Starting from this article, we will tell you how to do a full SEO-audit of an online store, although for information or service sites many parameters overlap. Technical parameters are the same everywhere, but usability and semantics can differ in content, because sites solve different tasks (buy a product, order a service and get an answer to a question), but the approach for the most part will always be the same everywhere.
We will not talk about external optimization in the form of communities in social networks, the reference mass and other methods of this type of promotion, but only about internal factors.
The full material will consist of three pages:
1. Technical SEO-audit
2. Usability Audit
3. Semantic audit
And in this article it will be a question of technical optimization and setting up an online store site. As an example, consider an online store for the sale of sanitary ware. The content of the article can be considered as a checklist of the main technical parameters that need to be adjusted / corrected for normal indexing by search engines and site users.
The content of the article:
1. General information about the domain: age, TIC, pages in Yandex, pages in Google
2. Server Response Time
3. Speed of loading pages
4. Presence of a mirror of a site (with "www" or without "www")
5. Redirect with http version of the site on https (if available)
6. Duplicates of the main page (index.php, index.html, index.htm, etc.)
7. Testing the robots.txt file
8. Site map check sitemap.xml
9. Checking the CNC pages
10. Checking the error page 404
11. Analysis of the pages on 3XX, 4XX, 5XX responses
12. The presence of broken links, images and other garbage in the HTML-code
13. The presence of external links, their number and response code
14. Blocked pages in robots.txt
15. The presence of pages with meta noindex
16. Indexing of official pages
17. Indexing of duplicated navigation elements
18. Indexing of Service Text Elements
19. Presence in HTML-code Display: none
20. Check for paginator and rel = canonical
21. Check meta title: availability, length, duplication
22. Check meta description: availability, length, duplication
23. Header check h1: number, length, duplication
24. Image analysis for the presence of the Alt attribute
Technical SEO-audit of the online sanitary ware shop svdd.ru
General information about the domain
Age: 9 years 2 months
Pages on Google: 4,890
Pages in Yandex: 13,923
The age of the domain can talk about the authority of the resource. Typically, the older the domain, the better it is ranked in the search. If the domain is young, up to 1 year, then there should not be a sharp increase in search traffic.
TIC - thematic index of citation. This indicator also tells us about the authority of the resource, or at least the presence of an external reference mass, which also needs to be analyzed.
The age of the domain and the TIC are primarily needed for comparison with competitors, to identify the youngest and weakest, but who is along with the flagships of the niche and see what he does with the site in order to rank so high.
The big difference in the number of pages in the index of Google and Yandex may indicate the incorrect configuration of robots.txt, filtering in one of the search engines, or about some error that needs to be found and corrected.
Server Response Time
Home - 700-800 ms.
Category - 500-600 ms (re-check)
The product page is 500-600 ms. (recheck)
The response time should be 200-300 ms, i.e. Now the answer is too long and you need to strive for a normal rate.
Recommendations: Reduce server response time to 200-300 ms.
Page loading speed
Home - 10,408 sec.
Category - 8,096 sec.
Product page - 8.5 sec.
The optimal loading time is 2-3 seconds, but for online stores with a lot of goods it is difficult to reach this value, so let's compare the time of loading pages with the sites of competitors, which we take from the top-10 Yandex issuance for one of the target requests.
Because The lion's share of traffic falls into categories, then it will be enough to compare the download times of similar pages for output.
Competitor 1 - 4,593 seconds.
Competitor 2 - 5,964 sec.
Competitor 3 - 9.485 seconds.
Competitor 4 - 5,401 sec.
Competitor 5 - 2,314 seconds.
Competitor 6 - 7,446 seconds.
Competitor 7 - 6,999 sec.
We see that in most cases the download time is less than that of the site under investigation, therefore, by all possible means it is necessary to strive to reduce it and achieve at least an average value of 6 seconds.
Recommendations: Reduce the load time for category pages up to 6 seconds.
Presence of a mirror of a site (with "www" or without "www")
The gluing of mirrors was carried out correctly (through 301 redirects).
To avoid duplicating the pages of the site in the search engine index, you need to glue the domain with the WWW and without. Also, gluing is necessary for users who enter the domain name of the site in the address bar of the browser, so that the domain name with the zone is sufficient to hit the site of interest. The glue must be carried out using 301 redirects.
Redirect with http version of the site on https (if available)
The redirect is configured correctly.
The https protocol for the user is a guarantee of the security of its data (including payment details), and for search engines - a signal for ranking. In 2016 Google representatives officially announced that since January 2017 a secure connection will become one of the official algorithms. That is, sites on http will gradually lose their positions.
Main page dubs (index.php, index.html, index.htm, etc.)
Checking your robots.txt file
The file is available.
The session begins by downloading the robots.txt file. If the file is missing, it is not a text file or the HTTP status is returned to the robot request, other than 200 OK, the robot believes that access to the documents is unlimited. This can lead to the entry in the index of unwanted files and pages of the site.
Site map check sitemap.xml
The file is available.
The presence of a site map allows the search engine robot to quickly learn and index as many pages of the site as possible, in the absence of a site map, the search for pages of deep nesting can be delayed.
Checking the CNC Pages
The CNC is in order.
All URLs must be logical and easily perceived by visitors. CNC better ranked in the search results and look more pleasant on sites and social networks (if you do not hide the address of the site in the anchor). If possible, it is better not to use underscores in addresses.
Error page check 404
The page is working and displayed correctly.
404 page is necessary to ensure that after disabling or deleting a page from the site, at a previously working address, a message was displayed to users and there was a corresponding 404 server response code. 404 server response will eliminate the need to monitor garbage pages in the index, because the search engine will exclude them.
Analysis of pages on 3xx, 4xx, 5xx responses
0 Pages have a 3xx response
0 Pages have a 4xx response
0 Pages have a 5xx response
The presence of broken links, images and other garbage in the HTML-code
1 Picture with 4xx response
The problem with the encoding of the email address in the footer of the site:
The presence of external links, their number and response code
962 External links with rel = "nofollow"
27 External links with rel = "dofollow"
0 External links with 4xx response
0 External links with 5xx response
Locked pages in robots.txt
0 Pages blocked by robots.txt
The presence of pages with meta noindex
0 Pages are not indexed by meta noindex
Indexing of official pages
List of service pages that you need to close from indexing via robots.txt
Indexing of duplicated navigation elements
Close the indexing of duplicating elements via <! - noindex ->:
Paginator in the category under the goods.
Indexing of Service Text Elements
Close frequently encountered service words (highlighted in red rectangle) via <! - noindex ->:
On the product page in the "Related Products" and "Products in this category" blocks:
HTML: Display: none
All the attributes are placed correctly on the allowed elements.
Check for paginator and rel = canonical
There is a paginator. The rel = canonical attribute is registered correctly.
Below are the checks, each of which needs to provide examples of pages, but here we do not publish them for obvious reasons, but there is such a list of pages in the audit report.
Check meta title: availability, length, duplication
0 Pages with an empty or missing title
222 Pages with a long title> 70
23 pages with short title <10
28 Pages with duplicate title
Check meta description: availability, length, duplication
98 Pages with blank or missing description
50 Pages with a long description> 320
497 Pages with short description <50
19 Pages with a duplicate description
Header check h1: number, length, duplication
0 Pages with an empty tag H1
0 Pages without H1 tag
27 Pages with a long tag H1
50 Pages with duplicate tag H1
Image analysis for the presence of the Alt attribute
5032 Images with missing ALT text
Fill in the attribute for payment methods in the footer of the site:
This technical SEO-audit is completed, then you can proceed to usability, and then to a semantic audit.