So, we have the first result, which is not ashamed to boast. That's why we present you a case on SEO-optimization on the example of one article from our website. Thanks to the analysis of queries in Yandex Webmaster and Search Console, for which our article was shown in the search results and competent article scrolling, we managed to increase traffic to the page with the article by almost 6 times! 6 times, Carl!
Initial data: Article "". Published on December 14, 2016. Indexed by Yandex on December 21, 2016, at the 13th position of the main issue, at the request that the article is titled, in a week, namely on December 29, it rises to the 4th position.
Finalization of the article
After the analysis of search queries, which revealed additional keywords and phrases, it was decided to make changes to the content and SEO settings of the article.
1. Changed title and snippet. Removed the words that users do not use in their queries, if the original title was: "How to set up Yandex Direct - instructions for use", now it has become "How to set up Yandex Direct - step by step instructions IM".
What means? Using the prefix IM, we made the title unique and indicated that the instruction was step-by-step.
We identified site articles that also received a high rating from Yandex and are indexed at least in the TOP-10, and also somehow relate to Yandex Direct or contextual advertising. From these articles, links to the promoted article were affixed, key phrases were used as anchors: "Direct, Yandex Direct and Yandex Direct Settings".
3. Improved the behavioral factor.
Article SEO results
Starting from January 20, we began to appear in the TOP-10 of the main issue for such queries as “Yandex Direct Settings”, “Direct” and even for the query “Yandex Direct”!
As a result, from December 14 to January 14, this article had 166 visitors, and from January 15 to February 15, already 976. It is not difficult to calculate that traffic growth was 587%.
As of February 16, 2017, we are in the TOP-10 Yandex search results in Russia and Moscow, on request, the frequency of which is more than 114,000 per month. And in the TOP-3 by request, which has a frequency of more than 3000 per month. Moreover, in both cases, these are not general indicators, but an exact occurrence. Therefore, we can safely add a huge tail of low-frequency queries to this.
Imagine businessman Boris. He has 30 employees in the company, an office in the center of Moscow and a website through which he sells refrigerators. Boris wants to get more sales through the Internet. He is sure that this is as easy as shelling pears: it is enough to stuff the site to the eyeballs with SEO texts, and the sales department can be increased due to the influx of new customers.
Boris assigns the task to SEO-specialist Gennady. He does everything as asked: the website of Boris's company is now overflowing with pages with useless text like "Buy cheap goods", "Buy goods in Moscow without registration and SMS". However, after a month there are no more visitors on the site, and the sales department is not torn apart by new calls. What's the matter?
Businessman Boris has become yet another victim of the incredibly enduring myth that SEO texts with high-frequency queries are a fail-safe way to increase traffic to your site from search engines. Let's reveal a secret that will save everyone a lot of time, money and nerves: no, it's not.
Multi-part SEO filters with higher conversions on the site
How can you really increase website traffic from search engines, what is needed for this and how to do it today?
We have already talked about increasing SEO performance with category filters. This method is suitable for any site and business and will allow you to fill the resource with pages that will work specifically for your product. If you have already applied the method described in it and want even more traffic, then here is a more advanced one for you - creating multi-part SEO filters on the site.
The technique of multi-component SEO filters has proven itself well with our clients. One of the most effective results was shown by the case described at the end of the article. The client company, on whose website this method was applied, increased its sales for the year by 50,000,000 rubles - only due to small changes on the website. Businessman Boris never dreamed of such a thing.
Once again about SEO filters and meta tags
SEO filters are the pages on your site that are designed to get organic traffic from search engines. In order for such pages to bring traffic, you need to correctly fill in meta tags and headings: title, description, keywords and h1. When filling them out, be sure to use the search query in the exact wording in which the user enters it into Yandex or Google - this will increase the chances that your site page will take a high position in the search results. Simply put, you will get more traffic to your site.
It is important to remember that meta tags for a particular page should correspond to the content of the page itself, and the content of the page should correspond to the site as a whole. Otherwise, you will soon get an increase in the bounce rate: realizing the deception, users will leave the site, and you will not get any increase in traffic, let alone sales.
Single and multipart SEO queries
Many online stores have SEO filters, for example: “street doors”, “boots with heels”, “Android smartphones”. These are pages that combine products that contain one common characteristic. Such filters are called single. From an SEO point of view, they are good - they meet the search query and the content of the site, include quality content.
However, single filters on the site end sooner or later. And then comes the time for our advanced method: creating SEO filters for multi-part search queries.
