Study guide · SEO and Target Groups
Staying visible: SEO, redirects and social media in TYPO3
A new website is of little use if it drops out of the search results after the relaunch. As a consultant, you do not have to do SEO yourself. But you do need to know the risks, explain to the client what TYPO3 offers, and build the right steps into the plan and the budget.
Our example is the Seetal holiday region. It is moving from an old system to TYPO3 14. Its old hiking page wandern.php ranks high on Google, and many hiking portals link to it. Such links from other websites are called backlinks. In the new system the page lives under the path erleben, wandern. And every day trip tip should look good when someone shares it on Facebook.
SEO and GEO: what machines need
SEO stands for Search Engine Optimization. It covers every measure that makes a website more visible in search engines. That includes technology such as loading time, clean URLs and a sitemap, content such as good texts, titles and headings, and trust through backlinks. TYPO3 supports this with the system extension seo.
GEO stands for Generative Engine Optimization. It targets AI-powered search systems that generate an answer from web content instead of just listing links. To get them to pick up and quote your content, the following helps:
- Answer typical questions clearly and directly, for example in a section with frequently asked questions.
- Structure the text clearly, with meaningful headings.
- Back up statements with sources, figures and quotes.
- Keep content up to date and make the author and the date visible.
Some tricks do harm instead. Invisible keywords in white text count as manipulation. Machines struggle to read text that only exists as an image. Many nearly identical pages for every search term create duplicate content.
Status codes: what the server answers
Every response from a web server carries a three-digit HTTP status code. The first digit tells you the group: 2 for success, 3 for redirects, 4 for client errors and 5 for server errors.
- 200 means the request was answered successfully.
- 301 Moved Permanently means the page has moved for good. Search engines update their index and pass the ranking signals on to the new URL.
- 308 Permanent Redirect is permanent as well. Unlike 301, the request method is kept, for example a POST from a form.
- 302 Found and 307 Temporary Redirect are temporary. The old URL stays in the index. 307 keeps the request method.
- 403 means forbidden, 404 means not found. With 404 it stays open whether the page will come back.
- 410 Gone says explicitly that the page was removed on purpose and for good.
- 500 stands for an error on the server. 503 reports temporary unavailability. Search engines then fetch the page again later. That is why 503 suits maintenance work.
Many 404 errors, for example after a relaunch without redirects, hurt the user experience and SEO.
Redirects and their pitfalls
A redirect automatically takes visitors and search engines from one address to another. The server answers with a code such as 301 or 307 and names the new address. The browser follows on its own. This keeps old links, bookmarks and search results working.
Badly set redirects do harm, though:
- A chain is several redirects in a row, for example from A via B to C. That costs time. It is better to send A straight to C.
- A loop goes round in a circle, for example from A to B and back to A. The browser never arrives and gives up. The page is broken.
- If you send old pages to the home page across the board, the target does not match the content. Search engines often treat this like a missing page, a soft 404. The ranking signals are lost.
- A temporary code for a permanent move delays the transfer of rankings.
Title, description and URL
With the seo extension, every page gets the SEO tab in its page properties.
The Title for search engines ends up in the title tag. It often appears as the headline in the search result. If the field stays empty, TYPO3 uses the normal page title. In addition, TYPO3 adds the website name from the site configuration by default. In the Core it comes first, separated by a colon. An integrator can change the order and the separator.
The meta description briefly describes what the page is about. It does not appear on the page itself. Search engines often use it as the text excerpt below the title, the snippet, but they may also pick an excerpt of their own. So title and description shape the search result, and with it the click rate. If the description is missing, the page is still indexed.
Handy for editors is the dashboard widget “Pages missing meta description” from the seo extension. It lists all pages without a description and links straight to editing. You need access to the dashboard and to the widget.
The URL segment counts too. TYPO3 generates it from the page title, and you can adjust it. A short, readable segment with the most important search term is good, without strings of numbers, spaces or capital letters. For Seetal it is simply wandern.
Robots, canonical and duplicate content
The SEO tab also holds the robots instructions. They are aimed at search engines, not at visitors.
- If “Index this page” is set to no, TYPO3 writes noindex into the HTML head. Search engines then leave the page out of their index, and TYPO3 leaves it out of the XML sitemap.
- If “Follow this page” is set to no, TYPO3 writes nofollow. Search engines should not follow the links on the page. Visitors can still click them.
Duplicate content means the same or very similar content is available under several URLs. Search engines then do not know which version to show, and the ranking signals are split. Canonical links, redirects and a clean URL structure fix this.
A canonical link is a reference in the HTML head with the attribute rel set to canonical. It names the preferred URL of a piece of content. With the seo extension, TYPO3 automatically outputs a canonical link to the current page. In the SEO tab you can enter a different original. TYPO3 leaves such a page out of its own sitemap.
