We analyzed 11.8 million GOOGLE search results, giving me a deeper understanding of SEO
It has been stopped for more than a year, but the fans are still growing every day. I decided to bring a high-quality research article to all fans.
The following research content is mainly for the Google search engine. In fact, the essence of all search engine rankings is the same. I hope that the following research content can bring new ways of thinking to all students.
Recently, 11.8 million Google search results were analyzed to answer the following questions:
What factors are related to the ranking of the search engine homepage (top 10)?
Not only analyzed the content, but also analyzed the backlinks, and even analyzed the page loading speed...
With the help of our data partner Ahrefs, we found some interesting findings.
Today, I will share our findings with you.
The following is a summary of our main findings:
1. Our data shows that the overall link authority of the site is closely related to higher rankings.
2. Pages with a large number of backlinks rank higher than pages without backlinks. In fact, in Google's top ranking results, the average number of backlinks is 3.8 times more than the links ranked second to tenth. (Baidu has weakened the role of external links, so you don’t need to do external links everywhere for Baidu)
3. Comprehensive content with in-depth research and comprehensive introduction is significantly better than content without in-depth introduction of the topic.
4. We found that there is no correlation between page loading speed and Google ranking (weak correlation).
5. Obtaining backlinks from multiple different sites seems important for SEO. We found that the number of domains linking to the page correlated with ranking.
6. The vast majority of title tags in Google completely or partially match the keywords they rank for. However, we found that the correlation between using keywords in title tags and higher rankings on the Google homepage is basically zero.
7. The correlation between page authority and ranking is weak.
8. We found that the word count is more evenly distributed among the first 10 results. The page content on the first page of Google shows an average of 1447 words.
9. HTML page size has nothing to do with ranking. In other words, regardless of page size, they all have the same chance of ranking in Google search results.
10. We found that there is a small correlation between URL length and ranking. Specifically, short URLs tend to have a slight ranking advantage over long URLs.
11. Sites with above average "time on site" rank higher in Google. Specifically, increasing the time on site by 3 seconds will result in a higher ranking in search results.
Detailed data and information are provided below.
Authoritative domains rank higher in Google search results
In other words, the more times you browse on the homepage, the higher the domain rating.
In fact, compared with the authority of the page itself (URL score), the overall authority of a website is more relevant to its ranking.
Summary of key points: A higher domain name rating is related to a higher ranking on the Google homepage. Therefore, the domain has obvious advantages in SERP.
Pages with more backlinks tend to rank higher than pages with fewer backlinks
One of the most interesting findings of this analysis is that very few pages have any backlinks.
In fact, we found that about 95% of backlinks in all pages are zero.
This finding is consistent with Backlinko-BuzzSumo's analysis of 912 million blog posts, which found that 94% of all content had zero backlinks.
In fact, only a few pages have any backlinks, so that "zero backlinks" pages begin to distort the data. Therefore, for analyzing ranking factors, we decided to conduct a specific analysis, but exclude pages with zero backlinks.
We found that the page with the highest total number of backlinks ranks highest in Google.
Key takeaway: Even if Google continues to increase diversity in its algorithm, backlinks still seem to be a crucial ranking signal. (Baidu's role in external links has been weakened)
Comprehensive content is closely related to higher rankings
Many SEO experts claim that comprehensive content performs best in Google. (Note: Comprehensive content refers to the in-depth, multi-level, multi-angle, and multi-dimensional analysis of a topic, and the combined content)
In other words, the content covering the entire topic on a single page may be directly or indirectly related to rankings.
Our analysis found that there is a clear correlation between "content rating" and Google's ranking in PC and mobile results.
In fact, when looking at the top 30 results, increasing the "content level" by 1 will increase the ranking by 1 position. This shows that there is an important relationship.
Of course, it is not clear whether comprehensive content directly affects rankings.
Google may prefer more comprehensive content. Or, it may be that users are more satisfied with the search results, thus providing complete answers to their queries.
Since this is a correlation study, our data alone cannot determine the root cause behind this relationship.
Key points: Writing comprehensive, in-depth content can help a page rank higher in Google. The content of the whole people is not only in the main text, but refers to the content displayed on the entire page that must be related to the topic.
Page loading speed is not related to ranking
Since 2010, Google has been using website speed as an official ranking signal.
Google’s recent speed-related update, the "Speed Update" in 2018, aims to provide mobile searchers with faster loading pages. Baidu also has similar official announcement content, and the loading speed of mobile pages is one of the main factors in ranking.
