SEO KPI (Key Performance Indicators) are essential to systematically measure the success of SEO projects, to monitor your own optimization processes, and to derive new actions. SEO key figures must always be viewed over time in order to be meaningful and external factors such as seasonal fluctuations, holiday periods or the performance of competitors must be taken into account when measuring success.
In this article, you will learn which SEO measurements are important, where you can see them, what you can derive from the results and which SEO KPI can be actively influenced at all. I have divided the key figures into onpage SEO KPI and Offpage SEO KPI.
Onpage SEO KPI
With the help of these key figures, the performance of optimization measures on your own website can be measured. Active influence on these values can be exerted via classic onpage measures such as the optimization of title, meta description, heading structure, WDF*IDF content optimization and appealing CTA (call-to-action) elements.
Search engine rankings
When talking about search engine optimization, it is usually the goal to be displayed as high as possible in the search results for lucrative keywords. The monitoring of keyword rankings is one of many ways to measure the success of SEO measures.
Search engine rankings can be used to determine which subpages of a website perform particularly well in comparison, how the rankings of the entire website are distributed as a percentage on page 1, 2, 3, X, how the ranking develops over time and how the average ranking positions are determined.
The problem with this KPI is that there are tons of different search queries that allow a visitor to access the site via Google and the search results are now displayed individually (depending on location, search behavior, etc.). Therefore, when using these KPIs as measured values, it is important that the measurement is not limited to individual keywords, but that a large keyword group/keyword set is considered.
To get a quick overview of all rankings you can use the SEO Tools Sistrix, or my personal secret tip metrics.tools. For deeper analysis and average positions by keyword or subpage, the Google Search Console is recommended. The keyword monitor, for example, is suitable for manually tracking individual terms.
Even though Google currently has a worldwide market share of 87 percent (1st quarter 2018), Bing, Safari, Yandex or Baidu may also be relevant for the assessment of the current website status, depending on the target group and region. Bing and Yandex offer their own webmaster tools to get a deeper knowledge about the performance of the website. At least from time to time, it is worth taking a look at the ranking position and possible error messages for index coverage.
Click-Through Rate (CTR)
The CTR indicates the percentage of users who have actually reached a website via the search results. The CTR is calculated from the number of clicks in relation to the impressions. An example: A web page is displayed with Title + Meta Description 100 times in the search results and clicked 10 times. The click rate would, therefore, be 10 %.
The click rate can be used to measure which subpages are clicked most frequently and least frequently in the search results. Since Title and Meta Description are the first things a Google user sees before deciding to click on a search result, the click rate can be improved by optimizing them. A/B tests provide valuable insights into the extent to which certain formulations reach visitors and which should be better avoided.
There is a correlation between the average positioning of a website and the click rate. If the average positioning is low, the click rate is also lower. This should always be taken into account.
The click rate is also an important indicator of whether the meta and description accurately reflect the content of the page when associated with the bounce rate and dwell time. If visitors jump directly back to the site after an extremely short period of time, the description texts shown in the search result lists may not match the actual content of the page enough – or not at all. The website visitor interrupts the page visit in frustration if he does not immediately find the hoped-for information on the website. The situation is similar with the length of stay: Extensive texts or videos only capture the website visitor when it is clear at first glance that he will actually find the information he had hoped for.
The development of the click rate and average position of subpages over time can be viewed via the Google Search Console & Google Analytics and filtered by search queries, pages, countries, devices and search type. Please note that the data in the search console only covers a period of 90 days. Google Analytics has a longer database.
Traffic/page views from the organic search
Traffic is the pure number for the number of visitors on a website. The higher this value is, the better. The traffic should always be considered in connection with a certain dimension, otherwise, it has no significance. The questions that can be answered with this key figure are meaningful:
- On which days and which months is the flow of visitors at its highest?
- How has the visitor flow changed in period A compared to period B?
- Which subpages have the most visitors?
- How does SEO perform compared to other online marketing channels?
- Which country of origin and which mobile device do I use to generate the most visitors via the organic search?
