10 Technical Mistakes in Turkish SEO and Practical Solutions
Errors like canonical tag mistakes, missing hreflang, and the i-İ problem on Turkish sites can reduce organic traffic by up to 40%. Here are the solutions.
Over the past 2 years, I've conducted technical SEO audits on 6 different Turkish sites. The common thread was this: the same 3-4 errors repeated in almost every project. One e-commerce site could only show 12% of its 18,000 pages to Google due to canonical tag mistakes. Another news site was showing .de domain competitors to Turkish readers in Germany because it didn't use hreflang.
Most SEO agencies in Turkey focus on content and backlinks. Technical infrastructure is either ignored or remains superficial. Yet according to Google's 2024 report, 73% of technical errors waste crawl budget. In this article, I'll share the 10 most common technical mistakes I've seen on Turkish sites and their correction methods. Each includes real case examples and code snippets.
1. Incorrect Canonical Tag Implementation
The most common error I see on Turkish e-commerce sites: separate URLs for each product variant, but no canonical tags. For example, in the diolivo.com.tr project, there were initially 2,400 product pages, but Google was only indexing 890 of them. Reason: URLs created for color and size filters like /urun?renk=kirmizi, /urun?beden=m.
The solution was simple:
- We added self-referencing canonical to the main product page:
- We used the same canonical on all variant URLs
- Within 6 weeks, the number of indexed pages increased from 890 to 2,180
Another common mistake: pagination canonical. They make the canonical for /blog?sayfa=2 point to /blog. In this case, Google can't see the content on the 2nd page at all. Correct usage: each page should have a self-referencing canonical to its own URL, prev/next tags can be added (no longer mandatory but helpful).
Checking Method
Scan all pages with the "Canonical" filter in Screaming Frog. If there's a "Non-Indexable Canonical" warning, it means the canonical URL itself is non-indexable. This is a serious error.
2. Complete Absence of Hreflang Tags
The thing that surprised me most while serving from the Netherlands to Turkey: 90% of Turkish sites don't have hreflang. They say "we only use one language anyway." But what's forgotten: there are over 5 million Turkish speakers in Germany, the Netherlands, and Belgium.
The kamupersonelhaber.com project initially had no hreflang. Turkish users searching on Google.de were seeing German news sites. After adding this implementation, traffic from Germany increased 180% in 3 months:
Important detail: you need to add both tr-TR (Turkey Turkish) and tr (general Turkish). x-default is the fallback for users without geographic targeting.
Common Hreflang Errors
- Missing reciprocal reference: page A points to B but B doesn't point to A
- Forgetting self-referencing hreflang: the page should also specify its own language
- Wrong language codes: "tr-tr" instead of "tr-TR" (ISO 639-1 standard)
You can see hreflang errors in the "International Targeting" section of Google Search Console. You typically get a "Return tag missing" warning.
3. Turkish Character Issues (i-İ Problem)
This is a headache specific to Turkish. Google treats "istanbul" and "İstanbul" searches the same, but problems can arise in URL slugs. We experienced this error on memuratamalari.com:
- /İlan-detay/memur-alımı → 404 error
- /ilan-detay/memur-alımı → works
Reason: the server was processing URLs as case-sensitive. Solution:
1. We converted all URL slugs to lowercase (for WordPress: Settings → Permalinks → Custom Structure) 2. We converted Turkish characters to ASCII: ı→i, ş→s, ğ→g 3. We added a 301 redirect rule: uppercase URLs redirect to lowercase
Another issue: Turkish characters in title and meta description sometimes appear broken. Charset should be set to UTF-8:
Using mb_string in PHP is also important. I prefer the mb_strtolower() function in all my projects because it handles Turkish characters correctly.
4. Incorrect Robots.txt Configurations
Last month a client came saying "Google isn't indexing us." I looked at robots.txt:
User-agent: * Disallow: /
The entire site was blocked. The developer added it while testing in the staging environment and forgot to remove it when moving to production. This is an extreme example, but similar errors are very common:
- /wp-content/ is blocked instead of /wp-admin/ (theme and plugin files are inaccessible)
- /kategori/ is blocked but product pages depend on these categories (internal link value is lost)
- Sitemap.xml is not specified in robots.txt
A correct robots.txt should look like this:
User-agent: * Disallow: /wp-admin/ Disallow: /sepet/ Disallow: /odeme/ Allow: /wp-admin/admin-ajax.php
Sitemap: https://sitename.com/sitemap.xml
The doktorbul.com project has 79,000 doctor profiles. We disallowed /doktor-ara?filtre= parameters in robots.txt because they were creating duplicate content. But we left the actual profile pages (/doktor/dr-ahmet-yilmaz) open with allow.
Testing Method
Test your URL in Google Search Console → Robots.txt Tester. If a URL shows as "Blocked" but you want it indexed, review your rules.
