Shopify's native translation storage. Output is indistinguishable from hand-entered translations, and just as portable.
  • Continuous localization workflow: Content-change webhooks re-translate only the diff, a review queue gates money pages, and per-locale SEO fields (titles, metas, alt text) get deliberate handling instead of raw machine output.
Effort band
Translate & Adapt costs nothing to adopt; the pipeline runs an estimated $10,000–$30,000 (Deploi estimate, illustrative), landing in the $10–25K contact-form band, edging into $25–75K with review tooling and many locales
Typical timeline
Same-day for Translate & Adapt; 3–6 weeks for the pipeline and review workflow (Deploi estimate, illustrative)
Maintenance, honestly
Roughly 15–20% of build cost per year (Deploi estimate): API version bumps about every six months, model and prompt tuning, glossary upkeep; call it $2,000–$6,000/yr (Deploi estimate, illustrative). Translation storage and localized rendering are Shopify's to maintain.
What you own — and what you take on
You own: every translated string (stored on your Shopify resources), the glossary, the tone rules, and the review workflow. Shopify owns storage and localized rendering. You take on: translation-quality QA in languages you may not read, and the upkeep above.

3-Year Total Cost of Capability

Buy (app path)Build (custom path)
Year 0 (setup)$0–$2,000$10,000–$30,000
Years 1–3 (recurring)$7,200–$18,000$6,000–$18,000 (AI usage + maintenance)
3-year total≈$7,200–$20,000≈$16,000–$48,000
Illustrative cumulative cost over 36 months$0$8k$17k$25k$33kMo 0Mo 12Mo 24Mo 36Buy (app path)Build (custom path)
Illustrative cumulative cost: at mid-band pricing the subscription stays cheaper inside the three-year window, and we're showing you that honestly. The customize case is what the invoice never shows: translations you keep, flat economics as locales grow, and no exposure to the exit Weglot customers just took. A Translate & Adapt-only scope costs about nothing on either line.
  • All figures illustrative samples for the reference scenario — not quotes, not verified pricing.
  • App path: a mid-tier multi-language subscription held flat; real word- and language-tiered pricing scales with locales and catalog, which is conservative for the customize case.
  • Customize path: Translate & Adapt floor (free) plus the AI pipeline and review workflow; recurring line includes AI usage and maintenance; three-year horizon.

What the Sticker Price Hides

On the buy path

  • Proxy architecture leaves nothing behind at exit: Weglot's early-2026 delisting turned a tool swap into a relocalization project (July 2026 research)
  • Word-count and per-language tiers creep upward as markets and catalog grow (community-reported pattern)
  • Translation widgets add script weight and break at theme updates and breakpoints (community-reported theme)
  • Stale vendor assets linger after exit: a DTC beauty brand's speed program included cleaning leftover Weglot assets out of the theme

On the build path

  • AI translation without a glossary and review queue ships confident errors in languages you can't read
  • Bulk translation writes hit the GraphQL cost-based rate limit, and THROTTLED errors arrive inside a 200 response (documented dev trap, July 2026 research)
  • Per-locale SEO fields (titles, metas, alt text) are the forgotten 20% of the scope
  • Roughly 15–20% of build cost per year in upkeep (Deploi estimate)

What Merchants Say

The post-delisting complaint shape: translations, localized subdomains, and their SEO signals lived on the vendor's side, so losing the app meant restarting localization rather than swapping a tool.
community-reported (2026 research corpus)
The steady theme for translation apps generally: auto-translated pages read fine until a native speaker checks the money pages, and word- and language-tier pricing climbs as the catalog grows.
app-store 1–2★ review theme

If You Change Your Mind Later

If you bought and outgrow it

Ask where translations live before signup, not at exit. If the app writes to Shopify's native translation storage, leaving is an uninstall plus cleanup. If it proxies or holds translations server-side, leaving means relocalizing from scratch: that's what the Weglot delisting forced on its merchants, and a DTC beauty brand's speed program was still cleaning stale Weglot assets out of its theme afterward.

