SEO Checklist for an On Page Audit

In order to carry out an SEO check, you do not necessarily have to resort to expensive service providers. There are a lot of things that are pretty easy to test if you have the right tools. A comprehensive SEO check is important in order to be able to recognize whether a page already offers the maximum performance in the search engines or whether there is still room for improvement. Our checklist enables a comprehensive on-page audit.

The following SEO checklist is divided into different sections. First there is a quick check. This enables serious problems to be identified quickly – for example whether there is a Google penalty. The other parts of the checklist then gradually examine technical aspects, metadata, the page structure and the content. This is followed by tests for display on mobile devices, for international websites and for integration in Google News.

SEO quick check

With the SEO-Quick-Check you can quickly identify the most serious problems on your website that prevent optimal indexing in Google. Do the quick check before you worry about the details.

Do a Google search for your own domain

Search Google for pages from your domain. To do this, use the “site:” parameter. Example: “site: hassanbilal.net”.

Expected result: One or more pages are returned as a result.

What to do in the event of an error If no result was found, your domain is not indexed in Google. Possible causes: technical problems or a Google penalty. Then go to Google Webmaster Tools and see if you can find information about a penalty under “Manual Actions”. Otherwise, continue to review your site’s technical points.

Search Google for your brand or terms related to your brand

Expected result: Your page appears as the first search hit or at least in the top 10.

What to do in the event of an error If your branded search page is not at the front of the results, there is a possibility of a Google penalty or a technical problem. In this case, proceed as described above.

Check the pages in the Google cache

Expected result : When pages are accessed from the cache, the content of the page is displayed and contains all the links and the navigation.

What to do in the event of an error If the page contents do not appear or appear only incompletely, this indicates past problems with the indexing. In this case, you can try to delete the old content in the cache so that the new content is written. See also here .

Is the page signed in to Google Webmaster Tools?

The Google Webmaster Tools are the central instrument with which the integration of a website in the Google search can be controlled and monitored. A lot of important information about crawling errors, search queries or links can be called up here.

Expected result: The page is logged into Google Webmaster Tools.

What to do in the event of an error If the page is not yet registered in Google Webmaster Tools, this should be done as soon as possible.

Check the Google results for any existing sitelinks

For pages that are considered important enough, Google provides additional deep links to individual subpages in the search results snippets.

Expected Result: Sitelinks are displayed.

What to do in the event of an error If you don’t see any sitelinks, the website is probably not popular enough yet. This is especially the case with new pages. In addition, the page structure and internal links should be checked to rule out errors.Example of sitelinks in Google

Example of sitelinks in Google

Googlebot check

Retrieve the page like a googlebot. You can either do this with a User Agent Switcher (available as a plug-in for Firefox), or you can use the “Access as Google” function in Google Webmaster Tools.

Expected result: The page looks the same for the Googlebot as it does in the browser.

What to do in the event of an error If other content is displayed for the Googlebot than in the normal browser, this should be corrected urgently. Otherwise, there is a risk of Google penalties for so-called cloaking.

Bounce rate and length of stay

The bounce rate and the length of stay are now among the most important SEO metrics . The higher the bounce rate and the lower the length of stay, the more likely Google is to assume that users are not satisfied with a page. Depending on the type of page, the bounce rate shouldn’t be higher than 50 to 70 percent. A dwell time of well over a minute should be aimed for.

Expected result: bounce rate less than 70 – 50 percent, dwell time is well over a minute. The numbers can be determined using Google Analytics, for example.

What to do in the event of an error If the bounce rate is too high or the length of stay is too low, this may be due to technical or content-related reasons. A technical reason for a high bounce rate is, for example, if the page loads too long. Content that is of no interest to the user leads to the user leaving the page earlier. To increase the length of stay, additional, thematically appropriate content can be offered that appears above or below the articles (“similar articles”).

Technical exam

Measure the page speed

Various tools are available to determine the page speed , for example Google Page speed Insights or webpagetest.org .

Expected result: The loading speed is sufficient to good. The tools mentioned show corresponding ratings.

