March 25, 2019 at 10:57Balanced site performance. Part 5: Usability
This is the final part of the story about the concept of balanced site metrics. In previous issues, we phased out all the main aspects of promotion and determined the relationship between these aspects. In this, we will consider the last block of indicators and draw final conclusions.
Previous parts:
- First part: Strategy
- Second part: Technical optimization
- Third part: Content
- Fourth part: External optimization
Now I suggest recalling our main scheme:
Let me remind you that the main element that immediately affects everyone else is Strategy. And the elements themselves are interdependent on each other. Therefore, in fact, in the name of the concept there is the word “balanced”: any serious changes in each of the aspects of promotion automatically take the whole system out of balance and require immediate restoration of balance.
If with the technical side of the site everything is basically obvious, but with the content and the linking environment a little less, but still understandable, then there are often problems with usability. They are connected with the fact that the very concept of "usability" is not perceived by many as very true, and therefore it is applied incorrectly to SEO and becomes a missing link.
Let's see what it is and why it interests us so much.
In a word, translating this term into Russian is impossible. Translated literally, it is “usability”. A less accurate but more familiar option is “usability”.
Usability is a conditional quantity that determines how capable the average user of a tool is to achieve the desired result using this tool in typical conditions. And also how easy he will be able to do it, how much he will be satisfied with the process of work.
In the most deplorable cases, the user is not able to use the tool at all, in the most successful ones, he is able to master it directly during work without any basic knowledge, due to his own life experience.
The site on our scheme is no coincidence, although it is in the central circle, but it is not a headline: it is a tool. And the goal, as we said in the first article, is to perform targeted actions by users (receive conversions). So in our case, usability will determine how fast, with what emotions and how much the targeted users will be able to achieve the goal.
Today, search algorithms take user expertise very seriously, the very behavioral factors. This is quite logical from the point of view of the logic of search engines: the orientation toward satisfying user interests is openly declared by them in their own license agreements. Obviously, how much the site is made “for people” is better to ask the people themselves.
Therefore, usability is very important for search promotion: it is an opportunity to give search engines a free and unbiased reason to consider a site better. Of course, for this it will be necessary to work and closely monitor the fate of your project, but in the framework of a single strategy, this is not as difficult as it might seem.
Understanding that the site has usability problems is not so difficult. This information is given to us by metrics. Moreover, they are constantly developing and improving their own usability indicator, so the most popular and necessary reports are already configured out of the box: traffic sources, entry and exit points, page time, viewing depth - all this can be viewed and analyzed the very next day after installation code. You can even see in the first person how the user scrolls the page and moves the mouse cursor.
And from this moment you can already draw conclusions about how “suitable for use” the site is. Of course, it will be necessary to set up event chains, goals, and if you are engaged in online sales, then e-commerce. But this is not so difficult, and the principles of analysis there are about the same.
Metrics give us the opportunity to look at the site through the eyes of the user. And then all its weaknesses, design errors, incorrectly guessed intent, and much more that can be overlooked at the development stage are revealed. Many will be scared by what they saw, but believe me, it’s much worse to not understand what you are doing wrong.
Particularly dramatic data is provided by end-to-end analytics systems, which I also mentioned in the second part of our series. They allow you to calculate business indicators such as ROI (return on investment) for your investments in the development of the site. So you can see in real numbers how profitable, say, was the decision to change the design of the pages, how quickly it paid off and whether it paid off at all. Based on these data, you will be able to more accurately plan budgets, understand exactly which solutions allow you to spend one time and reduce regular costs and payments.
End-to-end analytics allows you to quickly identify, for example, that you are ordering work for the most unprofitable of the available schemes (for example, for a monthly fee, rather than buying hours of work, or vice versa). For especially large projects, this can save huge amounts.
Metrics reveal errors in various aspects of promotion. We can see incorrect clustering of keywords, due to which the user gets to the wrong page that better meets his specific task. From here we get the need for re-clustering, changing the content of the pages (and sometimes creating new sections and changing the structure), changing incoming links - in a word, the whole system instantly responds to this error.
We may find that the forms do not work correctly on the site and the user cannot perform some simple action. We set off to study the code, check the server settings, and immerse ourselves in technical optimization.
