- demanded interesting content;
- diversity;
- opportunity to find new sources of information.
There are new communities in the recommendations, individual posts with interesting information, and they all correspond to your interests and needs. Of course, it also contains advertising, but in it all communities, goods and services relate only to topics that interest you.
The link to the section is located to the right of the news feed, almost immediately under your name and photo. The first line in this menu is just news, and below it - recommendations. By clicking on this inscription, you will be taken to the section with recommended posts.
Actually, how to get into the recommendations in VK, an attentive user will understand at a glance. It looks very similar to your news feed - posts from different public pages are placed under each other. They, just like in the news, are sorted by publication time. But, unlike news, here you will only see one post from each community. It is the records that are selected by the program.
By clicking on a post, you will open it in reading mode, and following the links, visit the recommended public pages, you can view other information in them and, if you wish, subscribe.
If the news feed displays posts only from those groups to which you are subscribed or from those personal accounts that are friends with us, then in the "recommended" feed - for the most part you will find posts from the wall of those who have not yet become your friend or with communities in which you are not yet a member.
A complex algorithm is integrated into the site, which every minute "spies" on you, your friends and subscribers. Any activity is taken into account: viewing information, comments, likes, reposts on the wall or in your public, forwarding posts in private messages, etc. This is how the program identifies your preferences and interests, and based on this information, the system issues a list of entries that owe you please.
Immediately, I note that the algorithm works on the site not for the purpose of spying on you, but for the selection of important information. It is not worth worrying about the safety of personal data and the secrecy of correspondence. No information other than adherence to a particular topic is retained by this program.
Yes, it can be customized according to personal preferences, and even a person who is on the site for the first time can handle it. There are no complicated menus here, remembering how to delete recommendations in VK is easy the first time. A simple menu automatically appears next to each entry. To expand it, hover your cursor over the small checkmark to the right of it. The menu will open itself. You just have to choose an action.
- It is not interesting. The post will simply disappear from view. If you immediately click the blue hide from feed button, posts from this community will no longer appear in the recommendations. If you change your mind at this point, click Mark Action.
- Save to bookmarks. The post will be saved in a section, which you can find by following the link in the menu on the left.
- Complain. If you see information that is contrary to the provisions of the Criminal Code, contains insults or any negative information, you can. Click on the line corresponding to the line, then select one of the items in the window that opens. The admin will see this message, analyze the data and take appropriate action. Unfortunately, you will not be able to find out anything about whether the offender has been punished.
- Notify about entries. This feature is akin to a subscription. You will receive notifications that new publications have appeared in the community you like. You will not miss them, as is the case with a crowded news feed. These notifications come to the "bell" - which is located at the top in the middle, if you open VK on your computer.
When you choose to opt-out of these notifications, you can do so as follows:
- go to notifications by clicking on the bell icon at the top of the page;
- find the title of the post from the community you need;
- hover the cursor over the checkmark opposite it;
- select "Hide notification";
- click on the inscription "Do not receive notifications from" under the message that the message is hidden.
The settings for each entry and group are automatically saved and remain unchanged until you reconfigure the tape yourself. You can return entries to the recommendation list by going to filters.
For this:
- click on the funnel icon next to the news feed link in the same menu as the recommendations;
- find the name of the group;
- click on the empty white circle opposite it so that a checkmark appears in it;
- click save.
Musical recommendations in VK: how this tape works when selecting tracks
This section works in a similar way, but is located in a different location. To enter it open your music. Here you will see four tabs: My Music, Playlists, Recommendations, and Friends Updates. The third tab will contain tracks that, according to the VK robot, are not yet known to you, but you will definitely like them. This is a great opportunity to learn about new recordings and artists.
This section is located in the right menu, if you do not have it, then you need to enable it. To do this, go to "settings".
In the window that opens, select: Customize the display of menu items \ and select the "Music" item - put a tick in front of this item and click "Save". Now the item "Music" will be displayed in your left menu.
The algorithm on the basis of which music recommendations work takes into account:
- the content of the user's tracklist;
- likes and comments left by the user from communities and on the walls of other people;
- reposts from communities and other personal pages;
- reposts from private messages and other activities related to music or information to which it is attached.
The more varied your tracklist and wall, the more varied the music (any audio) will be offered to you in the recommendations.
Important! Some users have noted imperfect recommendations, especially with regard to music. In their opinion, they come across tracks that are far from their preferences. The same claims are made by those who regularly read the recommended posts. However, such complaints are rare. Sometimes it is not the algorithm that is to blame for this, but the user himself, who is not aware of the intricacies of the "Recommendations" section in VK, how this tape works and how he selects tracks.
