DE
EN
1. Accidentally blocking everything: robots.txt, Meta-Robots, HTTP headers
Staging has gone live. Disallow: / stayed in. Or a build sets <meta name="robots" content="noindex,nofollow"> global. Sometimes even a reverse proxy shoots via header. X-Robots-Tag: noindex Across. Crawlers are gone. Sales are following suit.
Your checkAutomate checks in CI/CD. Test for "noindex" in HTML and headers, check... robots.txt on Disallow: /, and set staging via Basic-Auth or IP restriction instead of Noindex in Production.
# robots.txt (Production) User-agent: * Disallow: /checkout/ Disallow: /account/ Allow: /media/ Sitemap: https://www.deinshop.de/sitemap.xml2. Block resources: JS, CSS, media
If crawlers are not allowed to load CSS and JS, they have a poorer understanding of layout, lazy loading, and navigation. Rendering problems lead to misinterpretations. Content, LCP and CLS.
Your checkAllow static directories. Check the "Fetch as Google" or Search Console render reports. Do not block any sensitive assets, except for truly private paths.
# robots.txt User-agent: * Allow: /static/ Allow: /assets/ Allow: /bundles/ Disallow: /admin/You can find more information about JavaScript SEO in the Google documentation. Read the basics before you start tinkering with rendering again next week. JavaScript and Google Search.
3. Canonical Chaos: Duplicate Content due to filters, sorting, and pagination
Shops generate variations like confetti: ?sort=, ?color=, ?page=Endless facets. Without self-discipline, URLs compete internally.
Your checkUse self-referencing canonicals for page states that are allowed to rank independently. Do not index pages that are purely sorted. Use stable canonicals and clear parameters for pagination.
<link rel="canonical" href="https://www.deinshop.de/kategorie/" /> <!-- Seite 2 der Kategorie --> <link rel="canonical" href="https://www.deinshop.de/kategorie/?page=2" /4. Hreflang incorrectly wired
Internationalization sites love hreflang, until traffic disappears. Common mistakes: references to non-canonical URLs, incomplete backlinks, incorrect country codes, missing x-default.
Your checkAlways set hreflang to the canonical version and link them reciprocally. Automate the generation in the build.
<link rel="alternate" href="https://example.com/de/produkt" hreflang="de-DE" /> <link rel="alternate" href="https://example.com/at/produkt" hreflang="de-AT" /> <link rel="alternate" href="https://example.com/en/product" hreflang="en" /> <link rel="alternate" href="https://example.com/" hreflang="x-default" />
SEO Shop Cleanup – E-Commerce News – Tips & Tricks – 🛠️ Why shop developers often unknowingly destroy SEO🚦
5. JavaScript rendering eats content
Important text, links, and meta tags only appear after client-side rendering. If Google throttles resources or the app is slow to update, the bot sees too little. Result: poor indexing.
Your checkDeliver critical content rendered server-side. Progressive enhancement is your friend. Check if <title>, <meta description> and the main content is in the initial HTML.
6. Ignore Core Web Vitals
LCP, CLS, and INP depend on real user data. Missing image dimensions, dynamically loading web fonts, third-party scripts in the main thread, and gigantic JS bundles harm the system. Performance .
Your checkSet dimensions for images and placeholders. Load fonts with font-display=swap and preload. Split bundles, remove unused JS. Miss Web Vitals in production with RUM.
<img src="/media/hero.webp" width="1280" height="720" alt="Sommer-Sale">A concise Canonical how-to will help you clean up the parameter hell. Read the guide: SISTRIX: Canonical Tag Explained.
7. Lazy loading without fallback
Only JavaScript lazy loaders without loading="lazy" or <noscript> risk blank pages for crawlers and users without JS.
Your checkUse native lazy loading and add a <noscript>-Fallback.
<img src="/img/p1.webp" alt="Produkt 1" loading="lazy" width="600" height="600"> <noscript><img src="/img/p1.webp" alt="Produkt 1" width="600" height="600"></noscript>8. Pagination and filters without a clean information architecture
Endless scrolling, unclear URLs, and filters that filter out every facet click as an indexable page. This produces thin content and cannibalization.
Your checkDefine which filters are indexable. Use consistent URL parameters. Implement good internal pagination linking. Only offer a "Show all" page if it performs well.
9. Soft 404s and broken redirects
Product offline? Please, no 200 errors with "Sorry, sold out." That looks like a soft 404. Or a string of 302 errors after a relaunch. Both cost budget and trust.
Your checkSet 404/410 correctly. Use clean 301s to successors. Avoid chains and loops. Check regularly with Crawl.
# NGINX Beispiel location = /altes-produkt { return 301 https://www.deinshop.de/neues-produkt; }10. Sitemaps as an Afterthought
Ancient URLs, 404 errors, and noindex pages in your sitemap? Then Google trusts your signals less.
Your checkGenerate sitemaps daily. Include only indexable, canonical URLs. Keep each file under 50.000 URLs and 50 MB. Separate by type.
