Salesforce Commerce Cloud SEO (2026)

Salesforce Commerce Cloud SEO means making your product and category pages easy to find, crawl, index and buy from. For B2C Commerce, start with URL rules, canonical tags, search refinements and XML sitemaps. Then check the storefront itself. PWA Kit adds server rendering and browser hydration, so a correct Business Manager setting still needs a live-page test.

The expensive mistake is fixing the catalogue in one place while the storefront keeps publishing something else. A category gets a new address. Navigation follows it, but the sitemap still points elsewhere. Or the HTML contains the correct canonical until JavaScript changes it.

That gap is where this article starts.

In summary

Salesforce Commerce Cloud SEO works when catalogue settings and public pages agree about which URLs deserve search traffic.

• Check which storefront handles the request before changing Business Manager rules.
• Give useful categories stable URLs; keep sorting and disposable filter combinations out of the search strategy.
• Test sitemap inclusion against real catalogue availability and regional settings.
• Compare PWA Kit's initial HTML with the page after hydration and internal navigation.
• Judge progress by product discovery and organic sales, with each market separate.

Which Salesforce Commerce Cloud setup are you optimising?

This article covers Salesforce B2C Commerce SEO, including conventional storefronts and headless storefronts using PWA Kit. Start by identifying your architecture. Instructions for B2C Business Manager aren't interchangeable with instructions for Salesforce B2B Commerce.

Ask your engineering team where each public response comes from. The answer determines where a fix belongs.

SFRA means Storefront Reference Architecture. PWA Kit uses React, with a server-rendered first request followed by browser rendering, as Salesforce's rendering documentation explains.

That's the important distinction. Business Manager records the configuration. The public response reveals what happens when a customer or crawler requests that address, including anything the storefront changes before displaying the page.

Searchmaxxed's technical SEO work covers crawl and indexing fixes, redirects, canonical tags, XML sitemaps, schema and mobile rendering. The architecture sets the implementation scope.

How should you configure Commerce Cloud URL rules?

Commerce Cloud URL rules define patterns for product, category and content addresses, which you configure in Business Manager before testing and replicating to production.

Find the controls under Merchant Tools, SEO, URL Rules. Salesforce's Configure SEO URLs lesson recommends configuring and testing in a sandbox before exporting to staging, where your administrator replicates the configuration to production.

Keep that sequence. The rule may also change categories outside the section you are testing.

What belongs in a URL rule?

Use stable catalogue information. A product's name, category assignment and pageURL override can affect the generated address, so record which field controls each part.

Salesforce documents an exception worth bringing to your developers: lowercasing URL text leaves product IDs unchanged, even when those IDs contain capital letters. Preserve their case. A blanket edge rule that lowercases every character could therefore change more than intended.

The platform also adds the required product or content asset ID automatically within its documented URL-rule system. Headless routing needs its own check. Avoid transferring every conventional storefront rule into a React application without inspecting its actual routes.

What happens when categories share a name?

Shared names can create URL conflicts when the rule omits their parent categories. Salesforce recommends considering category-path where that ancestry separates otherwise identical names.

For a manufacturer, both power tools and laboratory equipment might contain a category called “accessories”. The category name alone loses the distinction.

Before publication, check overrides and hidden categories against the conflict reports, rather than accepting an accidental suffix as the permanent address for that category.

How do you change URLs without losing existing routes?

Map each old address to its relevant replacement, then test the redirect on the public hostname. Include old product URLs, category paths and incoming campaign links.

A configuration save doesn't prove that. B2C Commerce's URL Request Analyzer helps trace request handling, which the Bee IT implementation guide demonstrates.

For headless routes, inspect the application handling that request as well. Keep internal links and sitemap entries pointed at the final destination, rather than making visitors traverse the redirect.

Which facets and product URLs should Google index?

Index categories and selected refinements that answer distinct buying searches, while treating the remaining filter combinations as navigation with a separate crawl policy.

Facets filter products by their attributes. For example, filters might select material, size or compatibility, but replacing their query strings with tidy paths doesn't create useful landing pages by itself.

Consider a safety equipment manufacturer. “Chemical-resistant gloves” can deserve a permanent category page with relevant products, specifications and buying guidance. Sorting doesn't create another buying search.

The distinction matters more than the punctuation.

A canonical tag names your preferred URL for duplicate or very similar content. It sends a signal, rather than forcing Google's choice. Google's canonical guidance also advises consistent internal links and sitemap URLs.

Keep the controls separate. Robots.txt manages crawling, while a noindex instruction excludes a page from search only when Google can retrieve that instruction on the page. A canonical addresses duplication.

Google's faceted navigation guidance recommends restricting unwanted crawl paths and returning HTTP 404 for empty filter combinations. Plan those rules around useful categories. A blanket block can catch pages you wanted people to find.

The wider decision process belongs in our faceted navigation SEO guide. Here, the SFCC-specific job is checking how refinement rules, metadata and storefront links express that decision.

What should you check in SFCC XML sitemaps?

Compare the generated sitemap with the pages you intend to rank, checking the public host, included locales, product selection and the resulting addresses.

An XML sitemap helps URL discovery. Its generation job can succeed even when the file omits valuable products, includes unwanted pages or sends search engines to the wrong regional addresses.

Salesforce's Generate Sitemaps lesson documents per-host sitemap files, included locales and product-selection settings. It also explains the sitemap-included-flag attribute and generation logs.

That product setting deserves attention. If you select “Available and Orderable Products Only”, compare the output against your policy for temporarily unavailable products. A useful product page may still deserve discovery while stock replenishes.

What does the sitemap acceptance check include?