For example, you read an interesting article with beautiful photos, but did not listen to the song that was attached to it. The algorithm took it into account, and tracks similar to it will now appear in recommendations. The VK administration regularly updates and improves the work of the site. It is quite possible that the answer to the question of what are the recommendations "in VK, how this tape works, will be different.
Evgeniya Kryukova
We asked the representatives of the VKontakte team burning questions about Prometheus, the smart feed and their plans for the future. Serhiy Paranko, Media Ecosystem Director, and Andrey Zakonov, director of growth and social networking research. Hurry up to familiarize yourself with the unique information from the original source.
1. How do your algorithms understand that a post is interesting to a specific user? Can you list the factors that most affect a record's reach? The quality of the content, the response of the audience is understandable. Anything less obvious?
We analyze the user's activity in the feed and raise those posts with which he is more likely to interact higher: put "Like", write a comment, "stick", open a photo, read an article, watch a video. These actions are very different, we predict them separately. Further, all this is taken into account by the general algorithm, which, based on all factors, carries out the ranking.
2. At your speech at “MEN2»You said that even little things like the weather and the user's mood affect the reach of a post. How does the tape understand this?
There are a lot of ranking signs behind any post. Each component is a certain value that describes the post itself, its author, the time of day at the time of viewing the recommendations, the speed of the reader's Internet connection, and many other criteria.
If a user went to a news feed or a recommendation section while in a deep forest and with a bad Internet connection, we will not show him a bulky video at the beginning of the feed. The algorithm also takes into account the day of the week and the time of day. So in the morning feed of recommendations there will be more news and informational content, and in the evening - media and entertainment.
3. Do you have a document that lists all the factors that the algorithm takes into account? Or is he already so trained that even you do not know by what parameters he gives the coverage of a particular record?
We know the list of factors, but everything is unique for each user-record pair. Therefore, just by looking at the post, it is impossible to say how good it is, this is subjective. A post about local news from Moscow in St. Petersburg is absolutely irrelevant to me, and you may not be interested in my news feed about machine learning and social networks.
4. Is it possible to artificially increase the coverage of a record by purchasing likes and shares on promotion services?
Not worth it. In the long term, this will definitely lead to a decrease in coverage. In addition to the fact that likes and shares do not play a central role in the ranking of a post, such cheats are highly likely to be detected by us.
However, traffic to a post from references from other social networks can really improve the organic reach of the post and there is no penalty for that.
5. Is there an optimal text length for maximum reach in Smart Feed? For example, in an article announcement - no more than 500 characters. And in the longread, created in the editor, no more than 10,000?
There is no optimal length apart from the focus of the reader's attention. Everything is based on engagement, if you have an informational message, try to write succinctly, in an infostyle. If you are trying to involve a person by clicking on a snippet - the same principle, succinctly and engaging. Just don't fall for clickbait, this will negatively affect your reach.
If you're telling a story, get the person involved. It is better to typeset all large forms, conventionally more than 1000 characters in the editor of articles, splitting the text with multimedia inserts.
6. Smart feed gives preference to high-quality videos (over low-quality). And what is “high-quality video” for VKontakte? What characteristics does it have?
Videos 720-1080 r.
7. Does VK try to detect fake news? Is the truthfulness and reliability of the data important, or is it just the audience's interest in the recording?
VKontakte does not have pre-moderation of content - as in other sites, users can freely publish and discuss topics of interest to them. In this case, any person can report unlawful, offensive or unreliable content in his opinion using the "Complain" button. We consider all complaints without exception. VKontakte has one of the largest moderation services and we respond as quickly as possible.
8. How do you feel about contests for the most active subscribers and contests for subscription + repost? They are allowed by the VK rules, but does this somehow affect the coverage of records (positively or negatively)? Do you recommend using contests to promote on VK?
This is a legal way to promote, but you need to keep in mind that this way you attract an irrelevant audience to your subscribers. This can have a negative side effect - followers who are not engaging in your group's main content can have a negative impact on the reach and virality of your posts.
9. Articles created in the editor receive more coverage than articles leading to a third-party site. What should commercial sites and blogs do in such conditions? How can they attract VKontakte users to their site?
A business group in any social network is primarily a community. If you treat it as a media asset, you will succeed. Forced withdrawal of a user to another site, without providing an alternative, in communication can upset a person. As they say: you want to sell me your product, but you do it without respect.
