Performance is the single most common reason mid-market merchants replatform between Magento and Shopify in either direction. The story is more nuanced than the marketing on either side. Shopify Plus delivers strong performance out of the box for most catalogs. Adobe Commerce with Hyvä delivers comparable or better performance after the migration is properly done. The differences matter for specific store profiles, and the merchants who pick well do so based on what each platform actually controls versus what they have to control themselves.
Bemeir is a Shopify Plus partner and a leading Adobe Commerce agency, and we run both architectures in production for clients every week. This comparison is from inside the codebase, not from a vendor pitch deck.
What “performance” actually means for an eCommerce store
The conversation gets sloppy fast unless the metrics are defined. The four numbers that actually correlate with revenue:
Largest Contentful Paint (LCP). How quickly the main content of a page becomes visible. Target: under 2.5 seconds on mobile. Below 4 seconds is poor and measurably hurts conversion.
Interaction to Next Paint (INP). How responsive the page feels when a customer taps a button, types in a field, or opens a menu. Target: under 200 milliseconds. INP replaced FID as a Core Web Vital in 2024 and is the most underrated performance metric on eCommerce.
Cumulative Layout Shift (CLS). How much the page jumps around as it loads. Target: under 0.1. CLS above 0.25 is the most common cause of customers tapping the wrong thing on mobile and abandoning.
Time to First Byte (TTFB). How quickly the server starts responding. Target: under 600ms. TTFB is controlled by hosting, caching, and platform internals more than by frontend code.
Lighthouse Performance score correlates with these but is itself not the metric to optimize. What matters is the field data from real customer devices.
What each platform actually controls
Shopify Plus and Adobe Commerce with Hyvä have fundamentally different ownership boundaries. The merchant needs to understand which side of the line each performance dimension sits on.
| Performance dimension | Shopify Plus controls | Merchant controls | Adobe Commerce + Hyvä controls | Merchant controls |
|---|---|---|---|---|
| Hosting + CDN | Yes (Cloudflare global) | No | Partial (cloud) or No (on-prem) | Hosting choice |
| Server response (TTFB) | Yes | App + theme bloat | No | Yes |
| Asset bundle size | Theme dependent | Theme + apps | Hyvä baseline ~25KB | Custom code + extensions |
| Image optimization | Yes (built-in) | Image sizing | No | Yes (must configure) |
| Third-party app/extension impact | High | High | High | High |
| INP (interactivity) | Theme dependent | Apps + custom JS | Excellent (Alpine.js) | Extension JS |
| Database query performance | Yes | No | No | Yes (DBA work) |
| Caching strategy | Yes | No | No | Yes (Varnish + Redis) |
The headline is: Shopify Plus moves performance ownership to the platform and away from the merchant. Adobe Commerce moves performance ownership to the merchant and away from the platform. Hyvä on Adobe Commerce closes most of the frontend gap; hosting and database performance remain merchant responsibilities.
Real benchmark numbers
Synthetic Lighthouse runs are notoriously noisy. Here are typical field-data ranges we see across the production stores our team runs and audits, measured through Google’s PageSpeed Insights and Chrome User Experience Report.
| Stack | Median mobile LCP | Median mobile INP | Median CLS | Mobile Lighthouse Performance |
|---|---|---|---|---|
| Shopify Plus, Dawn theme, default apps | 1.8-2.4s | 130-180ms | 0.02-0.08 | 70-85 |
| Shopify Plus, custom theme, 8-15 apps | 2.4-3.8s | 200-350ms | 0.08-0.18 | 55-75 |
| Adobe Commerce + Luma (untouched) | 4.0-7.0s | 350-600ms | 0.10-0.40 | 25-45 |
| Adobe Commerce + Hyvä (baseline) | 1.6-2.4s | 90-160ms | 0.02-0.08 | 80-95 |
| Adobe Commerce + Hyvä (tuned) | 1.0-1.8s | 60-120ms | 0.01-0.05 | 90-100 |
Two things stand out from this data. First, Shopify Plus is consistently good at LCP but average at INP on heavily customized stores. Second, Adobe Commerce + Hyvä is the only stack in this table that routinely hits the green Core Web Vitals bands across all three metrics. Untouched Luma is in a different conversation entirely; nobody should be benchmarking against it.
Why Shopify Plus wins on simplicity
The performance advantages Shopify Plus delivers are real and worth naming honestly.
Built-in global CDN. Every Shopify store runs on Cloudflare with edge caching tuned for eCommerce. TTFB is consistently under 300ms from most US locations without merchant configuration.
Native image optimization. The platform serves WebP and AVIF automatically, with responsive image variants generated from the original upload. Merchants do not need to configure this and rarely break it.