5. XML Sitemap Deficiencies and Errors
Most Turkish sites use Yoast or RankMath, which automatically generate sitemaps. But they're not checked. On italyanmutfagi.com, there were 618 recipes, but only 340 were in the sitemap. Reason: old recipes remained in "draft" status.
Common sitemap errors:
- URLs returning 404 in the sitemap (Google marks these as "Submitted but not found")
- More than 50,000 URLs in a single sitemap (Google's limit is 50,000)
- Incorrect lastmod date (always showing "today")
- Meaningless priority and changefreq values (all pages given 1.0)
I organize sitemaps like this:
- sitemap-posts.xml → blog posts
- sitemap-pages.xml → static pages
- sitemap-products.xml → products
- sitemap-index.xml → main index
Especially on e-commerce sites, I add
This increases visibility on Google Images. On diolivo.com.tr, after product images were added to the sitemap, visual search traffic increased 340% in 6 months.
6. Not Using Schema Markup
80% of sites in Turkey have no structured data. Yet Google uses schema.org markup for rich snippets. On the italyanmutfagi.com project, we added Recipe schema for each recipe:
After this addition, star ratings, cooking time, and calorie information started appearing in Google searches for "karnıyarık tarifi." Click-through rate increased from 18% to 34%.
Important Schema Types for Turkish Sites
- Article: for blog posts (headline, author, datePublished)
- Product: for e-commerce (price, availability, aggregateRating)
- LocalBusiness: for local businesses (address, telephone, openingHours)
- FAQPage: for FAQ pages (shows expandable FAQ box in Google)
- BreadcrumbList: for breadcrumb navigation
We added NewsArticle schema to every news item on kamupersonelhaber.com. This was very helpful in the Google News acceptance process.
7. Mobile-First Index Issues
Google has been using mobile-first indexing since 2021. That means it indexes the mobile version of your site. But some Turkish sites still show different content for desktop and mobile.
On one client site, there was an 800-word product description on desktop, only 200 words on mobile. Ranking dropped because Google indexed the mobile version. Solution: show the same content on every device with responsive design.
Another issue: lazy loading incorrectly implemented on mobile. Images don't load at all, Google bot can't see the images. Correct lazy loading:
But don't add lazy loading to images in the critical viewport (first screen). This lowers the LCP (Largest Contentful Paint) score in Core Web Vitals.
PageSpeed Insights Errors
Performance issues I frequently see on Turkish sites:
- Unnecessary plugins (the average WordPress site uses 30+ plugins)
- Unoptimized images (2MB PNG files)
- Render-blocking JavaScript (jQuery, Font Awesome, etc.)
I use the Cloudflare + WP Rocket combination in all FUTIA projects. On memuratamalari.com, with these optimizations, the mobile PageSpeed score increased from 38 to 82.
8. Lack of Internal Linking Strategy
Most sites use internal links randomly. Yet proper internal linking directs PageRank to strategic pages. On doktorbul.com, we established this structure:
- Homepage → 10 popular city pages
- City pages → specialty pages in that city
- Specialty pages → doctor profiles
Thanks to this hierarchy, even newly added doctor profiles started getting indexed within 2 weeks. Because each profile received links from 2-3 levels up.
Internal link anchor texts are also important. Use descriptive anchors like "dietitian doctors in Istanbul" instead of "click here." But don't overdo it: if you use the same anchor 50 times, Google may perceive it as "over-optimization."
Orphan Page Problem
Orphan page: a page that receives no internal links. Google has difficulty finding these. Check the "Orphan Pages" report in Screaming Frog. On italyanmutfagi.com, there were 43 orphan recipes; after adding them to category pages and the "similar recipes" widget, their traffic tripled.
9. Duplicate Content and Parameter Management
Filter and sorting parameters on Turkish e-commerce sites create duplicate content:
- /urunler
- /urunler?siralama=fiyat
- /urunler?filtre=marka-x
- /urunler?siralama=fiyat&filtre=marka-x
Google sees these as separate pages. Solution:
1. Canonical tag: all variants should have canonical to the main page 2. Google Search Console → URL Parameters: mark the "siralama" parameter as "doesn't change content" 3. Robots.txt: disallow non-critical parameters
On diolivo.com.tr, after making this adjustment, the number of indexed pages dropped from 6,800 to 2,400 (this is good because 4,400 duplicate pages were cleaned up). But organic traffic increased 23% because Google spent its crawl budget on the right pages.