On the VK site, you can communicate with customers / readers, sell goods, receive money transfers and much more. We develop our ecosystem and make it as user-friendly as possible. And in this regard, it is also profitable for business to develop its site within VK. The article editor is another tool for communication and community building.
10. Will wiki articles now receive less coverage than articles made in the editor?
It's not the format that decides, but the user interaction with your content. An interestingly told story with a lot of interactions from readers will receive coverage regardless of the presentation format. Another thing is that keeping the audience's attention is more convenient just with the tools available in the article editor.
Specifically about wiki pages, they do not fall into the "Recommendations" section on mobile devices, this is basically a desktop tool, so yes, their reach is lower than that of other submission formats.
11. Some users complain that they do not see interesting and necessary posts. Why is this happening? And do you have any thoughts about how follow the path of "Facebook" - to make sure that there is more personal content in the users' feed, and not commercial?
We are constantly experimenting with different ranking models, and the main metric is user engagement. You can always send explicit signals to the neuron, hiding a couple of posts from objectionable publics.
12. You say that you cannot buy bots at all. How, then, does the community gain the first audience? If you sign relatives and friends, it turns out that the audience is still not targeted. And if you buy advertising, no one will subscribe to an empty community. How to find a way out?
People come to you for content. Therefore, your community will not be empty if you tell things interesting to the target audience there. Advertising for target audience is an ideal option for a start. This will give you a core audience that will help you grow organically.
13. How do I know if a community has been penalized? How to remove it?
Look at the statistics, you can see this by the decreased coverage. Analyze what you have been doing in recent days, which may be regarded by the platform as unwanted actions: clickbait, aggressive audience withdrawal, unoriginal content, too frequent posting. Stop doing it. And it will go away by itself in two days :-)
14. Are there filters based on topics? (For example, “adult content” goes down). If yes, what topics have decreasing coefficients.
Anything that is not prohibited by the legislation of the Russian Federation and the rules of the site is possible. Anything that is prohibited is prohibited.
15. A case from our practice. We publish unique articles on the blog every day, and they are immediately parsed by several VKontakte publics. It turns out that they are the first to announce this content, and we - after 5-10 minutes. Does this negatively impact the reach of our records?
16. That is, the first person to share the link is considered the primary source? If this is the case, we cannot compete with autoparsers, since it happens automatically for them, and for us manually, we write a unique announcement. What should we do in this case?
Didn't share the link, but natively uploaded the content. Embeds and links, in principle, do not make you the primary source of content in any way. What is there on someone else's video hosting or site, we do not know, and we do not parse or analyze this content. Therefore, it is important to upload your materials to the platform's native filing tools.
17. How will search engines index articles created with the editor? Wiki articles were well-searched, but what will happen now?
Articles are perfectly indexed by Yandex from the very start, and by Google already too.
18. Public LIVE is a VKontakte employee? Can you refer to him as an official source of information?
At one's own risk. This is an unofficial community.
How the Prometheus algorithm works
19. Are living people involved in the work of Prometheus? Or is the decision entirely left to the algorithm? How does he learn? Can you tell us in simple words?
The algorithm is automated and is looking for new authors on its own. The neuron sometimes makes mistakes. We mark each such mistake by sending a signal to the network about the wrong choice. Based on this feedback, "Prometheus" is trained.
20. We've been posting quality content that resonates with our audience for a long time, but we never got the Prometheus fire. How can we influence this? Can you give specific recommendations? I would like to be specific, and not vague advice like "post the content that your audience likes."
First of all, the Prometheus algorithm reacts to the original materials submitted in the native VKontakte tools. Native content delivery formats are extremely important because they allow the algorithm to evaluate the uniqueness of the content and its subject matter. The first is important in order to get the mark, the second is to provide the author with the “fire” with a good coverage boost. Records of such authors fall into the "Recommendations" section.
Almost all records
Probably many of us have noticed the VKontakte tab "Potential Friends", but not everyone knows what it is for and how it works. This is what will be discussed in this article.
Let's take a look at what the tab looks like "Potential Friends", perhaps someone did not notice her.
And how many of those who know about it have guessed how this function works, and by what principle does it define people with whom we may be familiar? Everything is very simple. Let's open this section and study it in more detail. Having done this, you will notice that most of the people who are there are those with whom we communicated, but did not add as friends, or we have mutual friends with them. Now it's a little clearer how this function works, but that's not all.