Multi-part queries- these are queries containing several characteristics or components at once. Such queries are aimed at finding a specific product. Accordingly, multicomponent filters are filters created by mixing characteristics or components related to one object.
Multi-part filters allow you to create new pages for SEO optimization that can bring relevant traffic. In addition, the conversion to purchase from such pages is higher than the average for the site.
Good Examples of Filter Page Titles
- Extension plastic with 4 sockets with fuse– 3 components
- Sony mobile phone with a 5 inch display with a long-lasting battery for android 5- 3 components + brand as a filter = 4
- Winter Jacket green with side pockets– 3 components
- Plastic double doors window heated doors with armored glass- 4 components.
- doctor visit at home in Moscow around the clock– 3 components
- tours Bali for 14 days all inclusive– 3 components
- shipping small cargoes in Russia - 2 components
Methodology for promoting SEO filters
To promote the pages of your site for low-frequency multi-part queries, you need to do the following:
- Collect all single queries that match your site
- Construct multipart queries from them
- Select the most popular queries
- Create web pages for multipart queries
Let's go through the points.
1. Collect single queries
Single requests - the characteristics of your products presented on the site, as well as the components of the search query. They are collected manually through Wordstat.yandex or using third-party servers.
2. Design multipart queries
Based on the list of single queries, you can now generate multipart queries. To do this, you need to take queries related to the same object and mix them up. As a result, you will get requests of three, five, seven words - and this works great for increasing traffic.
As in the first step, construction is easiest done using a script in which it is important to set a rule: only characteristics in a common basis should be mixed, otherwise, for example, “high-heeled doors” may turn out.
The result will be hundreds of groups with thousands of combinations of characteristics of one product, for example:
Word Order Variations in a Query
After you have collected all possible combinations of characteristics, you need to choose the most effective of them - the most popular in search engines.
3. Select the most popular queries
In 2017, Yandex introduced operator "", which made life easier for SEO-optimizers. This operator allows you to see the frequency of a request depending on the permutation of words inside it. However, for rare phrases, Wordstat.Yandex will still show 0, which does not mean zero demand: buyers still come to sites through such requests.
In the statistics of one of the clients, we saw that requests that Wordstat defines as zero still gave traffic.
Wordstat does not give a 100% accurate search rate for rare phrases. Therefore, it is worth evaluating demand and choosing the best word order through search results.
For example, you need to choose which of these queries occurs more often:
- "Buy green boots"
- "Green boots buy"
- "Green boots to buy"
Before the advent operator "" the path would be as follows. You access Wordstat using the "!" operator, which doesn't look at the exact word order in the query, and the system says:
We see that the number of impressions does not change from the rearrangement of words, but this cannot be.
After the appearance operator "" It has become easier to check multi-part filters in Wordstat, however, as we have already seen, for rare phrases the tool often shows false zero demand.
Determination of optimal queries
The process of determining optimal queries takes place in three steps:
1. Skip the collected requests through Yandex, select the most popular
Google produces results more accurately, but with a large number of requests, it bans more often. Yandex is more loyal in this regard, so we make the initial screening of requests through it. All requests must contain operator "".
At this stage, we refuse those combinations that give 0 at the output (this order of words was not found in the collection of Yandex documents).
2. We pass the remaining requests through Google, choosing the most popular
We use again operator "", we select the most frequent requests.
You can automate this process using the A-parser tool.
As a result, there will be 2-3 combinations of multi-part queries for each of the groups of site objects.
3. From the resulting selection, manually select the final queries, on the basis of which we will create filters on the site
We process requests directly with our hands and from each set of combinations we leave one best one at our own discretion.
For example, we need to understand the most frequently requested combination of words for the following concept:
We stop the choice on request No. 2: it has a higher frequency and is closer to what a living person will search for in a search engine.
4. Create web pages for multipart queries
Under the final combinations, you need to create site pages. These will be pages in which products are collected according to the characteristics specified in the requests. It is absolutely not necessary to place additional text on them, but content that meets the specified request, that is, product cards, must be mandatory.
The request received as a result of filtering is embedded in the meta tags of the page:
- H1 - final remaining request;
- Title - must be unique! We add an entry H1 - in fact, this is our main key, but it should not match H1;
- H2 - diluted H1 (for example, "Buy a white metal pipe 60 mm in Moscow");
- And do not forget about the static CNC (human-readable URL, consisting not of a set of characters, but of words). The address must be in Latin, it is better not to use Cyrillic.
