Technical SEO Consultant, Germany

Umlauts, compound words and URL slugs

German creates URL questions that English sites never face. Umlauts and the sharp s can be kept as characters, transliterated (ä as ae, ö as oe, ü as ue, ß as ss) or simply stripped, and each CMS has its own default. Compound nouns add a second decision: a slug for Lagerverwaltungssoftware can be one long word or split as lagerverwaltung-software, and searchers may type either form or a version with spaces.

Search engines can read encoded characters, so the risk lies in inconsistency. Sites that have changed CMS or plugins often carry several slug styles at once, producing duplicate URLs, broken internal links and redirect chains. The review covers:

• An inventory of slug styles across templates, including uppercase variants and encoded characters.
• Whether variants resolve to one canonical URL through permanent redirects.
• How sitemaps, hreflang, feeds and internal links write each URL.
• A rule for compounds: keep the form your buyers search for, which keyword research will show, and split with hyphens only where that matches real usage or helps readability.

The simplest durable rule is transliterated, lowercase slugs with hyphens, generated by the CMS and editable by authors. Changing established URLs needs a redirect map, so I recommend it only where the benefit is clear. Language and country questions across Germany, Austria and Switzerland belong to my international SEO work for Germany.

Shopware, JTL, Magento and ERP-fed catalogs

German ecommerce runs on a mix of platforms. Shopware and JTL both come from Germany, and JTL-Shop is designed to work with the JTL-Wawi merchandise management system; Magento and Adobe Commerce, Shopify, WooCommerce and custom builds on top of an ERP also appear. Technical patterns differ by platform:

• Shopware: category and product URLs generated by templates that can change between major versions, variant handling, filter parameters and plugins that add their own routes.
• JTL-Shop: product data synchronized from JTL-Wawi, where attribute and category settings in the merchandise system decide what reaches the page.
• Magento and Adobe Commerce: store views per language or country, layered navigation and indexing settings that can multiply URLs.
• ERP-fed B2B portals: products showing internal article numbers in titles, missing descriptions and attribute data that never reaches the template.

For manufacturers and wholesalers the catalog is the most valuable part of the site, so I trace issues back to the data source rather than patching templates. Discontinued articles deserve particular care: buyers still search for old part numbers, so a page explaining the successor product often serves them better than a redirect to a category. If data flows between ERP, PIM and shop need redesign, my system integration work in Germany covers that side.

Datasheets, PDFs and filter pages for technical products

German industrial and B2B sites publish technical documentation in volume: datasheets, operating manuals, safety data sheets, certificates, declarations of conformity and CAD files. Buyers and engineers search for these documents directly, often by article number or standard, so they can bring valuable visits. They also cause technical problems when every version of every PDF is indexed separately.

A document strategy for technical sites:

• Index current datasheets and manuals that answer real searches, each with a descriptive title in the file metadata and a link back to the product page.
• Give important documents an HTML product or download page that summarizes them, so search engines and visitors get context.
• Keep superseded versions out of the index, or redirect them to the current file, so outdated specifications do not rank.
• Exclude internal, customer-specific or login-protected files from crawling, and check that nothing confidential is exposed.

Filter pages need similar discipline. Technical catalogs often filter by dimensions, material, pressure rating, voltage or standard, creating enormous numbers of combinations. Only those matching real search patterns should be indexable; the rest stay usable without being crawled. Server logs show whether crawlers spend their time on documents and filters or on the products that matter. The manufacturing industry page describes the systems behind such catalogs.

Consent tools, self-hosted resources and page speed

Data protection expectations in Germany are strict, and site owners tend to respond conservatively: a consent tool on every page, scripts held back until consent and, after public discussion about loading fonts and other resources from third-party servers, more resources hosted on the site itself. Those choices have performance consequences that are easy to overlook.

The review starts with what real visitors experience in Core Web Vitals field data, then looks at causes:

• The consent tool: how it loads, whether its banner reserves space or pushes content down, and whether it delays the main content.
• Scripts released after consent, such as analytics, advertising, maps, video embeds, chat and review widgets, with an owner and a loading rule for each.
• Self-hosted fonts and libraries: whether they are compressed, cached properly and limited to the weights and character sets actually used.
• Click-to-load placeholders for embeds, which help privacy and speed if built without layout shifts.

Where the site is hosted, sometimes chosen in Germany for data protection reasons, and how a CDN serves visitors also affect response times. Which tags may run, and what the banner must say, are questions for your data protection officer; my work is the speed and measurement side. My Core Web Vitals consulting page explains the method in detail.

Relaunches, log files and working with German agencies

In German companies, a website relaunch is a familiar project: new design, new CMS or shop system and new URL structure, often all at once. It is also the moment when visibility built over a long period can drop, because redirects are incomplete, templates lose text or structured data, and staging settings such as noindex reach production. My SEO migration process writes SEO requirements into the specification (the Lastenheft or Pflichtenheft, if your project uses them), builds and tests the redirect map, crawls staging against the old site and checks production on launch day.

Server logs add evidence a crawler cannot. They show which URLs search engine bots request, how often and with what status, revealing crawl waste on filters, parameter combinations and outdated documents. Hosting contracts and data processing agreements govern how logs are shared; I specify an export with only the fields needed, and personal data such as IP addresses can be shortened or removed beforehand.

Implementation stays with your developers or agency. I write tickets in clear English with evidence, the change and an acceptance test, join review calls in German business hours and recheck fixes after deployment. Everything is remote. My general approach is set out under technical SEO, and the Germany overview gathers my other services there.