First, this list is formed based on the people with whom you have mutual friends. Further, this is the whole chain. Those users who have the same city as yours, the same job and other factors are searched for. That is, it is a clever algorithm that constantly updates your list of possible friends. Let's say you added someone as a friend and immediately, from the list of his friends, there will be those who have common friends with you, and they will be offered to you as your possible acquaintances. Here is the whole principle of the section. "Potential Friends".
Of course, accurate and reliable information is impossible to obtain. Only the developers of the VKontakte website know this. It can be assumed that VK collects anonymized data that is tied to an identifier, or buys it from other networks. But this is only a guess and do not be afraid, your personal data is not collected.
Conclusion
We hope you now understand how this function works. With the help of it, you will find your old acquaintances or even get to know people from your city, educational institution.
The social network we are talking about, and which we love so much, so far does not at all claim to be some kind of super-full-fledged music recommendation service. VKontakte is more specialized in other things. Such as convenient user communication, text entries with attachments, interesting games and applications. Nevertheless, the site has long existed a function that will allow you to find exactly the kind of music you prefer.
Unfortunately, not all users know that such a function is provided at all! Some people do not delve into the capabilities of a particular system, site, and just use it at a basic level, so to speak. It's good that there is our website, where they will explain everything in detail to you! Today we will tell you not only how the whole system of searching for similar musical compositions works, but also how to use it correctly. Read and tell your friends about what a VKontakte music recommendation is, as well as how to find similar audio recordings among themselves.
All these songs are collected taking into account not only your VKontakte tracklist, but also taking into account the likes that you put under certain records that contained music. And it doesn't matter which song you are currently playing. Unlike the next function, where this factor is dominant.
Also, VKontakte has a function that shows the user similar audio recordings by genre and performance. How can this function be used? And how does it work? Let's talk about this right now.Just play the song you like the most. Next to the title of this song, you will see a number of buttons. Among them there is a button similar to headphones, which is called "Show Similar". If you click on it, a list of songs will immediately be displayed, which, according to the administration of the VKontakte social network, is very similar to the audio recording you are listening to at the moment. With this method, you can really find a lot of interesting compositions that you will like, and even them. The author of this article himself has used this opportunity more than once. The developers themselves claim that the "Show Similar" function works even better than the recommendation of VKontakte music, which we talked about above.
How can the system find similar songs in real time? At the very beginning, the VKontakte system finds absolutely all of its users who have the same song in the list of audio recordings. The second step is to search for songs that occur most often with this audio recording. It is they who are ultimately displayed. In principle, everything is logical.
Thanks to the functions that are provided today in the VKontakte social network, you can always be in the search mode for new music. After all, with music it is more fun to live, work, and relax!
The benefits of the smart VK feed
According to the official statement of representatives of the social network VKontakte, most users independently switched to the smart feed. Among the users there are those who simply did not turn it off. Also, according to the above statistics, the coverage of records increased by more than 20% on average, and the number of targeted actions increased by 50%.
The old-style ribbon is still available to users. As before, it includes publications from friends and subscriptions in chronological order. When creating a VKontakte account, the smart feed is enabled by default. You can disable it in the "News" section by deactivating the "Interesting first" button.
The VK smart feed algorithm is interesting, first of all, to those users who are subscribed to a large number of groups and pages. Now they do not need to look through the entire chronology of publications in order to find the most interesting news and posts for themselves - it is enough to enable the button "Interesting first", and the algorithm will display the information that corresponds to the interests of the user. By the way, it is this fact that is the reason that the ratio of "Like" marks has increased by 50% since the activation of the smart feed.
It is worth noting that the algorithm collects and analyzes the behavior of users of the social network all the time: it takes into account not only comments and "Like" marks, but also hundreds of other factors, most of which are not disclosed by the VKontakte administration. Among the famous are:
Since everyone's interests are different, the smart feed builds content on an individual basis, and it improves over time thanks to a self-learning algorithm. In this regard, the owners of VKontakte groups and communities need to know how the smart VK feed works and how it can be used to maximize the reach of the target audience and increase loyalty to content.
How the smart feed VKontakte works
For users of the social network, work with the algorithm is limited to pressing the on / off button of the smart feed. The owners and administrators of VKontakte groups, on the contrary, in order to popularize their resource and raise its rating, should take care that the published content, be it photos, articles, audio or video recordings, gets into the ranking and is offered in the feed as the most interesting material.