What to do in the event of an error If the page loads too slowly, there are various options for optimization. Check the following points:

• Has caching been activated for static content such as images, CSS or Javascript?
• Are images transmitted in compressed form?
• Is a content delivery network (CDN) being used?
• Is the performance of the server on which the website is located sufficient?

Check redirects on your part

If there are redirects that are designed to be permanent, these must be implemented as 301 redirects. You can use various tools such as Screaming Frog or Xenu Link Sleuth for testing .

Expected result : All permanent redirects are set up as 301 redirects.

What to do in the event of an error If there are redirects that deviate from the default, for example 302 redirects (temporary), check whether these should really only exist temporarily. If this is not the case, adjust the redirects. This is possible via .htaccess file if your page is played by an Apache web server.

Check your site for redirect chains

Expected result: When a page is called, a maximum of one redirect takes place. This can be checked with suitable tools such as the Firefox plug-in Firebug .

What to do in the event of an error If there are unwanted redirect chains, these must be resolved. This is also possible with the help of the .htaccess file.

Is the robots.txt set up correctly?

Check the robots.txt to make sure that important pages are not unintentionally blocked. Also check that all important resources such as Javascript or CSS files have been released for the crawlers. Information on the syntax of the robots.txt can be found here .

Expected result: no blocked critical pages or required resources.

What to do in the event of an error Remove the entries from robots.txt that are blocking important pages or resources.

Are the required sitemaps available and valid?

Sitemaps are important for the most complete indexing possible because they tell the search engines which URLs are on a website. XML sitemaps are used for this purpose. HTML sitemaps, on the other hand, are primarily aimed at users and are intended to facilitate orientation on the website.

Expected Result: There is both an XML sitemap and an HTML sitemap. There is also a news sitemap for pages listed in Google News. All sitemaps are valid. There are many suitable tools for testing. For example, Google offers a function for testing sitemaps within Webmaster Tools.

What to do in the event of an error If sitemaps are missing or not valid, they must be supplemented or adapted. An HTML sitemap can be dispensed with if necessary, but it provides a good orientation for the user.

Are there broken links?

Expected result: There are no broken links – neither internal nor external links. This can be checked with suitable tools such as Screaming Frog or Xenu Link Sleuth.

What to do in the event of an error: Broken links must be adjusted or removed.

What happens if Javascript is deactivated?

Expected result: Ideally, a page can be reached and functions without Javascript.

What to do in the event of an error If the page does not work without Javascript, this is not optimal. However, more than 90 percent of people now use Javascript, so the damage should be limited.

Does the page contain Flash?

Expected result: Since Flash cannot be displayed on some devices (for example on iOS) and Google cannot completely index this content, such content should be avoided.

What to do in the event of an error: If possible, Flash content should be replaced by other content.

Does the page contain iFrames?

Expected result: Since the crawlers do not read iFrames content, iFrames should not be played out.

What to do in the event of an error iFrames should be replaced and the corresponding content integrated directly into the website.

Page structure

A website should have a clear hierarchy of category and subpages. The internal linking should be thematically appropriate. This can be used to ensure optimal indexing by the search engine crawlers.

Is the page hierarchy flat enough?

Expected result: Depending on the greetings of a website, there should be a maximum of four hierarchy levels according to the scheme Home> Category> Sub-Category> Landing Page .

What to do in the event of an error: If the structure is too deep, there is a risk that the pages on the lower level will not be reached by the crawlers. In this case, the hierarchy should be made flatter.

Error pages

Meaningful error pages (e.g. for 404 errors) with helpful navigation and a search function help users to find their way around – for example, if a desired page is not available. Error pages must meet technical, design and usability requirements so that they can fulfill their purpose, because search engines are also dependent on such pages in order to be able to recognize, for example, that certain content has been deleted.

Expected result: The error pages must meet the following technical and content requirements:

Navigation elements to help users find the content they are looking for (search function, menu, suggestions for similar content, etc.)

The page must report back the relevant server status in terms of content, for example 404 in the case of a page not found. Soft 404 pages should be avoided. These are pages that return the server status 200 for “ok” despite the content not being found.