It also happens that the page is loaded with errors, and the user leaves it immediately after opening, because he sees the impossibility of interacting with it, or simply perceives it as an indication of the quality of the site and decides not to have anything to do with it.
Remember, when we talked about content, I mentioned that it should answer the questions that a user has as fully as possible while studying? Now we can find out how well he does it. And if not enough, then make the appropriate changes. Can't do this within the existing page structure? So, we are trying to change this structure. Or maybe it will be easier and more promising to change the structure of the site itself?
Content with usability is generally very closely linked, in many aspects it is impossible to distinguish between them: content is an important part of the page and one of the main goals of the user. And it's not just about the text: well-placed infographics or other visual content captivates the visitor, creates a sense of harmony, helps to get what you want. Illiterate visual content interferes with the perception of information, draws attention to the poor quality of performance, negatively affects the perception of the credibility of the resource and encourages him to leave.
Of course, usability is not limited to content. This is a combination of all the factors that translate into the final picture: the logical structure of the site and pages, the correctness and relevance of the forms, ease of navigation, the presence of full-fledged interaction scenarios designed for different user groups, and much more.
It is no coincidence that all large and popular projects constantly refine the interface: the Internet is constantly evolving, so the user's perception of certain decisions is changing. Familiar tools become obsolete, new ones become more familiar.
Trends in the design and layout of pages live according to the general laws of fashion: if everyone makes a “Buy” button in green and with rounded edges in the lower right corner, then the user will look for it there, and that’s it.
Therefore, an ideal option would be to conduct A / B testing, when different versions of pages are shown to different users. Based on the results of this testing, the most successful solutions remain, and those that have not proved to be the best are changed or deleted. Unfortunately, this is a rather costly method and requires a representative sample. Therefore, usually large sites with high traffic resort to such testing.
There are no trifles: if you have a large failure rate, then this is bad not only in terms of conversions, but also in terms of promotion. Users come from the search and immediately leave - what should this tell the search engine? That's right: that the site is in the Top 10 undeservedly.
But in fact, the evaluation of your site by the user begins even before he clicked on the link. Let's talk about snippet, a small announcement that the search engine shows on the results page.
The user’s behavior in search is analyzed by search engines in no less detail than his actions on the site. These data allow the search engine to once again check the quality of the results, exploring the experience of living people.
The fact is that algorithms are far from ideal, they make mistakes. And it happens that the results fall on sites that do not belong there. Therefore, data on clicks on links is also an indicator of the relevance of the page to the request.
If you are in the Top 3 and nobody clicks on you (and yes to your two competitors with whom you share this Top), then this is a sign that your site for some reason is not suitable as an answer to the question asked in the search line. Even if it is not, you have every chance of getting a demotion.
Moreover, you need to understand that there are no “ideal values”: otherwise, in unpopular topics, no site would correspond to them (too high requirements), or vice versa, in popular ones all would correspond (too low). That is, several types of data are analyzed at once, a certain “corridor” is calculated within which the site indicators should be based on its positions.
Therefore, it is imperative that the snippet prompts you to click on the link. And this is possible only if the user sees in him the most complete and close to the asked question answer. The inclusion of query keywords in the title, mention of them and thematically related words in the snippet text, additional links, relatively recently appeared online chat directly from the search - this all works for you. And with all these factors it is necessary to work and monitor them.
Search and site behaviors are two factors that intertwine with each other. If the user goes to the site, but then returns to the search and clicks on the next result, then this is also a signal for the search engine. If a site collects more clicks from the fourth place than the first three, then it has reason to claim the top lines. User actions are considered as some kind of communicative act between a machine and a person, in which a person sequentially searches for an answer to his question, and the machine offers options and tries to learn. And your site in this chain is both a goal and a source of statistics.
Working with a snippet includes search services for a webmaster, directories, content, internal optimization, work with title and meta tags.
We found that formal factors are not exhaustive for search engines in rankings. Special attention is paid to the analysis of user behavior.
We also found that behavioral factors do not exist separately from the promotion process, on the contrary, they are organically integrated into the system. This is not surprising, because the user's behavior directly depends on what the search system offers him and then the site itself. The search for the desired content or service is a single and indivisible process into which we must integrate and which we must subordinate to the scenario that we have previously thought out.