The main thing that group administrators need to know is how to prepare and design a publication so that it is suitable for getting into the smart VK feed. First of all, you need to pay attention to the content itself. In particular, it must be unique, that is, it must be published for the first time, without any borrowing (even its own, previously published materials will be recognized by the algorithm as copied). It is not known to what extent the uniqueness of the content is taken into account, but the network representatives themselves state that original posts are shown much more often than duplicate material. Among the equally important conditions for the preparation of content are:
- Publication format. Highlighted short and long materials, with the presence of audio and video materials, photographs, links, documents, etc. The bottom line is that a person will be shown mainly what they most often pay attention to. For example, if he most often reacts to various selections of photographs of cats, similar material will always be in priority until the algorithm notices any changes and reacts to them; conversely, if a person prefers long posts, they will be shown more often.
- High visual quality of posts. Photos and images with poor resolution (less than 700px) do not rank or rank poorly in smart feed. Also with the video. This is done to ensure that the user has the most aesthetic in the feed.
- The ability to comment on records. User feedback is a sign of interest. If the comments are closed, the post is much less likely to reach the top positions in the rankings. If the administrator is worried about the presence of obscene language and insults in the comments under the posts, they can be easily suppressed using filters.
- Number of entries per day. This is bad news for those who publish a lot of material every day: the priority of the smart feed is infrequent, but high-quality posts. Conversely, the more activity on the group wall, the less audience coverage will be in the end. The recommended frequency of posting is no more than one post within 3-4 hours.
In addition to complying with these conditions when compiling and publishing content, community administrators should carefully monitor and analyze statistics, observe user reactions to different types of publication formats and build strategies in such a way that the proposed material satisfies not only the preferences of the audience, but also corresponds to the parameters of the algorithm. Together, this will lead to the fact that the news feed of each user will become more interesting and attractive, which means that VKontakte, firstly, will improve its position and will be able to attract much more business investment in advertising; second, it will force authors to take the interests of subscribers more seriously.
How to get into the smart feed on VKontakte
As noted above, a smart feed is an algorithm created specifically to generate a more interesting user news feed. The hierarchy of posts here, in contrast to the standard chronological order, depends on the degree of usefulness of the published material and its ability to interest the user.
In addition, recommended posts sometimes appear in a person's feed, which, according to the algorithm's data, may be of interest. This assumption is made, of course, on the basis of previously collected and analyzed data. At the same time, such posts can be published in a community to which the person is not subscribed. In other words, the algorithm generates a semblance of personal search results within a social network. In this regard, the tasks of group administrators wishing to get into the smart feed should be aimed at improving the quality of content:
- First, the posts should be higher than the posts of other groups.
- Secondly, so that not only subscribers, but also random people interested in a similar topic, can receive the group's notes as recommendations.
Collectively, these actions benefit the community statistics by increasing the reach of views and subscribers.
The main factors of getting into the VK smart feed
Above, we have described the factors that allow you to display group records in the smart tape. However, the fulfillment of these conditions does not give a 100% guarantee of success. The behavior of the target audience also plays an important role here. In particular, this includes the following parameters:
It is worth noting that these factors operate for a limited amount of time: the more a post collects user reactions within a day after publication, the more likely it will be in the smart feed. In other words, a record that is a month or even three days old, even if it suddenly starts to gain popularity, will not qualify for ranking by the algorithm.
Among the typical posts that, according to statistics, are most often found in the smart feed, one can mark posts with text and attached video material (one or more), text with a poll or vote, a selection of pictures and photos in high resolution, etc.
Recently, VKontakte has the ability to write articles using an internal editor. According to popular belief, text materials intended for long-term reading take priority in the ranking of the smart feed. This assumption is due to the fact that large articles are most often found in recommendations, increasing the natural increase not only in the reach of the target audience, but also in the group's subscribers.
Smart feed on Instagram
The introduction of a smart feed in Instagram is due to the fact that the application was bought by the largest social network Facebook, where for many years it has been ranking news feeds by the degree of interest in posts from users. In other words, Instagram now only shows content that gets the most audience engagement. As a result of the innovations, not only the ranking conditions have changed, but also the overall level of subscriber engagement - many popular accounts have noticeably decreased reach.
The principle of forming a smart feed on Instagram is as follows: the algorithm, based on the individual characteristics of the user's account, selects the most interesting material. In particular, it takes into account: user activity (likes, comments), real geographic location, etc. Several users can be subscribed to the same profiles, however, the chronology of displaying posts will be different for each of them: one person will have a photo in priority and rank higher than that of another user.
When posting a photo on Instagram, the algorithm shows it to a small part of followers (about 10-20%). If the audience's response is high, the photo will move to even greater coverage and, having passed the "exam", will move on. Due to this, profiles with winding subscribers (not showing activity) lose popularity and audience coverage.