With SEO extensions, check who generates canonical links. If the Core and an extension both output one, there are two in the head. That can confuse search engines. Only one source should be in charge.
Sitemap and language versions
An XML sitemap is a file that tells search engines about the pages of a website and its structure. It lists the URLs, optionally with the date of the last change, the change frequency and the priority. It has nothing to do with a visual map for visitors. The content element “Sitemap” only creates a menu.
Since TYPO3 13.3, the seo extension ships the site set “SEO Sitemap”, technically typo3/seo-sitemap. Once you add it to the site, the sitemap is available as sitemap.xml directly after the domain. Since TYPO3 14 the site set also brings the routes it needs, so no entry in the site configuration is required. You adjust its settings in the settings editor of the site.
In the SEO tab you maintain the change frequency and the priority of each page. The priority says how important the page is compared with other pages of the same website. The default is 0.5. Both are only hints that search engines may ignore. A priority of 1.0 does not lift a page above pages of other websites.
On multilingual sites, TYPO3 with the seo extension automatically outputs hreflang links to the translated versions of a page. The values come from the language configuration of the site and from the existing translations. A canonical link to the default language would be wrong, because translations are content in their own right.
Two kinds of metadata
There are two kinds of metadata.
- HTML metadata sits in the head of the page: title, description, robots, canonical, hreflang and the tags for social networks.
- Structured data describes content in a machine-readable way using Schema.org, for example events, products or organizations. The usual format is JSON-LD inside the HTML. Search engines build rich results from it, such as dates or star ratings.
The Core offers no editor fields for structured data. The output comes from the template or from extensions. So for the Seetal events, the consultant plans it as a development task.
Meta tags can come from several sources. TYPO3 sets them in a fixed order: plugins first, then the page properties of the seo extension, and TypoScript with page.meta last. A tag that is already set is not overwritten by a later source. So the page properties take priority, and page.meta acts as a default for pages without their own value.
If TypoScript should win, the integrator sets the option replace to 1. It only replaces the value if TypoScript actually delivers one. An empty value does not remove an existing tag.
page.meta.description = Holidays in the Seetal region
page.meta.description.replace = 1Headings and images
Headings form the outline of a page. Search engines recognize the main topics from them, and screen readers recognize the structure. That is why a logical hierarchy from h1 to h6 matters: one h1, below it h2, below that h3. The body text of course still counts.
The level of a heading belongs to its meaning, not to its looks. The looks are handled by the CSS in the template. Editors choose the level in the content element or in the rich text editor. Integrators can sensibly restrict the choice.
Images need metadata too: an alternative text, a meaningful file name and a suitable size. You maintain these in the Media module.
Social media: shaping shared links
Social media platforms are online channels where people share content, comment and connect with each other, such as Facebook, LinkedIn, Instagram or X. When someone shares a link there, the platform shows a preview with a title, a description and an image.
The Open Graph standard controls this preview with the tags og:title, og:description and og:image. Facebook and LinkedIn read them. X uses its own tags, the X Cards. The XML sitemap and robots.txt, on the other hand, are aimed at search engines, and hreflang at language versions.
In TYPO3 you maintain these values in the “Social media” tab of the page properties. It also comes from the seo extension and has fields for title, description and image, both for Open Graph and for X Cards.
- The fields are independent of the title for search engines. If a page should show a different title on LinkedIn, you only enter your own Open Graph title.
- If the fields stay empty, the page can still be shared. The network decides on its own and often uses the search engine title and the description.
- A dedicated image makes the post stand out. The networks expect certain aspect ratios. TYPO3 ships common crops, and the integrator adds more. The image has no effect on the ranking in search engines.
Redirects in TYPO3
Redirects are managed by the system extension redirects. In TYPO3 14 you find them in the module Sites, Link Management, Redirects. Next to it are Short URLs and QR Codes. Short URLs are short, easy-to-share addresses that redirect to pages, files or external URLs, for example for business cards. You can change the target later without changing the shared address. QR codes do the same for printed material.
A redirect consists of a source and a target:
- The source domain is an asterisk by default. This wildcard makes the rule apply to all sites of the installation. In a multi-site setup it can hit other brands by accident. So you usually pick the specific domain.
- The source path is the old address, optionally as a regular expression. Query parameters are not taken into account by default.
- The target is a page, a file or a URL. For a page, TYPO3 stores its ID.
- On top come the status code, an optional start and end date, and a hit counter.
Automatic redirects: When an editor changes the URL segment of a page, TYPO3 helps on its own. Two site settings control this, both switched on by default. With autoUpdateSlugs, TYPO3 adjusts the segments of all subpages. With autoCreateRedirects, it creates redirects for all changed pages. A notice then offers to undo everything or to remove only the redirects. No automatic redirects are created in workspaces.