What to do in the event of an error If your own error pages do not meet the requirements professionally or technically, the pages must be adapted. More information on setting up 404 error pages can be found here .

Horizontal links to related topics

Expected result: Pages have an internal link to other pages with similar topics.

What to do in the event of an error: If there is no horizontal link, depending on the content management system used, suitable plugins can be used that display links to related pages (for example “similar articles” based on headings or keywords).

Not too many links on the page

The link power that a page can pass on is distributed over the total number of outbound links on this page. The more links there are, the weaker each individual link is. In addition, Google takes a critical look at pages with too many links.

Expected result: Depending on the size and scope of the page, the number of outbound links is no more than 20 to 50.

What to do in the event of an error: Less important links must be removed. Setting “nofollow” is not enough, since such links are fully included in the distribution.

Internal Page Rank

The most important sub-pages should also be the best-linked pages on a website. A landing page that is important for sales should, for example, be linked more strongly than an imprint or the page with the terms and conditions. It is not only a question of the number of links that refer to a page, but also the strength of the links. As a webmaster, you can influence internal links in particular, both through the navigation structure and through additional links.

How strongly a subpage is actually linked cannot be easily seen. For this it is necessary to calculate the internal PageRank. This provides a value for each subpage. The larger the internal PageRank, the more prominent the page is with regard to its linking.

The internal PageRank can be determined with the help of common tools. This requires a crawler such as Screaming Frog and software suitable for statistical calculations. The website to be examined is crawled using a screaming frog. Then you export the outgoing links of all pages. The result is then analyzed with a statistics tool such as “R”.

A detailed description for the determination of internal PageRank is here to find.

Content

Is there duplicate content on the site?

Duplicate content can take various forms.

Duplicate title or description: it is best to use a crawler such as Screaming Frog or Xenu Link Sleuth for checking.

Pages that can be called up under different URLs: Pay particular attention to URLs with parameters here – printer pages are also dangerous.

Reuse of texts on several pages: the best way to check it is to do a Google search for an entire text or section.

Domain can be reached with and without “www”: call up the domain with and without “www” for testing. One of the variants should pass on to the other.

Domain can be reached under HTTP and under HTTPS: Comparable to the case of the domain with or without “www”.

Expected result: Duplicate content in the form described does not occur.

What to do in the event of an error There are numerous ways to avoid duplicate content:

• If there are several URLs per page: Set a canonical attribute that refers to the URL that is to be represented in the index
• In the case of “www” versus “without www”: Set up forwarding of one version to the other. This is possible via .htaccess.
• Ditto for the HTTP / HTTPS case.
• Duplicate titles and descriptions have to be adjusted.
• Pages with duplicate content that should remain (for example printer pages) should be set to “noindex”.

Headings neatly divided.

Expected result: Each page has exactly one H1 heading and possibly one or more H2 signatures. If content is to be broken down further, H3 headings can also be used.

What to do in the event of an error Missing H1 headings have to be added, superfluous H1 headings have to be removed.

Target keyword included in H1?

Expected Result: The target keyword that you want the page to rank for is in the H1 heading.

What to do in the event of an error If the target keyword is not included in the H1 heading, it must be supplemented.

Integrated headings as text?

Expected result: Headings are included as text and not as graphics so that the search engines can interpret the content.

What to do in the event of an error Graphic headings are to be replaced by text.

Do headings encourage you to click?

Expected Result: Headings must be formulated in such a way that they create an incentive to click. They have to arouse curiosity without making promises that the contents of the page cannot deliver.

What to do in the event of an error Headings without an incentive to click should be reformulated accordingly.

Optimized images?

Images must be integrated in such a way that their relevance for the search engines can be recognized. The files must also not be too large so that the page can load quickly.

Expected result:

• Images are transmitted in compressed form .
• Images have an alt tag .
• The file names of the images are meaningful and relate to the content.
• Images are optimized to limit file size.

What to do in the event of an error If the images do not meet the requirements mentioned, at least the images for the most important pages should be optimized.

Structured data set correctly?