Why doesn't everyone do this?
The problem is that parsing even filtered requests and integrating them into the site is a huge job. But it can be partially automated: create a matrix of optimal queries in Excel with the necessary data and ask the programmer to write a script that will implement queries on the site. After SEO-specialist Gennady, one click will be enough for businessman Boris to enjoy the results of competent SEO-optimization and count sales on the site.
Case: how SEO filters brought 50 million rubles to a client
The company is engaged in the manufacture and sale of doors. Single SEO filters were created on the site - “street doors” and “insulated doors”, which immediately began to bring good traffic.
Traffic statistics for SEO-filter "street doors"
Traffic statistics for the "insulated doors" SEO filter
To further increase traffic from search engines, we decided to experiment with crossed filters.
Step-by-step algorithm of actions
1. We divided the collected door parameters into groups
Conditions were set: parameters within the same group cannot be crossed; parameters of different groups can intersect with each other.
2. Mixed those pairs of characteristics that can be mixed
For convenience, you can operate with the numerical values of the groups. For example, the page under the query-filter “doors to the house is insulated from the street” has the index ;;. And the query "insulated street doors" index.
Thus, having recreated the valid combinations, we got all possible crossing options. Let us dwell in more detail on the combination - insulated street doors.
3. Generated all sorts of word order options in the query
After the system has shown us that the combination "insulated street doors" has the right to be (previously set conditions are met - the characteristics are in different groups, see the figure above), we need to generate all possible variants of word order in the phrase.
At this stage, it is important to mention additional service words, without which the phrase will not have a complete meaning. In the subject of doors, the word “doors”, “entrance”, “metal”, “steel” can be used as service words. Moreover, the word "doors" in the final phrase is always used, and the rest of the service words may not be used.
Using a simple script, we generate combinations:
- insulated street doors
- insulated street doors
- insulated outdoor doors
- outdoor insulated doors
- insulated street doors
- insulated street doors
- insulated street doors
We remind you that the first three points can be done manually, but this is a very voluminous work, so we recommend that you ask a programmer for help.
4. The final list of requests was processed through Yandex and Google
At this stage, requests that show a zero result were cut off. The resulting list of queries was passed through Google.
5. Selected final queries manually
As a result of parsing, we still have the most popular requests in the list. Among them, there may be queries with the same number of results, for example:
- outdoor insulated doors (100 results in search results)
- insulated street entrance doors (100 results in search results).
In each such case, we independently decided which phrase would be embedded on the site. We chose the query "insulated street doors".
6. Prepared a table of meta tags
In each meta tag, we used the phrase "insulated outdoor doors." This table was imported to the site.
7. Imported to the site
After preparing the matrix in Excel, the programmer wrote a handler that allowed him to import new pages to the site.
Bitrix filter page interface
8. Updated sitemap
Don't forget to add new page URLs to your XML sitemap and speed up search engine indexing if necessary
results
The multi-component filter "insulated street doors" began to bring the client 20 users a week - from only one filter!
In total, we created 300 filter pages on the client's website. Each of them gives an average of 7 people a week. The site collects 6,000 people a month from all filter pages. Recall that conversion to sales from such pages is higher than in general on the site, because the most relevant traffic goes to them.
During the year, taking into account seasonality, 2,016 sales were made from multi-part filter pages! With an average check of a client of 25,000 rubles, filter pages per year brought sales to the client in the amount of 50,400,000 rubles!
If businessman Boris knew about the multipart query method, then instead of useless SEO texts, his site would have effective filter pages that would bring profit to his business.
(Visited 5 002 times, 1 visits today)
Dear friends, I decided to make a selection of useful cases on how to increase website traffic.
For many years I have been storing the best materials in various areas of Russian and English SEO in my Evernote bookmarks, and from time to time I look through them to brush up on interesting developments.
It's time to share your bins with the readers of the site 😉
In fact, there are a lot of cases on how to increase website traffic in Runet, much more than I cited in the article. But a significant part of them is uninformative and does not carry any value.
I collected only those cases that I liked and that have interesting tips and best practices. All cases relate to promotion in the Russian-speaking segment of the Internet.
I ran into a little problem while compiling the list. Many sites in the articles do not have a normal header image. It’s just some kind of trouble 🙂 Therefore, for screenshots, I just used the title of the article or statistics with a traffic graph.
It is advisable not only to look at the listed articles, but to briefly write down the ideas you like, and then be sure to apply them to promote your sites. I do it myself.