- Which keyword is used to generate the most visitors?
These questions can be used, for example, to draw conclusions about which subpages should be prioritized in optimization measures, how the flow of visitors has changed after measures A or B, and to which end devices and countries of origin content should be optimized in any case.
This KPI can be split up more finely by additionally following the
- Landing Page
- Place of origin
sorted. The view by landing page provides information about concrete pages and measures taken there, which in turn are reflected in the ranking of the landing page and the defined conversion rate. My tip: Consider the Landing Page as a separate page within the performance analysis – not just as part of the website itself.
If the organic traffic comes from an unexpected specific region, the local search engine optimization can be adjusted accordingly. Particularly after local events such as trade fairs and congresses, it is worth keeping an eye on regional organic traffic in order to evaluate the effects of offline measures such as participation in trade fairs.
The traffic via the organic search results can be measured via Google Analytics, viewed from different dimensions and compared between different online marketing channels.
Conversion Rate (CR)
The conversion rate indicates the percentage of page visitors who perform a pre-defined action on the website. Typically, they are:
- a completed contact form
- a newsletter subscription
- a completed product purchase
- sharing content via Social Media channels
The prerequisite for monitoring the conversion rate is the prior creation of targets in the respective Web analysis tool. A conversion can and should always be assigned a (fixed) monetary value.
Please note: Only one conversion should be possible on a Landing Page. On a typical content page, on blogs or in e-commerce projects, several conversions – i.e. defined targets with a defined monetary value – can sometimes be used. An example: On a content project with affiliate links, the conversions are in a single blog post:
- Click on the affiliate link and forward it to the item you are ready to buy.
- Distribution of blog content via social media buttons
- Click on further articles at the end of the article
With the Landing Page with the goal of newsletter registration, no further elements should distract from the expected conversion!
The conversion rate usually gathers all conversions of a page. This SEO KPI is an important indicator of how well a page is tailored to the user’s search intention and how convincingly content, USP and CTA are communicated on target pages. A comparison can show which subpages and keywords have the highest value for the success of a company and which have less.
Finer distinctions can be made by determining the conversion rate based on
- Landing Page
and examined more closely.
The following applies to the conversion rate per landing page: Different values can apply depending on the destination and purpose of the landing page. If the same project produces very different values on different landing pages, it is worth taking a closer look at the differences in content design, CTA and the layout of the texts. It is best not to look at these values in isolation, but to look at the KPI for organic traffic, the clickthrough rate and the dwell time in parallel.
The conversion rate per location provides more accurate indications of how effective local search engine optimization is. If the newsletter landing page of a Denver pizza service converts perfectly, especially in the greater Boston area, something is definitely wrong. Local offers on a landing page should convert much better, especially in the targeted region.
With the conversion rate per device, it is possible to filter out exactly on which end devices display problems occur. The first indication of this would be conversions whose values vary greatly depending on the device used. Conversion via browser also makes such suboptimal representations clearer. Simulators such as Browserstack can be used to test a specific browser and device environments.
You can influence this value through:
- Changes in content and structure
- A more prominent CTA
- More convincing sales arguments
Optimization potential always exists on sites with a high number of visitors and an extremely low conversion rate.
The conversion rate can be monitored via Google Analytics, but must first be implemented manually in the source code. For online shops, this requires the special integration of e-commerce tracking. Afterward, you can check how high the conversion rate per subpage is and how much revenue has been generated by a page. For optimization approaches, the conversion rate is usually used as a measure to evaluate the success of individual measures.
The bounce rate indicates what percentage of page visitors leave a page directly without interaction. This SEO KPI must be interpreted differently, depending on the page focus. While a high bounce rate is bad in an online shop, it can be a positive indicator on content pages because the visitor has found what he is looking for. If the goal of the information page is to keep interested parties on the page about a larger content offer, a high bounce rate would still be critical. Then measures to increase the retention period would be important, for example by expanding the internal linking in the articles in the content itself, in the page margin areas or in the form of further articles at the end of the content.