Structured data enable the content of elements of websites such as images or headings to be assigned. This helps the search engines to assess relevance. In the worst case, incorrectly set structured data can lead to a Google penalty . The implementation of structured data can be checked, for example, using Google Webmaster Tools.

Expected result: Structured data was implemented without errors and is not misused. There are no unnecessary ratings or ratings from category or list pages.

What to do in the event of an error Should Google display incorrect structured data, the errors must be corrected or the markup removed.

Are the URLs optimized?

Expected result: The page URLs each contain the target keyword, work without parameters such as session ID and are static, i.e. they do not change. URLs that are too long should be avoided. In addition, it is important to ensure that no spaces are used in URLs .

What to do in the event of an error If URLs have to be adapted, make sure that a redirect to the new URL is set up (301, i.e. permanent).

Structure of the texts

Expected result: Texts, especially longer texts, are divided into several sections and are visually clear. The important sections have appropriate headings.

What to do in the event of an error Wastes of text make legibility more difficult and thus lead to a higher bounce rate. Therefore, it is essential to pay attention to well-structured texts!

Texts long enough?

Expected Result: An article should be at least 100 to 120 words long – especially after the Panda update , Google attaches great importance to the content. If a page is listed in Google News, the minimum text length is even more important.

What to do in the event of an error Texts that are too short must be expanded. If in doubt, consider taking pages with texts that are too short offline.

Ad placement

Expected result: no or only a few ads are displayed at the top of the page (“above the fold”).

What to do in the event of an error Remove ads or move them to the bottom of the page.

Ratio of content to ads

Expected Result: The main body of the page consists of content that offers users added value. Too many ads are harmful (keyword Google Panda). No more than two ads should appear per page.

What to do in the event of an error: Expand content or remove excess ads.

Spell and grammar check

Expected Result: The page is free of spelling and grammar errors.

What to do in the event of an error Spelling mistakes and incorrect grammar such as incorrectly placed or forgotten commas give a bad impression and appear unprofessional. It is therefore imperative to rectify these errors or seek advice from a specialist.

Metadata

Metadata is one of the most important SEO tools. You alone can decide about the success or failure of a side.

Title

Expected result: There is an individual title per page that contains the target keyword. Since the widening of the Google results column, tiltles can now be displayed with a maximum of 70 instead of the previous maximum of 55 to 60 characters. The display can be tested with this tool .

What to do in the event of an error Titles that are too long, too short, or poorly worded need to be corrected. Missing titles are to be added.

Description

Expected result: There is an individual description per page , which is a maximum of 175 characters long (previously: 155 characters), so that it can be displayed in full in the search results. The target keyword should be included in the description. The description should arouse curiosity and encourage people to click.

What to do in the event of an error Descriptions that do not meet the requirements mentioned must be adapted. Missing descriptions must be added.

Meta keywords

Expected result: Meta keywords are no longer relevant for ranking in Google, but can still be useful. For example, some content management system plugins use meta keywords to identify related articles and display them. Beware of too many meta keywords.

What to do in the event of an error If no meta keywords are available, selected keywords can be set. Too many meta keywords should be avoided and reduced accordingly. Meta keywords can, however, be dispensed with.

Canonicals

Expected result: The canonical attribute (link rel = “canonical”) determines which URL should be used for the Google index when there are multiple URLs per page. It must be ensured that the set canonical refers to the correct URL.

What to do in the event of an error A canonical does not have to be used. However, if it is used, make sure to use the correct URL, otherwise the website can be seriously damaged. In the worst case, most of the pages will be dropped out of the index.

Meta robots

The robots attribute indicates whether a page should be indexed and its links followed. A page that is to be indexed and whose links are to be followed has the attribute meta name = “robots” content = “index, follow”.

Expected result: All pages to be included in the Google index have the attribute meta name = “robots” content = “index, follow” , all other pages either meta name = “robots” content = “noindex, follow” or meta name = “robots” content = “noindex, nofollow” .

What to do in the event of an error Incorrectly set meta robots must be adapted, because otherwise no desired indexing is possible.