I placed the cases in random order.
A selection of cases to increase attendance
So let's go! 🙂
Taxi website promotion case
An interesting point in this case is that before creating the site, the necessary search queries were collected and it was found out that in the subject of a taxi there is a very large proportion of queries with the word "cheap". Based on this, a domain was registered that contained this word.
Based on the analysis of competitors' sites, they created approximately the same number of pages on the site being promoted as the average for competitors.
Services pages have H2 subheadings in the form of questions that users often ask along with the search query being promoted for this page. From my own experience, I can say that this has a positive effect on increasing traffic.
Promotion in RuNet.
@burzhunet - English SEO.
Taking advantage of my official position, I remembered my own case 🙂
Back in 2008, I set up a public experiment and increased my blog traffic from 350 to 1000 people per day in 10 weeks and on time. Then he wrote a series of 4 articles, describing in detail what and how I did to increase traffic to my site.
Although it was written a long time ago, much of the information in it is still relevant today.
Plus, this is the only case in RuNet, which consists of 4 parts 😉
Liberta case: promotion of a mirror production site with 501% ROI
I have seen many times in practice that creating and regularly blogging on a commercial site is an effective way to increase traffic. In this case, it was just the use of this method.
The article discusses the method of content re-optimization, which I also actively practice, I wrote about this in a post. This approach works great both in RuNet and in the English-speaking Internet.
SEO for a Moscow online spare parts store with a budget of 15,000 rubles per month
- The reference mass decreased gradually.
- A lot of work has been done to improve the optimization, functionality and structure of the site.
- Commercial factors have been improved.
As a result of the actions taken, site traffic in Yandex increased from 2,500 to 10,500 visitors per month. And this is on a limited budget.
Case: How did site traffic increase by 8 times in 4 months?
Petr Alexandrov has an interesting approach - he collects the semantic core not immediately for the entire site, but for its individual categories, so as not to waste time.
The post mentions that the strong keyword highlights have been removed. I also often notice that many site owners still highlight the queries promoted on the page in bold. It's been a few years since it stopped working. Moreover, because of this, the site may fall under the filters of search engines.
Increase in search traffic from 60 to 4500 clicks per month
To promote this site, articles were also used, of which 15 were published per month. Some of the articles had numerals in their titles, for example, "5 stunningly unusual ideas for curtains in the living room", "7 details that will make your curtains irresistible."
As you know, articles with numerals in the titles are more effective than regular ones. They are especially well suited for attracting visitors from social networks.
I liked that the case describes the creation of an e-mail newsletter on the site being promoted. This is an effective way to form a permanent audience of site readers. Some of the visitors will visit the site again and again thanks to the newsletter.
Search engines take into account such repeat visitors, and this works well for website promotion. Plus, search engines take into account the share of non-search traffic from sites.
The more non-search visitors (referrals from emails, social networks, other thematic sites, etc.), the better.
Case: from 106,276 to 411,956 search visitors per month
E-mail marketing was used, as well as a blog was created and a technical audit was carried out.
Along with other measures, this provided the necessary impetus for a tangible increase in attendance.
Case: 3.5 times increase in traffic to a news resource in 6 months
In the promotion of news sites, there are some peculiarities that need to be taken into account.
I have a news site in progress. The key to success in this area is to do a thorough job of internal optimization and elimination of technical problems in order to competently prepare the site for inclusion in major news aggregators.
If everything is done correctly, then in a few months the site will receive good traffic from news aggregators:
- Yandex News
- Google News
- Rambler News
- News
- Novoteka
With getting into news aggregators, the issue is also resolved with obtaining high-quality links - the news will quote other news sites and put a link to the source.
Judging by the described actions in the case, everything was done correctly.
Case: Growth of a content project by 1.5 times to 1 million visitors
It was decided to move the promoted site to a dedicated IP in order to avoid a "bad" neighborhood with sites of gray topics.
Detailed instructions for 5 pages were written on the design of articles and their content.
Also, an analysis and optimization of old pages was carried out, the positions of which were in 20-30 places. When a query is already on the 2nd or 3rd page of the search results, in many cases a small adjustment of texts and metagets is enough to get the query to the top.
When creating new articles, we studied popular articles of leading competitors that attract traffic, and based on them, topics for new materials were selected.
Case: how to increase traffic for an online store of home appliances by 4 times
In this case, I liked the following points:
- Used reviews on product pages, which gave additional traffic.