Tip: Use our free SEO audit tool here for a website check
SEO Audit Tool
Comprehensive website analysis with real performance data & AI optimizations
11. Structured data with errors
Product, Offer, Breadcrumb, FAQ, Organization. It's all there, but incorrectly linked or contradictory to the visible data.
Your checkGenerate JSON-LD consistently from the same data sources as the frontend. Validate using rich result testing. Keep prices, availability, and currency synchronized.
<script type="application/ld+json">{ "@context":"https://schema.org","@type":"Product", "name":"Basic Hoodie","sku":"BH-001", "offers":{"@type":"Offer","priceCurrency":"EUR","price":"49.90","availability":"https://schema.org/InStock"} }</script>12. Migration without a redirect plan
New architecture, new URLs, old rankings gone. Without 1:1 redirect mapping, you lose link juice and users.
Your checkCreate a mapping from old to new URLs. Test with 100% coverage of critical pages. Transfer metadata, hreflang, canonical tags, and structured data.
13. Trailing slashes, upper/lower case, http/https
Small inconsistencies create large duplicates. Example: /Kategorie vs. /kategorie, /pfad vs. /pfad/.
Your checkEstablish a standard. Enforce server-side redirects. Consistently set canonical tags. Test the internal link graph with a crawler.
14. Session IDs and tracking parameters in links
If internal links contain UTM parameters or session IDs, the URL space is unnecessarily multiplied.
Your checkRemove such parameters from internal links. Use server-side attribution or first-party storage instead of URL overload.
15. Internal search indexed
Search results are sparse, volatile, and do not belong in the index.
Your checkBlock internal search via robots.txt and set noindex on search result templates.
Disallow: /search Disallow: /suche16. Cookie banners and overlays that obscure content
If a consent layer obscures content or makes links unusable, user signals decrease. Bot detection should hide the banner, but cleanly, without cloaking.
Your checkServer-side consent tuning, no blocking of main content. Ensure that HTML content remains accessible regardless of the banner.
17. Product variants incorrectly modeled
Sizes and colors as separate indexable pages without added value create cannibalization.
Your checkBundle product variants on one strong product page. If separate pages are needed, provide clear, standalone content or set the canonical tag to the main variant.
18. Images without alt text, incorrect format, no CDN
Images help rank, speed up search results, and provide information. Without alt text, using huge JPEGs instead of WebP/AVIF, and without caching, you're throwing away performance and search traffic.
Your checkUse modern formats, contextual alt text, width/height settings, and sensible caching. Deliver responsively. <source>-sets.
<picture> <source srcset="/img/p1.avif" type="image/avif"> <source srcset="/img/p1.webp" type="image/webp"> <img src="/img/p1.jpg" alt="Herren Sneaker weiß" width="800" height="800"> </picture>If you want to delve deeper into hreflang, the Ryte wiki provides a good overview: Ryte Wiki: Hreflang.
19. HTTP headers and caching rules conflict
Misconfigured proxies or CDNs overwrite metadata. Suddenly, your category page has an outdated canonical tag or an incorrect cached status code.
Your checkVersion HTML, differentiate cache policies for HTML vs. assets, and precisely invalidate after deployment.
20. Monitoring is lacking
Without alerts, you won't notice problems until revenue disappears. You need metrics for indexing, status codes, internal linking, web vitals, and SERP clicks.
Your checkSet up health checks. Automate crawls after deployment. Log bot traffic. Build dashboards for 404 trends, 5xx spikes, and sitemap errors.
Practical quick wins for your next sprint planning
Checklist “Before going live”
- robots.txt checked, staging remnants removed
- No global noindex, clean header
- Core URLs rendered server-side
- Canonicals consistent, parameterization defined
- Sitemap up-to-date, only indexable URLs
- Redirect mapping is available and tested
- Hreflang validated, reciprocally
- Web-Vitals RUM active, budgets set
What you automate in CI/CD
- Linting for Meta-Robots and Canonical
- Snapshot tests for critical templates
- Crawl smoke test after deployment
- Sitemap validation and link graph checks
I could write a book about SEO fails. My favorite: A developer builds an amazing checkout process. One problem: Everything runs through /checkout#step1, /checkout#step2, etc. Google sees: One URL. We lose: All conversion data in Analytics. Tracking is broken, optimization is impossible.
I work as an SEO consultant and am constantly cleaning up after developers. The sad thing is: 80% of the mistakes could have been avoided. A little basic knowledge would be all it takes! But many developers are resistant to advice. I constantly hear, "We've always done it this way." Yeah, and that's why the shop ranks on page 8 of Google!
Super important topic! What particularly annoys me is that many developers don't even understand the basics. I recently had a discussion with a 'senior developer' (10 years of experience!) who tried to explain to me that meta keywords are super important for SEO are. In 2025! When I explained to him that Google has been ignoring them since 2009, he was completely surprised. How can you develop online stores for 10 years and not know something like that? This illustrates the fundamental problem: SEO is completely ignored in developer training. Yet it's essential! An online store without SEO is like a shop without an entrance – technically perfect, but nobody can get in.