Check with Google Webmaster Tools

Google Webmaster Tools can quickly identify some of the most important problems that can arise with indexing. This includes, for example, incorrect sitemaps, penalties or duplicate content.

Crawl errors

Problems that occurred while crawling appear under this point:

• Inaccessible pages
• Server error messages
• Soft 404, these are pages with not found content, for which the server does not provide a suitable error code.

Crawl statistics

Here you can see how many pages the Googlebot visited per day and how large the volume of data was accessed. A sudden drop in the crawl rate indicates problems. It also shows how long it took to download a page.

HTML improvements

Here you can find references to duplicate or missing titles or descriptions. Even if these are too short or too long, it will be displayed here.

Clicks, impressions and click rate

How often your own page appeared in the search results and how often it was clicked on can be seen here. The click rate also shows the relationship between clicks and impressions.

 Blocked resources

If important files such as CSS or Javascript cannot be crawled, Google may not be able to index a page correctly. You can find out which elements are blocked here.

Left

Incoming links are still of great importance for the ranking. Google provides an up-to-date list of backlinks in the Webmaster Tools. Here you can regularly see which links have been added.

Video SEO

The integration of videos is often neglected on websites. This is a big mistake, because videos are a good prerequisite for more visibility on the search results pages. However, in order for this potential to be exploited, a few things must be observed:

• The better the quality of a video , the greater the likelihood that users will watch the video through to the end. This in turn flows into the evaluation of the videos as a ranking factor. It is therefore best to always create videos in HD and have good sound quality .
• For the video, a transcription should be offered on the page on which the video is integrated , i.e. the content contained in the video as text. This improves the relevance rating of the site and helps users with hearing impairments as well as non-native speakers.
• In addition, a detailed description of the video should be included in the website, ideally in the form of a brief summary of the content with 200 to 300 characters .
• The video should not be integrated into the website as a YouTube version and via iFrame, but rather hosted itself . This means that the video is loaded and played from your own server. This increases the chances of a universal search hit in Google (with the associated video thumbnail in the snippet).
• A video sitemap should be offered for the video .
• Marking videos via schema.org: Meta information, so-called structured data, is integrated here so that Google and other search engines are able to recognize the content and evaluate it accordingly.

In addition, the video should definitely also be uploaded to YouTube. The following things are particularly important for embedding videos in YouTube:

• Provide and upload subtitles for the video: In order to reach non-native speakers and people with hearing impairments, subtitles are a good addition to a video. Subtitles can easily be added to YouTube .
• Use call-to-action on YouTube: You can include a call-to-action overlay (CTA) in the video. CTAs can be created in the YouTube account . An AdWords account is required, the Google AdWords guidelines apply . Information on integrating CTAs is available here.
• Use description for videos on YouTube: The description on YouTube has space for 5000 characters. This space should be used as much as possible because the relevance rating of videos includes the content of the description. In addition, care should be taken to integrate the most important statements and content in the first 100 to 120 characters of the description, because that is the part of the description that is displayed directly below the video without further clicking.
• Use meaningful thumbnails: Thumbnails are the small preview images that are displayed for videos. The better and more meaningful a thumbnail, the greater the likelihood that the video will be viewed. YouTube allows custom thumbnails to be uploaded . The thumbnails should have a resolution of 1280×720 pixels or a minimum width of 640 pixels, be in one of the formats JPG, GIF, BMP or PNG and have a maximum file size of 2 MB. The recommended aspect ratio is 16: 9.

Google News

For pages that are listed in Google’s news search Google News, there are additional requirements that must be met. Google pays particular attention to the quality of the content, which must meet journalistic standards. The frequency of publications also plays a role.

General guidelines

• Does the thematic focus of the website fit? Is the content newsworthy? Is the content interesting?
• Does the quality of the content meet journalistic standards? Are sources properly cited and referenced?
• Are the texts correct in terms of orthography (spelling) and grammar?
• Are the texts structured and written in such a way that they can be read fluently?
• Do the authors of the texts have the necessary expertise and knowledge for the topics covered?
• Are the authors named in the text?