However, we want to know:
Is website speed related to Google's actual ranking?
We use Alexa's domain speed to analyze the average load time of one million domains in the dataset. In other words, we did not directly measure the loading speed of each page in the data set. We just looked at the average load speed of the entire domain.
Overall, we found that the correlation between website speed and Google ranking is zero:
Discovering this may subvert our cognition. After all, PageSpeed is a confirmed Google ranking signal. When we know this, we hope that faster pages will generally rank better than slower pages.
However, the data paints another picture. However, when we researched deeply, we found that this lack of relationship makes sense.
When Google announced page load speed as a ranking factor, this largely affected extremely slow pages.
(Translation of the above image: After a period of time in ranking, this signal is mainly focused on desktop search. Today we announced that starting from July 2018, page speed will become a ranking factor for mobile search "speed update, because we only Will affect those pages that provide users with the slowest experience, using the same standards for all pages, regardless of the technology used to build the page. The purpose of the search query is still a very strong signal, so if a slow page has a large number of Related content, it may still rank high)
In short, Google's algorithm seems to reduce the ranking of extremely slow pages, and benefit fast pages.
Our analysis found that the average page load speed of the homepage results was 1.65 seconds.
Our previous website speed analysis found that the average page load time on the desktop was 10 seconds, and the load time on mobile devices was 27 seconds.
Compared with this benchmark, the average load speed of 1.65 seconds is very fast.
Moreover, because the top 10 search results load relatively fast, they don’t seem to be affected by Google’s various speed updates.
Highlights: The average loading time of the first page of Google results is 1.65 seconds. However, we found that there is no direct correlation between website speed and Google ranking. But fast loading speed is better than slow loading speed page ranking.
The number of recommended domains seems to have an impact on ranking
Many SEO experts agree that getting multiple backlinks from the same domain will reduce revenue.
In other words, it is better to get 10 links from 10 different sites, rather than getting 10 links from the same domain.
According to our analysis, this statement is also confirmed. We found that field diversity has a significant impact on rankings.
Just like using backlinks, the top of the first page tends to have more link domains than the link domains at the bottom of the first page.
The top results in Google point to 3 times more domains than the domains ranked 2-10
Key takeaway: For SEO, it seems important to get links from different domains.
Most title tags on the Google homepage contain keywords that exactly or partially match the search
Since the birth of search engines, title tags have always been considered the most important page SEO elements.
Because the title tag provides users (and search engines) with an overview of the overall theme of the page, the keywords that appear in the title tag can have a significant impact on rankings.
In fact, Google’s own "Search Engine Optimization Getting Started Guide" recommends writing a title tag to describe the entire content of the page.
(Translation of the above picture: Information, such as the actual location of the business, or may be its main focus or some best practices provided. It accurately describes the content of the page. Choose a title that can effectively convey the topic of the page content. Choose a topic Title, which has nothing to do with the content on the default or vague title (such as "Untitled" or "New") on page 1)
Sure enough, we found that most of the title tags on the Google homepage contain all or part of the keywords they rank for.
Although most pages that rank for keywords contain the keyword in the title tag, the title tag for keyword optimization does not seem to be associated with a higher ranking on the first page.
In fact, our linear model predicts that the relationship between title tag matching and ranking is very small (the difference between the 1st and 10th results is only 1%).
But if there is no such keyword tag in the title, it will be difficult to rank well.
However, once on the homepage, using the exact same keywords in the title does not seem to help you improve your ranking. This is where other factors (such as backlinks, user experience signals, and domain authority) seem to play an important role.
Key points: Among the top 10 search results pages in Google, 65% to 85% of keywords are ranked in the title tag. However, we found that there is not much correlation between the keyword-optimized title tags and the higher rankings on the homepage.
Keyword-optimized H1 tags have nothing to do with higher homepage rankings
Similar to our title tag discovery, most pages in Google search results have matching keywords in the H1 tag of the page.
Key points: Like title tag optimization, only keywords in H1 tags will have a chance to enter the top ten. However, the keyword richness in the H1 tag is not enough to help the page improve its homepage ranking. Therefore, there is no need to piling up keywords in the H1 tag, and piling up keywords is meaningless.
Web authorization (URL rating) has a slight correlation with higher ranking
In addition to the domain rating, does the authority of the page's backlinks affect rankings?
In other words, is it more important to get backlinks to specific pages? Or is the overall domain authority of the site more important?