The bounce rate is a helpful indicator of the quality of the user experience and gives us the opportunity to derive actions to optimize usability. Typical reasons for high bounce rates are:
- Wrong keyword focus
- Bad layout
- Bad content
- Confusing navigation
- Loading times too long
Google Analytics can be used to measure the bounce rate. However, the measurement values are inherently somewhat inaccurate and should be refined with the help of events. Here you can use event tracking with the Google Tag Manager to define events that count as interaction instead of jump. Possible events would be for example a certain dwell time on the page or scrolling to a certain point on the page.
Number of pages per session
In addition to the bounce rate and dwell time, the number of pages per session can also be used as an SEO measurement for the quality of the user experience. The more pages a visitor visits on a website, the more interested he seems to be in the content or the services. Of course, this value can also be an indicator that the visitor has to navigate unnecessarily on the page due to a bad page structure. In order to interpret these SEO KPI correctly, one should therefore also keep an eye on and monitor the behavioral flow of the users. In this – context, it must be checked which pages are specifically addressed in such meetings.
The number of pages per session is a suitable measurement to check the quality of the internal links and page structure and to derive recommendations for optimizing the page structure. The value is positively influenced by:
- Merging page areas and categories
- The optimization of the menu structure
- The removal of internal links that do not lead to the desired result
- Or setting new and helpful links
For a pure content offer, however, the number of pages per session must be evaluated differently than for a landing page. The former should be designed to stimulate meaningful page impressions in order to provide website visitors with the best possible information.
A Landing Page, on the other hand, will be designed exactly in such a way that no further page views within the session should be necessary. In other words, if there is a large number of additional page views per session after entering the LandingPage, the Landing Page should probably be revised. Then elements like menus, help areas in the footer or other page areas may distract the page viewer from the contents of the LandingPage.
The number of pages per session and the customer journey can be monitored using Google Analytics.
Average session duration
The average session duration indicates the average time a visitor stays on a (sub)page. Like the bounce rate, this measurement is primarily an indicator of user experience and how target-oriented the content is optimized for specific keywords. If the user is disappointed by the content and his expectations are not met, this time goes to zero.
With the help of the average session duration or length of stay, it is possible to compare how qualified the traffic of the organic search results is compared to other online marketing channels, which subpages should be qualitatively revised and how optimization measures have affected the user experience.
This value can be influenced by the quality of the content. One way to optimize this would be to enrich the content with pictures, videos, and other helpful information.
The problem with this reading is that it is an average. If visitor A stays 10 minutes on the page and visitor B stays 1 minute, the average session duration at the end is 5:30 minutes, which distorts the result. To make the results more meaningful one can alternatively measure how high the percentage of users is who have been on the site for a certain time. This is possible by setting up events and using target actions. To measure this SEO KPI Google Analytics and the use of the Google Tag Manager is predestined.
The average session duration can be further limited:
- By device
- Via browser
- Per LandingPage
- By origin/language
Browser and device specifications provide information about where in the user experience display problems occur or where the performance of the website deviates in relation to the browser and mobile devices.
An example: If the high-quality content, including a video recording on Safari browsers of a certain iPad generation, fails, the session duration of this user group will be correspondingly short. Sometimes the reason is quite simple: The video is not loaded or is not played correctly. The content loses, the user jumps to the next information offer. The resulting average session duration distorts the real averages on the one hand, but on the other hand, provides an explanation for bad values especially for the group of iPad users with Safari browser. Once the display error has been corrected, the session duration values should be adjusted independently of the device or browser used.
Anyone who can detect noticeable differences between the average session lengths in terms of origin attribution and language should check the quality of multilingual offerings more closely.
In addition to the conversions generated directly via the organic search, the number of preparatory conversions can also be used to measure SEO success. The preparatory conversion counts if a visitor reaches the page as the first step via the search results, then jumps off and in the third step navigates directly back to the page and makes a purchase.
The organic search was thus the first channel through which the buyer became aware of a product before buying it. Especially for content pages that are optimized for organic searches and have the intention to indirectly encourage the visitor to buy, this measurement value is very well suited for success measurement.