Technical guidelines

• Can the articles be recorded by the crawler? Has all important content such as JavaScript and CSS files been approved?
• Are the articles displayed in HTML format? For example, PDF files are not suitable for indexing in Google News.
• Can the articles be reached under static and permanently valid addresses (URLs)? Dynamic URLs must not be used.
• Optional: Is there a news sitemap? There are special requirements for news sitemaps that differ from those for normal sitemaps. For example, only articles from the last two days can be listed. In addition, there can be a maximum of 1,000 URLs per sitemap.
• No longer valid: Until recently, Google required at least a three-digit number in the URL for each news item. This requirement does not apply.

Quality guidelines

• The loading times of the website must not be too high. Although there is no generally valid rule, loading times of more than five seconds are considered to be too long.
• Mobile-friendliness: The website must also be easy to read and use on small displays such as smartphones.
• No mixing of content: Articles intended for news searches may not be mixed with other content on the website. For example, no commercial content may find its way into the news search. This can best be achieved through a clear structural separation of the categories on the website.
• Publish new posts regularly: In the application phase for inclusion in Google News, at least one or two new posts should appear per day.
• Number of editors: Websites should contain contributions from at least three to four active editors. On the other hand, blogs written by just one person have a hard time getting included.
• The authors must be named on a central page – ideally in the imprint; ideally with a photo and email address
• Minimum Length: Articles must contain at least 80 to 100 words to appear on Google News.
• Articles cannot be too fragmented. Too short paragraphs must be avoided.

Inclusion in Google News can be done in the Google News Publisher Center .

Mobile SEO

Meanwhile, the number of mobile searches in some countries has exceeded the number of searches carried out on desktops. That is why it is important to ensure that the website is optimally displayed on smartphones. Google has now established this as a ranking factor. Mobile-friendly sites therefore benefit.

Run mobile search

The first thing to do is do a mobile search and find your site. See if Google is showing the “For Mobile” label.Google notice: “For mobile devices”Then everything is fine. In the worst case, the search result contains an indication that the page is loading slowly – Google is currently testing appropriate markings.

Faulty redirects

If your site has a separate version for mobile devices (for example at m.domain.com), make sure that the redirects for each subpage are set up correctly and that not all of them point to the start page . Because Google doesn’t like that at all.

Rel = “alternate” and rel = “canonical”

This is also relevant for those pages that have a separate version for mobile devices: Check whether the desktop version has the attribute rel = “canonical” and the mobile version the attribute rel = “alternate” . Details on the correct implementation can be found here .

Mobile friendliness and usability

The most important test for a mobile website is the one with Google’s specially designed tool . Here you will quickly get the result that tells you whether or not Google considers your page to be suitable for display on smartphones. Above all, the usability and the speed are evaluated. Problems can arise, for example, from horizontal scrolling, links that are too small or buttons that are too small.

Is the Vary HTTP header set?

For pages for which different HTML and / or CSS is displayed depending on the device used (desktop, smartphone, tablet …), a corresponding note should be set in the HTTP header . This informs the crawler for the desktop search that there is special content for the mobile version.

International sites

There are special challenges for pages that are played in several languages ​​and countries. The most important ones are as follows:

Use of rel = “alternate” and hreflang

In the case of multilingualism, the use of the rel = “alternate” and the hreflang attribute can be used to indicate which language and country versions are available as an alternative. An example : In addition to Spanish, a page is also available in French and Portuguese. This would be declared by the following statements, which have to be integrated into every language version:Correct use of hreflang and rel = “alternate”: Example

Check country alignment in Google Webmaster Tools

Under the item “International Orientation” you can check within the Google Webmaster Tools which country your site is aimed at. If the alignment is not correct, this can be adjusted under the “Country” tab. More information here .

Local language URLs?

In the case of multilingual websites, the text of the URLs should be formulated in the appropriate language.

Do you need help optimizing your website?

With the help of the points mentioned, a comprehensive on-site audit can be carried out. If you have any problems or questions, or if a more extensive check is required, please contact me using the contact form . You can find more information on SEO advice here .