October 24, 2014 at 18:10Something better than A / B testing
Hello! My name is Vasiliy. My main occupation is to increase website conversion, a key metric in any IT business.
According to statistics, the average conversion of sites in RuNet does not exceed 1 percent and this means that a huge number of Internet entrepreneurs lose a lot of money every day. These entrepreneurs can be mentally divided into two main types: the first are those who just sit and look at their non-optimized resource and do nothing. Unfortunately, they are doomed to be crushed by time by innovations and competitors. The second is those who are trying to do something about it. As a rule, for this they draw landing pages, experiment with advertising channels and conduct A / B testing. A / B testing is an essential tool for modern marketing, it allows you to determine which version of the selling page is the most effective for most users. To understandObama and $ 60 million . However, this method has a flip side, it is that all people are different, and if, for example, statistics show that women mostly love cats , it is not necessary that your girlfriend is not allergic to them. That is, if our A / B test showed that a certain version of the landing is ideal for 60% of users and we choose it for our advertising company, then we automatically lose the remaining 40%, which it does not catch on anything. So, A / B testing is not perfect and you need to urgently do something about it. And right now, because to hell with all the hell we lose 40% of leads and this is not cool at all!
In order not to lose customers, we need a system that not only conducts an A / B test and ex-post shows you the results, but is able to predict which version of the landing page your visitor will like and show him this particular option. It’s a pity that this is not possible, because no one knows until the end what your potential customers have in mind, is that right? Or not?
Although, of course, we can assume that a certain number of your visitors do not watch the presentation video during working hours, because their bosses are on fire, and if your grandmother visited your site, then most likely the selling text that was written for your son. Immediately a huge number of matches climb into my head. But what if you find these correspondences automatically, if you create an algorithm that finds similarity in the behavior of different users on the site for different parameters, for example, time, geolocation, browser, operating system, UTM tag from advertising companies? How many matches would he find?
I’ll say right away that we did this thing and got an interesting result, it turned out that with a probability of about 80%, the algorithm can predict which version of the landing a particular user will like more. Yes, we do not have data what your user ate for breakfast, also, we don’t know what is his dog or wife’s name. But even the available data was already enough to automatically look for matches. To make our algorithm work better, we attached a so-called weight system to it. If the landing option is not pleasant to users and they do not perform targeted actions on it, it slowly sinks in rotation and as a result does not appear at all to users or is shown only to the group that they like, only at a time when they like it more. Again, there is a small problem, it is that in order to identify conformities and normal operation of the system, it takes some time and corresponding attendance. And the more landing options for rotation, the more visitors are needed. For each option, at least 500 is desirable. Unfortunately, if you have an attendance of 200 people per day and 5 landing options, you are more likely to go nuts than wait until our algorithm segments your users. However, if you have equal attendance at 2000+, he will learn in one and a half days. And if you have this traffic from ads on which utm tags are installed, then the algorithm will do the magic for you. True, do not forget that even a magic genie does not fulfill his desire, if you do not rub his lamp, to put it simply, if you have made 5 variants of creepy non-usable pages that differ from each other only in the number of unicorns in the footer, do not wait,
In general, the thing turned out to be interesting and it seems to me that this looks like the future of marketing technologies.
I apologize in advance if I missed any errors in the text of the article, if you find it please write in PM.
PS I am the only dysgraphic marketer in this country.
According to statistics, the average conversion of sites in RuNet does not exceed 1 percent and this means that a huge number of Internet entrepreneurs lose a lot of money every day. These entrepreneurs can be mentally divided into two main types: the first are those who just sit and look at their non-optimized resource and do nothing. Unfortunately, they are doomed to be crushed by time by innovations and competitors. The second is those who are trying to do something about it. As a rule, for this they draw landing pages, experiment with advertising channels and conduct A / B testing. A / B testing is an essential tool for modern marketing, it allows you to determine which version of the selling page is the most effective for most users. To understandObama and $ 60 million . However, this method has a flip side, it is that all people are different, and if, for example, statistics show that women mostly love cats , it is not necessary that your girlfriend is not allergic to them. That is, if our A / B test showed that a certain version of the landing is ideal for 60% of users and we choose it for our advertising company, then we automatically lose the remaining 40%, which it does not catch on anything. So, A / B testing is not perfect and you need to urgently do something about it. And right now, because to hell with all the hell we lose 40% of leads and this is not cool at all!
In order not to lose customers, we need a system that not only conducts an A / B test and ex-post shows you the results, but is able to predict which version of the landing page your visitor will like and show him this particular option. It’s a pity that this is not possible, because no one knows until the end what your potential customers have in mind, is that right? Or not?
Although, of course, we can assume that a certain number of your visitors do not watch the presentation video during working hours, because their bosses are on fire, and if your grandmother visited your site, then most likely the selling text that was written for your son. Immediately a huge number of matches climb into my head. But what if you find these correspondences automatically, if you create an algorithm that finds similarity in the behavior of different users on the site for different parameters, for example, time, geolocation, browser, operating system, UTM tag from advertising companies? How many matches would he find?
I’ll say right away that we did this thing and got an interesting result, it turned out that with a probability of about 80%, the algorithm can predict which version of the landing a particular user will like more. Yes, we do not have data what your user ate for breakfast, also, we don’t know what is his dog or wife’s name. But even the available data was already enough to automatically look for matches. To make our algorithm work better, we attached a so-called weight system to it. If the landing option is not pleasant to users and they do not perform targeted actions on it, it slowly sinks in rotation and as a result does not appear at all to users or is shown only to the group that they like, only at a time when they like it more. Again, there is a small problem, it is that in order to identify conformities and normal operation of the system, it takes some time and corresponding attendance. And the more landing options for rotation, the more visitors are needed. For each option, at least 500 is desirable. Unfortunately, if you have an attendance of 200 people per day and 5 landing options, you are more likely to go nuts than wait until our algorithm segments your users. However, if you have equal attendance at 2000+, he will learn in one and a half days. And if you have this traffic from ads on which utm tags are installed, then the algorithm will do the magic for you. True, do not forget that even a magic genie does not fulfill his desire, if you do not rub his lamp, to put it simply, if you have made 5 variants of creepy non-usable pages that differ from each other only in the number of unicorns in the footer, do not wait,
In general, the thing turned out to be interesting and it seems to me that this looks like the future of marketing technologies.
I apologize in advance if I missed any errors in the text of the article, if you find it please write in PM.
PS I am the only dysgraphic marketer in this country.