If you built and want out

Nothing is stranded: translations live on your own Shopify resources and survive any tooling change, including a retreat to Translate & Adapt alone or a future app. You'd walk away from the pipeline glue and keep every translated string, the glossary, and the tone rules. Portable storage is the whole point of this architecture.

When This Answer Changes

We're watching for:

  • Shopify upgrading Translate & Adapt with stronger AI translation or workflow features, which would push this verdict from CUSTOMIZE toward WAIT
  • Surviving translation apps repricing or consolidating after the Weglot delisting; watch this category the way subscriptions watched Recharge and Skio (July 2026 research)
  • AI shopping surfaces reading localized content directly, which raises the value of translations stored as owned, structured content

Verdict change log:

  • 2026-01-01Weglot delisted from the App Store in early 2026 (per July 2026 research). The default answer used to be install Weglot. Its exit showed what proxy architecture costs: translations held on the vendor's side don't survive the vendor. The verdict moved to native Translate & Adapt as the floor with an owned AI pipeline as the ceiling, because storage location is now the deciding question.

Common Questions

Is Shopify's Translate & Adapt really free?

Yes. Translate & Adapt is Shopify's native translation app, included at no charge, and it covers basic manual translation plus auto translation while storing every string portably in Shopify (July 2026 research; re-verify). That storage detail is the quiet headline: work done in it survives any later tooling change, which is exactly what proxy-app customers lost when Weglot left the App Store.

What happened to Weglot on Shopify?

Weglot, long the category's default recommendation, was delisted from the Shopify App Store in early 2026 (July 2026 research; re-verify). For merchants, the mechanism matters more than the reason: proxy-served translations live on the vendor's infrastructure, so a vendor exit becomes your migration project. It's the clearest recent argument for keeping translations in Shopify's native storage, whoever creates them.

Can AI translate a Shopify store well enough to trust?

For the catalog long tail, yes, provided you add guardrails: a per-locale glossary, tone rules, and human review on money pages like PDPs, policies, and checkout content. The customize path bakes those in and writes output through the translation APIs, so every string lands as owned, portable content. Unreviewed machine output on legal or medical claims is the line we won't cross.

Your Next Steps

If you're going with CUSTOMIZE(matches your selected profile)

  1. Turn on Translate & Adapt today; it's the free floor and everything it stores carries forward
  2. Write the per-locale glossary and tone rules before automating anything; they're the quality contract
  3. Scope the pipeline: content-change webhooks, AI translation, a review queue, writes to native translation storage
  4. Route money pages (PDPs, policies, checkout content) through human review; automate the long tail
  5. Batch API writes against the cost-based rate limit and handle THROTTLED errors that arrive inside 200 responses

If you're going with BUY

  1. Make native-storage writes the first filter; reject proxy and script-render architectures outright
  2. Get the export path and cleanup terms in writing before signup; Weglot is the cautionary case
  3. Model pricing at twice your locale count and content volume; word and language tiers are where fees creep
  4. Cap script weight and test theme updates on a duplicate; translation widgets break at breakpoints
  5. Diary a yearly re-decision; this category just lost its biggest vendor

Official Docs & Sources

Official documentation linked for verification — our verdicts and estimates are our own.

Ready to own your translations?

Translate & Adapt is the free floor. We scope the AI pipeline on Shopify's translation APIs: your glossary, your tone, human review where it counts, and every translated string stored where no vendor exit can strand it.

Contact us today

Ecommerce development at Deploi

Verdict scored for the reference scenario above. Estimates are not quotes; app pricing carries its verification date and gets re-verified quarterly. Full scoring anchors: see the TCC methodology.

Read how we score these decisions (the TCC Framework). No affiliate links, no paid placement — no app vendor pays to appear here.

No affiliate links. No paid placement. We make money building and integrating solutions — not on referral fees.