Automatic redirects point to the page ID, in the format t3, colon, page with the uid. If the segment changes again, the old redirect still leads straight to the current target. This reduces chains and loops.
The status code: The site setting redirects.httpStatusCode is set to 307 by default, so temporary. It only applies to automatically created redirects. The documentation advises against 301 there. Browsers and search engines store permanent redirects, and the old URL can hardly be reused afterwards. But automatic redirects are often created in passing. Deliberate, permanent redirects may use 301, for example when a company moves for good from alt-firma.de to neue-firma.de.
Permissions: The documentation warns that the Redirects module opens up the redirects of the whole installation, including other sites. In TYPO3 14.3 the code filters this for editors. They only see and edit redirects whose source domain belongs to their sites or is an asterisk, and whose target they are allowed to reach. Admins still see everything. The risk remains, though: a rule with an asterisk applies to all sites, and mistakes such as loops can make pages unusable. So only give the module to trusted, experienced users.
To undo automatic redirects, an editor does not need the module. After their own change, they take them back through the notice. The documentation requires permissions on the table sys_redirect for this. In TYPO3 14.3, however, the code grants these permissions for this step on its own, just for a moment.
Maintenance: Over the years many redirects pile up. Redirects handled by PHP have limits, especially with regular expressions. The documentation recommends:
- Set a lifetime in days for new redirects with redirectTTL. The default 0 means no expiry.
- Clean up outdated redirects regularly with the command redirects:cleanup, for example by age and number of hits. For that, the hit counter has to be switched on as a feature toggle.
- Find loops with the command redirects:checkintegrity. The result also appears in the status report.
- Export very large numbers of redirects to the web server or load balancer configuration.
Since TYPO3 14, every new redirect stores who created it and when. The link validator can also report external links that redirect. The integrator switches this on in page TSconfig.
The relaunch roadmap
In a relaunch, many URLs often change. Without redirects, visitors land on 404 pages and rankings are lost. So the most important step is: every old URL gets a permanent 301 redirect to the matching new page.
- List all relevant old URLs, for example with a crawler, the old sitemap and Search Console. Pages with visitors and backlinks matter most.
- Keep content and page hierarchy as far as possible. A new structure changes URLs, internal links and how well search engines can crawl the site. Big rebuilds often cost rankings until search engines have picked them up.
- Map every old URL to the new one with matching content, not to the home page across the board.
- Create permanent redirects, one by one in the Redirects module or by import if there are many.
- Carry over titles and meta descriptions, or improve them on purpose.
- Avoid duplicate content with canonical links and only one domain variant.
- Update internal links and submit the new sitemap.
- After go-live, check for 404 errors and redirect chains.
Watch out at go-live: During development, the new website often runs on a staging domain with noindex. At launch, the block for search engines must be removed on the live website, both noindex and a blocking robots.txt. Canonical links must point to the live domain, and the XML sitemap must contain live URLs. In TYPO3, base variants in the site configuration let each environment use its own domain.
The consultant's role: You advise the stakeholders, explain the impact on visibility and anchor the SEO tasks in the plan and the budget. That includes mapping the old URLs, redirects, metadata, structure and the check after launch. And you advise against deleting redirects after a few weeks.
Key points
- SEO makes pages visible to search engines, GEO to AI answer systems. Classic SEO remains the foundation.
- 301 and 308 are permanent, 302 and 307 temporary. 404 means not found, 410 removed for good.
- Chains cost time, loops make pages unreachable, and redirects to the home page count as soft 404.
- An empty title for search engines falls back to the page title.
- noindex keeps a page out of the index and the sitemap, but does not protect it from visitors.
- Canonical links name the preferred URL and help against duplicate content.
- The XML sitemap comes from the site set “SEO Sitemap”.
- Page properties beat page.meta, unless TypoScript sets replace to 1.
- Open Graph and X Cards control the preview of shared links.
- Automatic redirects use 307 by default and point to the page ID. The module also preselects 307. For permanent moves you choose 301.
Terms
- Backlink
- A link from another website to your own.
- Canonical link
- A reference in the HTML head to the preferred URL of a piece of content.
- Duplicate content
- The same or very similar content under several URLs.
- GEO
- Generative Engine Optimization, optimizing for AI-powered answer systems.
- hreflang
- A reference to the language versions of a page.
- JSON-LD
- The usual format for structured data based on Schema.org.
- noindex
- A robots instruction not to add the page to the index.
- nofollow
- A robots instruction not to follow the links on the page.
- Open Graph
- Meta tags for the preview of shared links.
- Soft 404
- A redirect to an unsuitable page that search engines treat like a missing page.
- Status code
- The server's three-digit answer, for example 200, 301 or 404.
- XML sitemap
- A file listing the URLs of a website for search engines.