And work on this scenario begins already when we determine the main goals of promotion and develop a strategy. No further action is taken to achieve the first lines in the results without taking into account these basic provisions.
On the other hand, a strategy can and should be changed if you see its fallacy. Miscalculations can be at various levels, including the level of business planning. In any combination of circumstances, stopping deliberately disadvantageous measures and adjusting the plan will be cheaper than continuing unchanged.
BUT! Changes at any level automatically entail the need to review all related indicators. The system must be in constant balance, so you can’t change the semantics and not change the content, and therefore usability. Somewhere it will be necessary to redo everything, somewhere there will be enough cosmetic transformations, and somewhere, by a happy coincidence, it will do without changes at all.
In addition to the obvious consequences, such as “changed the structure and did not change the links” or “did not configure redirects”, there are those that will not immediately manifest themselves. They are the most dangerous, because if you did not immediately identify them, then your analytics is poorly built, and it means that at a later stage it will be even more difficult for you to identify them. This is especially true for behavioral factors.
A systematic approach to promotion means that there are no trifles in it: any trifle in one way or another will affect the balance of the system. Moreover, minor flaws should be considered more dangerous than large ones: they are more difficult to track and more difficult to correct when there are a lot of them.
For the same reason, it is important to track the source data (if you already have a site, and especially if it already has a promotion history). Even if you set goals and objectives, this does not mean that the site is suitable for their solution. It makes no sense to purchase links and write texts if the user cannot find the Buy button, or leaves without waiting for the page to load. Firstly, you still can’t get to the right positions, and secondly, promotion will not give conversions.
The one described in this series of articles may seem trite to someone: a systematic approach is good in any business. All this is logical and, it seems, should be intuitive. The problem is always in execution. We know how to, but we don’t always do that. There are a thousand and one reasons why it is not possible to work on the “maximum” program. There is no budget, time limits, the customer does not like it and so on, the list goes on and on.
But even if it is impossible to fulfill one hundred percent, this does not mean that ten is enough. On the contrary, the system needs to be balanced as much as physically possible. Because a stable system with any outcome has a better chance of success. If you know that you have a clear “bias” in some place, then its elimination should be kept in plans, albeit not for the next month.
Previous parts:
- First part: Strategy
- Second part: Technical optimization
- Third part: Content
- Fourth part: External optimization
Now I suggest recalling our main scheme:
Let me remind you that the main element that immediately affects everyone else is Strategy. And the elements themselves are interdependent on each other. Therefore, in fact, in the name of the concept there is the word “balanced”: any serious changes in each of the aspects of promotion automatically take the whole system out of balance and require immediate restoration of balance.
If with the technical side of the site everything is basically obvious, but with the content and the linking environment a little less, but still understandable, then there are often problems with usability. They are connected with the fact that the very concept of "usability" is not perceived by many as very true, and therefore it is applied incorrectly to SEO and becomes a missing link.
Let's see what it is and why it interests us so much.
Usability
In a word, translating this term into Russian is impossible. Translated literally, it is “usability”. A less accurate but more familiar option is “usability”.
Usability is a conditional quantity that determines how capable the average user of a tool is to achieve the desired result using this tool in typical conditions. And also how easy he will be able to do it, how much he will be satisfied with the process of work.
In the most deplorable cases, the user is not able to use the tool at all, in the most successful ones, he is able to master it directly during work without any basic knowledge, due to his own life experience.
The site on our scheme is no coincidence, although it is in the central circle, but it is not a headline: it is a tool. And the goal, as we said in the first article, is to perform targeted actions by users (receive conversions). So in our case, usability will determine how fast, with what emotions and how much the targeted users will be able to achieve the goal.
What is useful for SEO
Today, search algorithms take user expertise very seriously, the very behavioral factors. This is quite logical from the point of view of the logic of search engines: the orientation toward satisfying user interests is openly declared by them in their own license agreements. Obviously, how much the site is made “for people” is better to ask the people themselves.
Therefore, usability is very important for search promotion: it is an opportunity to give search engines a free and unbiased reason to consider a site better. Of course, for this it will be necessary to work and closely monitor the fate of your project, but in the framework of a single strategy, this is not as difficult as it might seem.