We once paid €20 for an 'SEO-optimized' shop. What we got: – Title tags all the same ('Shop | Shop | Shop') – Auto-generated meta descriptions with the first sentence from Content – No structured data – Images without alt tags – URLs with session IDs – No XML sitemap. When we complained, the agency said: 'But the keywords are in the content!' *facepalm*
We had a developer who delivered all images as WebP – without a fallback. Older browsers? Tough luck. Safari users? Can't see any images. Traffic plummeted by 30% because people were leaving immediately. SEO It's not just about Google optimization, but also about user experience!
I am a developer myself and I have to admit: SEO SEO was a long-standing blind spot for me. I always thought it was just marketing stuff. Until I saw the analytics for one of my shops. 90% bounce rate, average session duration 8 seconds. The shop was technically perfect – but an SEO disaster. Since then, I've been intensively studying the topic. What I've noticed is that there are hardly any good resources for developers! Most SEO guides are written for marketing people. We developers need concrete, technical instructions. Not just "Make good content," but "How to properly implement structured data," "How to optimize your crawl budget," "How to build an SEO-friendly URL structure." This article is a good start, but we need more like it!
Great article! I'd like to add: Infinite scrolling without a pagination alternative. Google only sees the first 20 products. The other 480? Invisible! Our developer: 'But that's modern UX!' Yeah, great, except that 96% of our products weren't indexed. Modern UX is useless if nobody can find the shop!
We have found that many developers SEO They see SEO as a 'brake'. 'It slows down the shop,' 'It complicates the code,' 'It's old school.' But modern SEO is the exact opposite! A well-optimized shop is faster, more user-friendly, and technically cleaner. The problem is the training. In most coding bootcamps, SEO isn't mentioned at all. People learn React, Node, Docker – but not how search engines work. This comes back to haunt them later. We now have a dedicated SEO consultant who is involved in EVERY shop project from the very beginning. Does it cost more? Yes. Does it save money in the long run? Absolutely!
Something that's often forgotten: Mobile First! Our developer optimized the shop perfectly for desktop. Mobile? 'We'll do that later.' But Google indexes mobile first! The rankings completely crashed because the mobile version was awful. No responsive images, horizontal scrolling, tiny buttons. A nightmare!
This article really resonates with me! I've been in e-commerce for 15 years and have seen it all. The craziest thing was a developer who claimed to have developed a 'brilliant SEO innovation'. His idea: All Product texts I wanted to generate dynamically using JavaScript, based on keywords. The problem? Google couldn't read any of it! The whole thing Content It was rendered client-side, without SSR, without fallback. 8.000 product pages, completely invisible to search engines. When I explained this to him, his response was: "But Google can handle JavaScript!" – Yes, it CAN, but not always and not immediately! It took us six months to fix it. Since then, I insist that every online store project has an SEO expert involved from day one. There's simply no other way.
I'm a shop owner and SEO enthusiast. What annoys me most: developers who think a fast shop is enough for good rankings. Yes, speed is ONE factor out of over 200! What good is a shop that loads in 0,5 seconds but has no proper title tags, no internal linking, no structured data? Last year we had a developer who was super proud of his Core Web VitalsEverything's green! Except: He'd forgotten to set up the XML sitemap, the robots.txt file was misconfigured, and he'd omitted the breadcrumb navigation 'for performance reasons'. I had to explain to him that breadcrumbs aren't just important for users, but also for Google! His comment: 'Oh, I didn't know that.' – Yes, that's exactly the problem! Developers finally need to understand that SEO It's not a nice-to-have, but essential for the success of an online shop. Without visibility, there are no customers; without customers, there is no revenue. It's that simple.
We once had a developer who thought robots.txt was a "security risk" and simply set everything to Disallow. After two weeks, we'd completely disappeared from Google. True story! When we confronted him about it, he just said, "But the shop is still accessible, right?" – Yeah, sure, but nobody can find it anymore! The worst part: To this day, he still doesn't understand what the problem was. Some developers really live in their own bubble.
Interesting point about JavaScript frameworks! We used Vue.js and had massive indexing problems. Google couldn't even crawl some of our product pages. The solution was server-side rendering, but that cost another three months of development. We could have known that beforehand…
I work as a developer myself and unfortunately have to admit: the article hits the nail on the head. During my training, I came across... SEO Exactly zero. We've learned how to build high-performing, secure, and beautiful online stores – but nobody has explained to us how Google actually works. The problem is systemic: Most developer bootcamps and degree programs treat SEO as a 'marketing topic,' not as a technical requirement. Yet, 80% of technical SEO is developer work! Canonical tags, structured data, Core Web VitalsCrawl budget optimization – these are all developer topics. I only acquired this knowledge myself after three years of professional experience, after realizing how many client shops I had unknowingly ruined through optimization. My suggestion: SEO should be a mandatory component of every web developer training!
Finally, someone said it! We relaunched our online store last year – €45.000 budget, great agency from Hamburg. The store looks amazing, runs flawlessly, superb UX. But: After three months, 60% of our rankings were gone! The developers had simply changed all the URLs, didn't set up any redirects, deleted meta descriptions, and completely messed up the entire site structure. We then had to spend another €15,000 on SEO recovery. My takeaway: NEVER again develop an online store without SEO support!