- Separate pages of the type were created: tag and tag + brand in order to attract targeted visitors by search suggestions.
- A/B tests were conducted to improve conversions and increase profits.
Case: promotion of construction services
Perhaps the most beautiful picture of all cases 😉
One of the main directions in terms of cash costs for this site were:
- Creation of video content
- Buying quality links, reviews and press releases
- Quality website content
The structure of the site and the formatting of the content on the pages have been redesigned.
To attract additional traffic, a blog was created in which about 300 articles were published.
How to increase traffic to the site by 2 times due to "unnecessary requests"
The emphasis in this example is on attracting targeted commercial traffic for low-frequency queries.
As a result of the analysis of search queries in the selected topic, it was found that 60% of traffic falls on queries with a length of 5 or more words.
Based on the data obtained, the tagging method was used.
What is tagging? This is the optimization of tag pages, as well as the creation of new tags for low-frequency queries, usually used on online shopping sites.
As a result, the site began to receive an additional 3,500 targeted visitors only for low-frequency queries, excluding medium and high-frequency ones.
Promotion of a medical site - a general clinic
Promotion of medical sites has its own characteristics. The case describes that work was carried out to add a site to medical portals, and activities were also carried out to attract targeted traffic from regional forums.
The site being promoted has been added to Google My Business and Yandex Directory.
In addition, press releases and articles about the clinic's services were posted in regional portals and news sites, which also brought additional visitors.
In this case, I liked that the page loading speed was accelerated.
Case: a new site in the Yandex TOP for a year and a half
Before the start of the site, the client did not yet have, therefore, a detailed analysis of the sites of the leading competitors in the field of heating was carried out.
With this in mind, the site was created and its internal optimization was carried out.
Later, based on the analysis of pages from the top 10, the semantic core was expanded for the necessary queries.
Seasonality in promotion is about sledges in summer and carts in winter. Of course, you won’t sell them, but you should prepare properly in order to squeeze the maximum at the peak of demand. Contextual advertising in this sense is more flexible, but seasonality must also be taken into account in SEO promotion. How? Let's talk about the example of promoting an online store of tires and wheels VIANOR.
To bookmarks
We started working on the site in June 2017. The site was over 4 years old at the time. The customer is a representative of a network of tire centers owned by the Nokian Tires concern, the largest network in the Nordic countries, Russia and the Baltics. Turnovers are significant, traffic - too. An error can result in the loss of six-figure orders.
The site is large even by the standards of an online store: about 6 million pages, of which more than a million are in the search engine index. An additional difficulty is seasonality, which is expressed not only in a sharp increase and decrease in traffic, but also in a change in the pool of active requests. In October, buyers are looking for winter tires, and in April - summer tires. Therefore, the standard approach to the formation and updating of the semantic core is not suitable - segmentation is required.
Segmentation of seasonal queries
From the point of view of the customer, it looks like this: the standard reporting file for positions is supplemented with tabs, where queries and indicators by groups are displayed. In our case, these were general requests (we work with them all the time), batteries (we are working on, monitoring, not a priority), winter and summer tires (we are actively working during the season).
Query Segmentation Work File
So it is convenient for the customer to track the results for each group and compare them with the general ones in terms of semantics.
What does this mean for us?
· Actualization of groups and frequency only in the same period.
· Group reporting.
· Selection of groups for work in accordance with seasonality.
results
First, let's evaluate the results of traffic from search engines. With pronounced seasonality, it is impossible to evaluate it solely for the current period: changes will be directly related to the interest in the product. So we compare year by year. At the peaks of seasonality in October and April, we received a good increase.
Traffic growth at seasonal peaks
Traffic must be targeted in order to convert successfully. You can evaluate the compliance of pages with the expectations of visitors, for example, by the percentage of bounces (the visitor immediately left the page). According to the results of April, the percentage of failures decreased from 14 in 2017 to 8.61 this year. This means that the likelihood of converting leads has increased.
The audience becomes more targeted
Let's check if this is the case - let's see the conversion charts:
Conversion growth
Even the graph shows that both the number of goal achievement and the % of conversions have increased. But do not believe your eyes - we have numbers.
We also evaluate the results of promotion on the requests of the semantic core. Visibility (the sum of exact search frequencies in the top 10) during our work increased from 282 to 30,737, and the average position decreased from 93.55 to 34.43.
Growing visibility
Decreasing average position
As mentioned earlier, we report by request type.
Loading...