Analysis
Understanding that the site has usability problems is not so difficult. This information is given to us by metrics. Moreover, they are constantly developing and improving their own usability indicator, so the most popular and necessary reports are already configured out of the box: traffic sources, entry and exit points, page time, viewing depth - all this can be viewed and analyzed the very next day after installation code. You can even see in the first person how the user scrolls the page and moves the mouse cursor.
And from this moment you can already draw conclusions about how “suitable for use” the site is. Of course, it will be necessary to set up event chains, goals, and if you are engaged in online sales, then e-commerce. But this is not so difficult, and the principles of analysis there are about the same.
Metrics give us the opportunity to look at the site through the eyes of the user. And then all its weaknesses, design errors, incorrectly guessed intent, and much more that can be overlooked at the development stage are revealed. Many will be scared by what they saw, but believe me, it’s much worse to not understand what you are doing wrong.
Particularly dramatic data is provided by end-to-end analytics systems, which I also mentioned in the second part of our series. They allow you to calculate business indicators such as ROI (return on investment) for your investments in the development of the site. So you can see in real numbers how profitable, say, was the decision to change the design of the pages, how quickly it paid off and whether it paid off at all. Based on these data, you will be able to more accurately plan budgets, understand exactly which solutions allow you to spend one time and reduce regular costs and payments.
End-to-end analytics allows you to quickly identify, for example, that you are ordering work for the most unprofitable of the available schemes (for example, for a monthly fee, rather than buying hours of work, or vice versa). For especially large projects, this can save huge amounts.
Work with the site
Metrics reveal errors in various aspects of promotion. We can see incorrect clustering of keywords, due to which the user gets to the wrong page that better meets his specific task. From here we get the need for re-clustering, changing the content of the pages (and sometimes creating new sections and changing the structure), changing incoming links - in a word, the whole system instantly responds to this error.
We may find that the forms do not work correctly on the site and the user cannot perform some simple action. We set off to study the code, check the server settings, and immerse ourselves in technical optimization.
It also happens that the page is loaded with errors, and the user leaves it immediately after opening, because he sees the impossibility of interacting with it, or simply perceives it as an indication of the quality of the site and decides not to have anything to do with it.
Remember, when we talked about content, I mentioned that it should answer the questions that a user has as fully as possible while studying? Now we can find out how well he does it. And if not enough, then make the appropriate changes. Can't do this within the existing page structure? So, we are trying to change this structure. Or maybe it will be easier and more promising to change the structure of the site itself?
Content with usability is generally very closely linked, in many aspects it is impossible to distinguish between them: content is an important part of the page and one of the main goals of the user. And it's not just about the text: well-placed infographics or other visual content captivates the visitor, creates a sense of harmony, helps to get what you want. Illiterate visual content interferes with the perception of information, draws attention to the poor quality of performance, negatively affects the perception of the credibility of the resource and encourages him to leave.
Of course, usability is not limited to content. This is a combination of all the factors that translate into the final picture: the logical structure of the site and pages, the correctness and relevance of the forms, ease of navigation, the presence of full-fledged interaction scenarios designed for different user groups, and much more.
It is no coincidence that all large and popular projects constantly refine the interface: the Internet is constantly evolving, so the user's perception of certain decisions is changing. Familiar tools become obsolete, new ones become more familiar.
Trends in the design and layout of pages live according to the general laws of fashion: if everyone makes a “Buy” button in green and with rounded edges in the lower right corner, then the user will look for it there, and that’s it.
Therefore, an ideal option would be to conduct A / B testing, when different versions of pages are shown to different users. Based on the results of this testing, the most successful solutions remain, and those that have not proved to be the best are changed or deleted. Unfortunately, this is a rather costly method and requires a representative sample. Therefore, usually large sites with high traffic resort to such testing.
There are no trifles: if you have a large failure rate, then this is bad not only in terms of conversions, but also in terms of promotion. Users come from the search and immediately leave - what should this tell the search engine? That's right: that the site is in the Top 10 undeservedly.
Work with snippet
But in fact, the evaluation of your site by the user begins even before he clicked on the link. Let's talk about snippet, a small announcement that the search engine shows on the results page.
The user’s behavior in search is analyzed by search engines in no less detail than his actions on the site. These data allow the search engine to once again check the quality of the results, exploring the experience of living people.