Compare the catalogue you want indexed with the sitemap Google can fetch. Investigate both missing pages and unwanted additions.

• Confirm the production hostname.
• Check included locales and their alternate URLs against the storefront's actual language and country routing.
• Inspect excluded products and categories.
• Check sitemap-included-flag values, including values arriving through catalogue feeds that could replace a manual setting.
• Review the generation log.
• Fetch a sample of listed URLs and compare their status, canonical and content with the intended page.

Salesforce supports up to 50,000 links per sitemap file through this setting. Treat that as the ceiling. Smaller logical groups can make diagnosis easier when a specific catalogue section stops appearing.

Schedule generation around catalogue changes. After a URL-rule release, check that the next sitemap uses the new addresses and that old links still resolve correctly.

How should Australian and US storefronts handle hreflang?

Hreflang connects equivalent regional pages through language-country codes and return links, helping Google identify which version suits the person searching in that market.

Australian English uses en-AU; US English uses en-US. A B2C locale value such as en_US needs the correct hreflang form, with a hyphen rather than an underscore.

Check the output, not the label.

Google's localised-version guidance requires each version to list itself and its alternates. Reciprocal links matter. If the US page names Australia but Australia never links back, that relationship is incomplete.

Match equivalent products across the regional catalogue, rather than substituting a homepage whenever the corresponding product doesn't exist or is unavailable in another market. Map genuine counterparts. Where no counterpart exists, omit that alternate.

A blanket canonical pointing Australian pages to the US site contradicts your intention if both regional pages should remain searchable for their respective markets.

Also test the market selector. A buyer arriving on an Australian product should reach its US counterpart, where one exists, instead of starting again at the homepage.

What changes for SEO when you use PWA Kit?

PWA Kit needs checks before and after JavaScript runs, because the server-rendered response and hydrated browser page should agree about which page they represent.

Salesforce documents server-side rendering in the application server, followed by hydration and client-side rendering. Managed Runtime can cache rendered pages, but your implementation may reserve personalised or frequently changing content for browser rendering rather than the initial response.

So inspect the distinction. “We use server rendering” doesn't tell you whether a particular product description, canonical or category link arrived in the initial response.

Hydration means React attaches browser behaviour to the server-rendered page. Google specifically recommends setting the canonical in source HTML and keeping JavaScript from changing it in its canonical documentation.

Test errors too. A missing product needs an appropriate HTTP response, rather than a cheerful success response around a “not found” message.

Give deeper products crawlable links. Google says its crawler generally follows linked URLs rather than clicking buttons to reveal more products. Our pagination SEO guide covers that design choice, including why later pages shouldn't all canonicalise to the beginning.

Performance belongs here as well. Check mobile Core Web Vitals alongside the route tests, especially product images, layout movement and interaction delays after hydration.

How do metadata and product schema scale across a large catalogue?

Use metadata rules for repeatable titles and descriptions, allowing deliberate exceptions for important products and checking the output against the current product and market.

B2C Commerce includes Page Meta Tag Rules for generating metadata at scale, as the Bee IT guide demonstrates. Start with useful source fields. The rule only has the fields you give it, so missing specifications and meaningless category names can spread poor metadata across an entire catalogue.

For a manufacturer's catalogue, a title can combine product type, model and a distinguishing specification. Keep empty attributes out. Category copy should explain differences shoppers need to choose, rather than repeat a list of product names.

The same discipline applies to Product structured data. Google's product documentation warns that dynamically generated markup can make shopping crawls less frequent and less reliable.

Compare visible product details with structured data after regional selection, variant selection and client-side navigation, paying particular attention to the price and availability.

Schema describes the current product. That means the product the customer can see, with its current details, rather than yesterday's inventory or the item they opened before this one.

How would you test an SFCC release across the whole catalogue?

Test representative URL behaviours together. Your acceptance record should connect each requested URL with the resulting page, its canonical, regional alternates and the sitemap entry describing that same page.

Here's a worked example. Imagine an enterprise manufacturer selling safety equipment in Australia and the US through a PWA Kit storefront. This is an illustrative release test, not a client result.

The team wants a permanent chemical-resistant gloves category, with size and material filters, while keeping a useful product page live during a stock shortage.

Repeat these tests on the public host after release, using a fresh session and a cached response wherever the application caches the page output.

Then measure the affected page group. Compare indexing, non-brand search traffic and organic transactions, with Australia separate from the US. Keep merchandising changes alongside the release record, because a stock outage can explain lost sales even when the technical SEO changes work as intended.

Searchmaxxed includes Search Console and GA4 setup checks in its managed work. The point is to connect technical changes with business results, rather than celebrate a clean crawl report while product discovery stays flat.

Is Searchmaxxed right for your Commerce Cloud SEO?

Searchmaxxed connects technical SEO with ongoing page improvements and measurement, while its Custom & Enterprise scope can include dedicated engineering, multiple markets and custom reporting.

Large catalogues need that combination. Fixing a canonical rule won't supply missing product specifications, and publishing better category copy won't repair broken rendering.

Our managed work includes implementation across pages, technical SEO, schema and internal links, with ongoing page releases through Growth System and Agentic Search Engineering. Custom & Enterprise is the scope to discuss when regional storefronts and engineering requirements need a different arrangement.

Keep the platform boundary explicit. If you need a Salesforce-certified implementation partner for checkout, order management or a full platform rebuild, confirm those requirements separately. Our technical SEO scope shouldn't stand in for that procurement check.

Bring the storefront architecture, affected categories and recent release history so we can discuss the pages customers struggle to find and the changes behind them.

Frequently asked questions