Pagination Duplicate Content
Blogs have URLs like /sayfa/2, /sayfa/3. For these:
- Each page should have self-referencing canonical to its own URL
- Add rel="prev" and rel="next" (no longer mandatory but helpful)
- Don't add "noindex, follow" to meta robots (Google stopped recommending this in 2019)
10. HTTPS and Security Certificate Issues
Although this seems basic, there are still Turkish sites using HTTP. Google has been using HTTPS as a ranking factor since 2014. But some sites have switched to HTTPS but have mixed content errors:
→ HTTP → HTTPS
The browser shows a "Not Secure" warning, users don't trust it. Solution: convert all resources (images, CSS, JS) to HTTPS. The Really Simple SSL plugin does this automatically in WordPress.
Another issue: SSL certificate has expired. I use Let's Encrypt in all FUTIA projects, it automatically renews every 90 days. If you're using Cloudflare, set their SSL/TLS setting to "Full (strict)".
Redirect Chains
This error is very common in HTTP → HTTPS transitions:
http://site.com → http://www.site.com → https://www.site.com
3-step redirect chain. Each step adds 100-200ms delay. The correct way:
http://site.com → https://site.com (single step)
Use this code in .htaccess:
RewriteEngine On RewriteCond %{HTTPS} off [OR] RewriteCond %{HTTP_HOST} ^www\. [NC] RewriteRule ^ https://site.com%{REQUEST_URI} [R=301,L]
How to Conduct a Technical SEO Audit?
I follow these steps in every new project:
1. Google Search Console and Analytics integration (basic metrics) 2. Site scan with Screaming Frog (broken links, duplicate content, canonical errors) 3. Performance test with PageSpeed Insights (mobile and desktop) 4. Structured data check with Schema Markup Validator 5. Robots.txt and sitemap.xml review 6. Hreflang and canonical tag verification 7. Internal linking analysis (orphan pages, link depth)
When we conducted this audit on memuratamalari.com, we found 87 critical errors. After a 6-week correction process, organic traffic increased 140%. The biggest impact came from canonical tag and sitemap corrections.
If you suspect your site has similar technical issues, you can contact me via WhatsApp: +90 532 491 17 05. As FUTIA, we offer site + automation + monthly maintenance packages, and technical SEO auditing is part of this package. Alternatively, you can send a detailed email to info@futia.net; I usually respond within 24 hours.
Frequently Asked Questions
How much do technical SEO errors affect organic traffic?
In my experience, serious technical errors can reduce organic traffic by 30-60%. Especially canonical tag mistakes, robots.txt blocks, and indexing issues are very critical. For example, on an e-commerce site, due to incorrect canonical usage, only 12% of 18,000 pages were being indexed. After this was corrected, the number of indexed pages increased 2.5 times within 6 weeks. But not every error creates equal impact. Schema markup deficiency doesn't cause traffic loss, but rather leads to lower click-through rates.
Is it mandatory to use hreflang on Turkish sites?
If you only target users in Turkey, it's not mandatory. But there are over 5 million Turkish speakers in countries like Germany, the Netherlands, and Belgium. If you don't use hreflang, when these users search on Google.de or Google.nl, they may see your local competitors. On a news site, after adding hreflang, traffic from Germany increased 180% in 3 months. Implementation isn't difficult either; just add 3 lines of code to the <head> section of each page. I recommend using tr-TR, tr, and x-default values.
What's the difference between canonical tag and 301 redirect?
301 redirect physically redirects the user and bot to another URL. The browser address bar changes. Canonical tag only sends a signal to Google saying 'the original version of this page is here,' the user stays on the current page. Canonical is preferred for duplicate content situations. For example, for product variants (/urun?renk=kirmizi, /urun?renk=mavi), you mark them all with canonical to the main product page. But if you're permanently moving an old URL, you should use 301 redirect. If a page has both 301 and canonical, 301 takes priority.
Do I need coding knowledge to add schema markup?
No, if you're using WordPress, plugins like RankMath or Yoast automatically generate schema. But I prefer adding JSON-LD manually because it provides more control. You can use Google's Schema Markup Generator tool; you fill out the form, it gives you ready-made code. You paste this into your site's <head> or <body> section. After adding, definitely check with Google's Rich Results Test tool. It will show if there are errors. On a recipe site, after adding Recipe schema, the click-through rate increased from 18% to 34% because star ratings and cooking time started appearing in search results.
How often should technical SEO audits be conducted?
It varies depending on site size. For a 100-500 page site, once every 6 months is sufficient. But for projects with constantly added new content like e-commerce or news sites, I recommend checking monthly. Especially urgent audits are needed in these situations: after theme changes, after hosting changes, if traffic suddenly drops by more than 20%, if there's a sudden increase in errors in Google Search Console. In FUTIA projects, I do automatic monitoring as part of monthly maintenance. Screaming Frog scans, GSC error tracking, and Core Web Vitals checks run routinely. If we detect a critical issue, we notify the client within 24 hours.
Want to apply one of the techniques from this post? Fill out a short form and we'll email you a free preview audit within 48 hours.