Preparatory conversions and conversion paths can be captured with the multi-channel funnel in Google Analytics. There is no universal Customer Journey. Depending on the industry, the country and the size of the company, purchasing behavior varies greatly. Google has nicely prepared this on the basis of the available data (and these are enormous amounts of data), what the usual customer journey looks like in each case: https://www.thinkwithgoogle.com/tools/customer-journey-to-online-purchase.html
Crawling and loading times
Crawling and loading times belong to the technical SEO KPI. The aim of technical SEO measures is to keep loading times as short as possible and to make it as easy as possible for the Google Bot to crawl the page so that changes and new pages are registered faster in the search engine index. Load times and the crawling behavior of the website should, therefore, be monitored regularly. Strong changes in these values indicate possible errors and should be investigated.
If Sitemaps of web projects are submitted to the Google Webmaster Dashboard, the number of pages already indexed can be compared with those submitted via Sitemap. Strong deviations usually also lead to explicit error messages, which are listed on the dashboard. Faulty pages and pages that have been excluded from indexing for performance reasons can also be found in the dashboard.
The loading time can be improved by compressing images, minifying CSS/JS files, swapping CSS and JS files, using a CDN and reducing HTTP requests (in this context I recommend using the http/2 protocol). To optimize the crawling, only the important pages should be indexed and included in the XML sitemap, 404 errors should be checked, the website should be checked for blocked content and the ratio of submitted and actually indexed pages should be checked.
An important tool for investigating crawling is the Google Search Console. Here you can monitor crawled pages, downloaded kilobytes as well as the duration of downloading one page per day. Detailed information on loading times can be found at GTmetrix and webpagetest.org as well as in the Google PageSpeed Tool.
For some time now, Google has been using its Pagespeed Tools to rate loading times for mobile devices and on the desktop. In mobile search, pages with optimized mobile presentation have been preferred since about 2015. With the rollout of the Mobile-First index in August 2018, long loading times and mobile display errors will lead to a downgrading of visibility in search engine rankings.
Offpage SEO KPI
With the help of these SEO indicators, the performance of optimization measures can be measured outside your own website. It is more difficult to directly influence the key figures in this category because they depend on external influences such as the behavior of the competition and the quality and quantity of external links. Offpage SEO KPI can be indirectly influenced, for example, by
- The creation of content with added value
- Satisfied customers
- High-quality products
- Outstanding USP (Unique Selling Proposition)
- The consideration of the onpage factors
Direct influence can partly be exerted with sustainable and targeted link building.
Visibility is an SEO indicator for the visibility of a complete domain in Google search results. It indicates how well a website is positioned for a set of X keywords in search engine results and is determined weekly. Higher rankings for keywords with a high monthly search volume increase the value more than good rankings for keywords with a low search volume. A thought-out example: A position 1 ranking to a keyword with a monthly search volume of 1000 increases the visibility value by +0.10 a position 1 to a keyword with search volume 100 by +0.010.
The visibility is suitable for obtaining a rough course an overview of the ranking development and as a measured value for competitive comparison. In this way, the ranking development of your own domain can be compared with competitors over time. In addition, you can use the development to determine whether there was a possible punishment on the part of Google.
The rankings in these tools almost never correspond to the rankings you get when you search the keyword yourself on Google. This is due to the fact that each person gets his individual search result. Depending on location, time of day, search history and the like. Nevertheless, these tools are very important, because you get an “objective” ranking for the keywords. As with many SEO KPIs, this is also about comparisons, not a single number. Above all, also to compare the own website with the competitors.
The disadvantage of this key figure is that the tools do not monitor all existing search queries and have not saved them in the respective databases. Especially in the longtail area, there are countless search queries which cannot be considered.
Furthermore, this figure is purely quantitative, not qualitative. The more rankings, the higher the visibility index. Domains that provide a result for every conceivable keyword (for example Amazon or eBay) have very high visibility. Web pages that deal with niche topics can have a visibility index of less than 1, even though they rank at the top of all important keywords.