No server-side cache invalidation to think about. Product updates, inventory changes, and content edits propagate to the edge without merchant configuration. There is no Varnish to tune, no Redis to configure, no nightly cache warm-up scripts.
Predictable scaling. Black Friday spikes are handled by Shopify’s infrastructure. The merchant does not get paged at 2am because the database hit a connection limit.
For a merchant whose engineering team is small, whose catalog is straightforward, and whose business model fits Shopify Plus’s storefront and checkout patterns, this is a meaningful set of advantages. Bemeir’s Shopify team runs many of these stores and the operational simplicity is real.
Why Adobe Commerce + Hyvä wins on ceiling
The performance advantages Adobe Commerce + Hyvä delivers, when the implementation is right, are also real.
Lower JavaScript baseline. Hyvä ships about 25KB of JavaScript on a typical page. A Shopify Plus store with 8-15 apps frequently ships 250-500KB. The Hyvä baseline is hard to match without aggressive app management on Shopify.
Better INP at scale. Alpine.js is a smaller, simpler runtime than the React-based components most Shopify apps inject. INP scores on tuned Hyvä builds routinely sit in the 60-120ms band, which is where the customer experiences the page as instantly responsive.
Total caching control. Adobe Commerce stores typically run Varnish at the edge, Redis for session and full-page cache, and a CDN of the merchant’s choosing. The performance ceiling is higher than Shopify Plus delivers because the merchant can tune every layer.
Custom database query optimization. Heavy B2B catalogs (50,000+ SKUs, complex attribute filtering, customer-specific pricing) can be optimized at the database level on Adobe Commerce in ways Shopify Plus does not allow.
For a merchant with the engineering team to operate the stack, the catalog complexity to need the tuning, and the business model that justifies the investment, Adobe Commerce + Hyvä reaches a higher performance ceiling than Shopify Plus.
The honest decision framework
The performance comparison should not be the only factor in the platform decision, but here is how it lands when isolated:
Shopify Plus wins for performance when: the merchant has a small engineering team, the catalog is under 5,000 SKUs, the storefront design is within Shopify’s theme conventions, the app footprint is kept disciplined (under 12 production apps), and the business does not need heavy custom merchandising logic.
Adobe Commerce + Hyvä wins for performance when: the merchant has at least one full-time frontend engineer, the catalog is large or complex, the business needs custom pricing or merchandising rules at the database level, the team is willing to invest in Hyvä migration and ongoing optimization, and the conversion lift from a tuned Hyvä build justifies the engineering investment.
Neither platform wins for performance when: the implementation team treats apps or extensions as free and stacks 25+ of them on the storefront. Performance dies on both platforms under that pattern.
What the migration math looks like in practice
A common scenario: a merchant on Adobe Commerce Luma with poor Core Web Vitals is evaluating whether to migrate to Shopify Plus or to migrate to Adobe Commerce + Hyvä. The performance outcome is similar at the median. The migration scope is very different.
The Shopify Plus migration involves rebuilding the storefront in Liquid, rebuilding the checkout in Shopify’s checkout extensibility model, re-implementing or replacing every Magento extension as a Shopify app or custom app, migrating customer data, and rebuilding any B2B or multi-store functionality. Timeline: typically 6-12 months for a mid-market store. We have covered this in detail with Bemeir’s Shopify Plus team.
The Hyvä migration involves replacing the Luma theme with Hyvä, verifying or rebuilding extension compatibility, optionally migrating to Hyvä Checkout, and tuning performance. Backend, catalog, customer data, and integrations all stay in place. Timeline: typically 3-6 months for the same store.
For a merchant whose only problem is Luma’s performance, the Hyvä path is faster, cheaper, and lower-risk. For a merchant whose problems include backend complexity, hosting cost, and operational overhead, Shopify Plus may be the right longer migration. The performance data alone does not pick the winner.
The reality the benchmark data does not show
A pattern from running both stacks in production: the second-year experience matters more than launch-day numbers. Shopify Plus stays performant by default as the merchant adds apps, but only if app discipline is enforced; we have seen well-built Shopify Plus stores degrade to 60-second Lighthouse scores by year two from uncontrolled app sprawl. Adobe Commerce + Hyvä stays performant by default if the team continues to test and tune; without that discipline, performance drifts as extensions are added and the catalog grows.
Both platforms reward operational discipline. Neither platform is a setup-and-forget performance solution. The choice between them should hinge on which set of trade-offs your team can actually operate, not on which one had the better Lighthouse score on demo day.
Our team at Bemeir ships and maintains both stacks in production. When merchants ask us which to pick on performance grounds alone, the honest answer is that both can win; the question is which set of ongoing operational responsibilities your team is best positioned to own.
