On Google’s phone test, your homepage asks for 20 seconds. Google gives it 2.5. SAP stopped fixing it.
That is how long your main image takes on Google’s standard test: a mid-range phone on a mobile connection, opening the page fresh. Real visits fare better today, but the storefront you are on is deprecated, so this number only moves one way.
What that score means for a buyer
The score is just the symptom. Here is how it affects someone opening your store on a phone, and makes ordering slower and more frustrating for them.
Loading on a phone
Real visits pass Core Web Vitals today, with the main image at 2.3 seconds for most of them. This is a first visit on a mid-range phone, which is where the storefront does worst.
How long the screen ignores a tap
On Google’s phone test. Google’s limit for a good score is 200 ms, and past 600 ms it is poor.
The weight of your homepage on a phone
Well past the median mobile page. Source: HTTP Archive, 2025.
JavaScript alone, across 110 files
Over 0.8 MB of it is downloaded and never runs.
Files the page must finish before it can show anything
Each dot is one request. The red ones hold the screen blank, which Google puts at about 2 seconds.
Why the test matters when real visits pass
Google’s test opens the page fresh, on a mid-range phone, on a mobile connection. That is how a buyer trying a new supplier meets you, and it is the visit this storefront handles worst.
Real visits pass today. On a deprecated storefront every new feature and tag eats into that, and no performance work is coming to win it back.
Nothing in this storefront will get faster on its own. SAP has deprecated it and no longer develops it.
Nobody is coming to fix this
Your team did not make this slow. The storefront did, and SAP has stopped developing it.
It is deprecated, so it will not improve.
SAP has frozen the Commerce Accelerator: no new features, no performance work. The removal date moved in June 2026. The dead end did not.
Every feature you add makes it slower.
In our experience the day-one score is the best a storefront ever gets. Each feature and each tracking tag takes its cut, and on a frozen platform nothing gives it back.
Whatever SAP builds next will not reach it.
New SAP capabilities, AI included, ship API-only to headless storefronts. The Accelerator cannot call them.
So the question is not whether you move. It is where to.
SAP’s recommended next step is Composable Storefront (Spartacus). We measured 200+ stores that took it.
It is a real step up in how it is built: headless, API-driven, modern tooling. Performance is where it stops. The two numbers below are the average across 200+ live stores, and almost every one scores about the same, no matter who built it.
Source: Alokai benchmark of 200+ public SAP Composable Storefront sites, 2026.
higher conversion, plus 9.2% higher average order value, from a single 0.1 s improvement in load time.
Google and Deloitte, Milliseconds Make Millionsof sales lost for every 100 milliseconds of added latency, in Amazon’s own testing.
Amazonthe conversion lift teams report after moving to a modern, lightweight frontend.
Composable migration benchmarksSpartacus caps your performance low from the start. Alokai is not bound by it.
Both demos ran the same mobile Lighthouse test, and both are full storefronts. Spartacus scores lower because of how it is built: a heavy JavaScript bundle it cannot shed, so teams fight the framework for every point. Great performance comes from the architecture underneath, which an accelerator does not have. You need a platform, not an accelerator.
What fixing this gets you
The slowness is the storefront, not your team, and tuning a frozen one barely dents it. A modern storefront on the same SAP Commerce Cloud does these three things on day one.
Every image is sized for the device asking for it and sent in a modern format, so a phone downloads a fraction of what it does today.
Far less legacy script, loaded after the page is usable instead of before it can show anything.
Far less code for the phone to run, so the page answers the first tap instead of ignoring it.
Your store turns more first visits into orders, earns more from every visit, and keeps the buyers it has. Your team stops maintaining a storefront nobody develops and runs the store for less.
Carolina Herrera kept SAP Commerce Cloud and moved the storefront. Two months later they were past their KPIs.
Carolina Herrera, part of the PUIG group, wanted an independent read on their SAP storefront. They moved the frontend to Alokai, kept the backend where it was, and two months after launch the results were already well past the targets they had set.
Pages Google now indexes
Every market finally has its own pages instead of one generic set.
Organic clicks from the US
Total organic clicks up 17%. The localised markets are where it compounded.
Localised click-through rate
Brazil doubled too, 0.6% to 1.1%.
Traffic to hero product pages
The pages that actually sell, in the markets that actually buy.
Migrating is not the year you think it is
Your backend does not move. The frontend goes market by market, and because Alokai is AI-native, agents do most of the rewrite in weeks instead of months.
Your backend stays. Your storefront moves, market by market. A migration map tells you what that looks like for your store, before you commit to anything.
Move once, and skip the storefront you would have to leave again.
The Accelerator is Gen 1. Spartacus is Gen 2. Alokai takes you straight to Gen 3 on the same SAP Commerce Cloud, for the same migration effort.
Everything that works stays.
Catalog, promotions, OMS, pricing, personalization: all stay in SAP Commerce Cloud. SmartEdit stays the tool your merchandisers already know.
First storefront live in 3 to 6 months.
Market by market, no big-bang cutover. Value ships monthly while the old store keeps selling.
Built in React (Next.js) or Vue (Nuxt).
More than twice as many developers work in React as in Angular (Stack Overflow Developer Survey 2025), and you will feel that in every role you fill.
The questions you will ask
Yes. In June 2026 SAP removed the fixed removal date for the deprecated Accelerator UI templates, and support continues with no set end date. But deprecated still means frozen: no new features, and SAP’s roadmap, including AI, ships API-only to headless frontends. The deadline moved. The dead end did not.
What happens when you reply
Reply to this scan.
Tell us who on your team should be on the call.
30-minute call with an architect.
Someone who has run this migration walks your SAP Commerce Cloud setup with you.
Your migration map.
What stays, what moves, in what order, and a realistic timeline and cost. Yours to keep.
First storefront live.
Market by market, while the old store keeps selling.
Map your migration path.
The migration is inevitable. Doing it twice is not. Book a technical deep dive with an Alokai architect who has run this migration before. We will walk your current SAP Commerce Cloud instance and give you a realistic timeline.