The trust is an SEO KPI that indicates how much trust a website enjoys in a subject area X. The trust is an SEO KPI. The higher this value is, the better. It is calculated on the basis of the proximity and trustworthiness of the link-giving environment. It is, so to speak, an indicator of the quality and topic relevance of your own link profile. In addition, more than 200 factors are now included in the trust valuation.
This SEO measurement provides a rough overview of how your own link profile performs on average compared to competitors and in which subject area (e.g. with Majestic Topical Trust) your own online presence is regarded as a trustworthy source. Don’t forget: The main focus here is on high-quality links, both incoming and outgoing links are evaluated.
This value can be improved by reducing spam links that are displayed in the search console, for example, or by creating links to relevant and high-quality pages.
The extent to which social signs play a role here is discussed again and again. What’s certain is this: Sites that do not receive any social signals at all, that are not discussed in the social networks and whose contents are not shared, will feel changes in the trust in the long run.
Also Onpage KPIs like
- Loading times
- Dwell time on individual pages
- A legal framework such as data protection declarations, about page, or cookie information
- Missing HTTPS encryption
- Low domain age or frequent owner changes
can affect Google Trust. Just as the frequency with which Chrome users block a website in the browser plays a role here.
The trust can be retrieved using various tools that calculate this value using their own algorithms. The best known are Majestic, Moz, and Linkresearchtools.
The number of backlinks indicates how many links exist from external websites to your own domain. At this point, it should be mentioned that it is not only the quantity that is important, but also the quality, and you have to distinguish between the number of backlinks and the number of links of different domains (domain popularity).
With this SEO key figure first insights can be derived about how one stands in comparison to the competition and how much backlog demand and budget would have to be spent in order to catch up to the competition. It can also be used to monitor link growth over time. The number of backlinks should never be considered alone. It is important to investigate very carefully how trustworthy the linking sites are (trust), whether they are spam links and which contents are linked exactly.
The number of backlinks at a glance can be viewed, for example, with the help of Majestic and Linkresearchtool. With these backlink tools, you have the possibility to display information about link text, target URL and trust of the link giving websites or to export it as Excel sheet. Besides the Google Search Console offers the possibility to export backlinks. However, information on link text and trust is missing here.
Visitors who come to the website via external sources far away from Google are referred to as referral traffic. This is created via external links such as price comparison portals, topic-specific search engines far away from Google or online business directories.
Referral traffic can be used to identify which external link sources on the website generate the most traffic, which subpages are the most popular externally, and what effects NoFollow links without inherited link power have. In addition, new, previously unknown cooperation partners can be found in this way. And as we all know: Networking is everything.
The only way to assess which factors Google actually includes in the evaluation of the search results is to use correlations. However, I have the suspicion that Google will include data on referral traffic in order to take into account the quality of NoFollow links in the link mix and ranking.
Referral traffic can be evaluated and filtered via Google Analytics. Since the traffic sources also contain spam pages, appropriate filters should be set up. You can read more about this in the article Getting rid of fake traffic and referrer spam in Google Analytics.
There are tons of SEO indicators for measuring SEO success. Some people would think that search engine optimization can only be captured through rankings and that rankings are all that matters. However, the Google algorithm has long since changed so much that different search results are displayed for person A and person B depending on their location and search behavior. What counts in the end is qualified traffic that converts. In order to achieve this goal, the continuous measurement of all SEO indicators is equally necessary to intervene directly in case of wrong developments. And don’t forget: The relationship between the KPIs should not be underestimated. Looking at KPIs in isolation may lead to erroneous interpretations. It makes much more sense to take a comprehensive look. For example, we provide a monthly report that contains all changes in this field. So you always stay up to date!
I hope this collection makes it easier for you to get started with SEO Success Measurement and gives you enough input to prove your SEO commitment.
If you have any other SEO figures that you think are missing, please leave me a comment. I’ll expand the list then. If you have further questions about SEO measurement you can always contact our SEO agency.