The fact is that algorithms are far from ideal, they make mistakes. And it happens that the results fall on sites that do not belong there. Therefore, data on clicks on links is also an indicator of the relevance of the page to the request.
If you are in the Top 3 and nobody clicks on you (and yes to your two competitors with whom you share this Top), then this is a sign that your site for some reason is not suitable as an answer to the question asked in the search line. Even if it is not, you have every chance of getting a demotion.
Moreover, you need to understand that there are no “ideal values”: otherwise, in unpopular topics, no site would correspond to them (too high requirements), or vice versa, in popular ones all would correspond (too low). That is, several types of data are analyzed at once, a certain “corridor” is calculated within which the site indicators should be based on its positions.
Therefore, it is imperative that the snippet prompts you to click on the link. And this is possible only if the user sees in him the most complete and close to the asked question answer. The inclusion of query keywords in the title, mention of them and thematically related words in the snippet text, additional links, relatively recently appeared online chat directly from the search - this all works for you. And with all these factors it is necessary to work and monitor them.
Search and site behaviors are two factors that intertwine with each other. If the user goes to the site, but then returns to the search and clicks on the next result, then this is also a signal for the search engine. If a site collects more clicks from the fourth place than the first three, then it has reason to claim the top lines. User actions are considered as some kind of communicative act between a machine and a person, in which a person sequentially searches for an answer to his question, and the machine offers options and tries to learn. And your site in this chain is both a goal and a source of statistics.
Working with a snippet includes search services for a webmaster, directories, content, internal optimization, work with title and meta tags.
Total
We found that formal factors are not exhaustive for search engines in rankings. Special attention is paid to the analysis of user behavior.
We also found that behavioral factors do not exist separately from the promotion process, on the contrary, they are organically integrated into the system. This is not surprising, because the user's behavior directly depends on what the search system offers him and then the site itself. The search for the desired content or service is a single and indivisible process into which we must integrate and which we must subordinate to the scenario that we have previously thought out.
And work on this scenario begins already when we determine the main goals of promotion and develop a strategy. No further action is taken to achieve the first lines in the results without taking into account these basic provisions.
On the other hand, a strategy can and should be changed if you see its fallacy. Miscalculations can be at various levels, including the level of business planning. In any combination of circumstances, stopping deliberately disadvantageous measures and adjusting the plan will be cheaper than continuing unchanged.
BUT! Changes at any level automatically entail the need to review all related indicators. The system must be in constant balance, so you can’t change the semantics and not change the content, and therefore usability. Somewhere it will be necessary to redo everything, somewhere there will be enough cosmetic transformations, and somewhere, by a happy coincidence, it will do without changes at all.
In addition to the obvious consequences, such as “changed the structure and did not change the links” or “did not configure redirects”, there are those that will not immediately manifest themselves. They are the most dangerous, because if you did not immediately identify them, then your analytics is poorly built, and it means that at a later stage it will be even more difficult for you to identify them. This is especially true for behavioral factors.
A systematic approach to promotion means that there are no trifles in it: any trifle in one way or another will affect the balance of the system. Moreover, minor flaws should be considered more dangerous than large ones: they are more difficult to track and more difficult to correct when there are a lot of them.
For the same reason, it is important to track the source data (if you already have a site, and especially if it already has a promotion history). Even if you set goals and objectives, this does not mean that the site is suitable for their solution. It makes no sense to purchase links and write texts if the user cannot find the Buy button, or leaves without waiting for the page to load. Firstly, you still can’t get to the right positions, and secondly, promotion will not give conversions.
The one described in this series of articles may seem trite to someone: a systematic approach is good in any business. All this is logical and, it seems, should be intuitive. The problem is always in execution. We know how to, but we don’t always do that. There are a thousand and one reasons why it is not possible to work on the “maximum” program. There is no budget, time limits, the customer does not like it and so on, the list goes on and on.
But even if it is impossible to fulfill one hundred percent, this does not mean that ten is enough. On the contrary, the system needs to be balanced as much as physically possible. Because a stable system with any outcome has a better chance of success. If you know that you have a clear “bias” in some place, then its elimination should be kept in plans, albeit not for the next month.