To find out, we studied the correlation between page authority and ranking.
Although we do find that URL ratings and rankings are related, the relationship is very small.
Specifically, the URL score of the top 6 pages (12) is higher than that of the pages ranked 7-10 (11).
However, this relevance is not as strong as the influence of the website's domain name rating on rankings.
Overall, most URL ratings are similar in the top 10.
Important: Compared with the overall domain authority of the site, the link authority of each page on the site seems to have a smaller impact on rankings.
The average word count of Google homepage results is 1,447 words
Is long content better than short 200-word blog posts?
Research has found that compared with short blog posts, longer content tends to accumulate more backlinks and user dwell time.
In fact, we found that the top-ranked content in Google tends to exist for a long time.
Overall, the average word count of Google's top 10 results is 1,447 words.
However, although long format content is best for link building, we found that there is no direct relationship between word count and ranking.
This may be due to the fact that, like keyword-optimized title tags, long-form content can make it into the top 10. But once it arrives, if you want to continue to improve the ranking, it has nothing to do with the word count of the page content.
This is a correlation study, and we are unable to determine why long format content is more likely to appear on Google’s homepage.
Summary of key points: Compared with pages with fewer words, pages with higher word counts have a higher chance of ranking on the first page. The average word count of Google homepage results is 1,447 words.
Page HTML size has nothing to do with ranking
Will streamlined pages (in total bytes) affect your Google rankings?
According to this analysis, no.
We found no correlation between page size and ranking.
There are often SEOers who want to optimize the HTML code of the page, and feel that the bloated HTML page is not conducive to page ranking.
However, according to the pages in our analysis, page size cannot be linked to rankings.
Key point: Page size does not seem to have an effect on Google rankings.
Short URLs rank better than long URLs
Google recommends the use of "simple URLs", especially recommends not to use "extremely long" URLs.
(Translation of the above picture: only provide a vague information, such as not found. 404, or no 404 page at all
Use a design that is inconsistent with other parts of the website for the 404 page
Simple URL to pass content information
Creating descriptive categories and file names for the documents on the website not only helps to better preserve the website, it can be created more easily. "A more friendly URL for those who want to link to your content. Visitors may be intimidated by extremely long and mysterious URLs that contain few recognizable words
The URL shown in the image below may be confusing)
However, these suggestions seem to be more suitable for optimizing URLs for user experience than SEO.
This is why we set out to study the link between URL length and ranking.
In fact, we did find that short URLs rank higher than long URLs.
The average URL length of Google’s top 10 results is 66 characters.
However, in general, most URLs on the Google homepage are roughly the same length (40 to 100 characters).
Short URLs may improve SEO for a variety of reasons.
First, short URLs may lead to higher organic click-through rates. In fact, our large-scale organic click-through rate study found that short URLs have higher click-through rates than long URLs.
Second, a short URL can help Google understand the entire content of your web page.
Finally, a long URL usually requires a few clicks from the homepage to access the page. Usually, this means that there is less weight to flow to that page. The lower the weight, the lower the ranking.
Key point: Shorter URLs are associated with higher rankings. The average URL length on the Google homepage is 66 characters.
Sites with above average "time on site" rank higher in Google
Many people who have been exposed to SEO speculate that Google uses "user experience signals" (such as bounce rate, time on site, organic click-through rate, and pogosticking) as ranking factors.
To verify this theory, we ran a subset of domains from the data set through Alexa to determine the site time within the site. Then, we check whether there is a correlation between the time on site and the Google ranking on the homepage.
From the test results, it is indeed found that there is a close relationship between the time spent on the website and the ranking.
Specifically, we found that the time on site is closely related to higher rankings.
Generally, the average time spent on the Google homepage results on the site is 2.5 minutes.
Of course, Google may use information such as time on site or bounce rate as ranking signals (Google has denied it before). Perhaps high-quality content will make users more actively participate in the interaction. Therefore, the long stay on the site is a by-product of high-quality content measured by Google.
Since this is a correlation study, it cannot be determined from our data alone.
Highlights: The average time spent on the Google homepage results on the site is 2.5 minutes.
We also found a strong correlation between time on site and Google ranking. However, it is not clear whether this is a correlation or a causal relationship.
The above content is the content I translated and compiled based on my own experience in combination with foreign research SEO articles. Although this article is only a test for the Google search engine, it still applies to domestic search engines, and we can also learn a lot from it.
If you have any questions about the above content, please actively leave a message below, and I will reply to your questions in time.