The secret to successful A / B testing - 5 questions to yourself
Abraham Lincoln once said: "Give me 6 hours to cut down a tree, and I will spend the first 4 hours on sharpening an ax." In this quote, there is a great analogy with A / B testing. If you want to break the bank during the A / B test, you will have to prepare very well. And exactly how to do this, you will learn in our article.
Michael Aagaard (one of the most famous experts in the field of conversion optimization and copywriting), decided to make a list of the most serious errors in A / B testing. Based on his many years of experience, he came to the conclusion that his main mistake was to conduct too many experiments.
')
It sounds very strange. What, then, should an expert concentrate on in A / B testing , if not on conducting experiments? The answer is inside the article.
Let's start with a short story. Working on increasing the conversion of one site, Michael launched an experiment in which the test version of the landing page was very different from the original. As a result of the experiment, the conversion increased so much that the client earned an additional amount with five zeros in just two weeks after the implementation of the test version on the site.
This result was not achieved due to the large number of experiments or ideas for testing . No, it has nothing to do with success at all.
It was possible to conduct a successful A / B test due to thorough preparation. Most of the time was spent on:
Only after collecting all the information did Michael start formulating ideas for increasing conversion and sales.
Surely, many of you think that A / B testing is the simplest way to increase conversion. And you are mistaken. Really successful experiments that allow you to increase sales and make a good big score with five zeros are based on what is listed above.
That is why, if Abraham Lincoln was an internet marketer of the 21st century, his quote would have something like this: “Give me 6 hours to increase the conversion of the landing page, and 4 of them I will spend on analyzing information”
Doing A / B testing without web analytics, understanding users is the same as trying to cut a tree with a blunt ax.
The success of A / B testing depends directly on how good the hypothesis is for the experiment.
It is to this conclusion that Michael came after running hundreds of tests, most of which were unsuccessful. And all just because he always chased the quantity, not the quality. And in cases where emphasis was placed on the preparation, the experiments were most often successful.
So why don't you learn this lesson. Or is it better to step on the same rake?
Most A / B experiments fail either because of a weak hypothesis for testing, or because of its absence. No matter how funny it sounds ...
The article in the topic - the 10 most common errors in A / B testing
The A / B testing hypothesis includes information about what you want to change on the page and how this change can affect the final result.
To simplify the process of generating good ideas for testing, we suggest that you use special techniques that Michael uses in his work.
This technique is very simple. Its essence is getting the right answers to the right questions in the right order. Who! Where? Why? What? How?
It all starts and ends with an understanding of the target audience. You must know your potential customer.
The question for which you need to get an answer: "Who visits my site (the page on which you are going to conduct an A / B test)?"
You need to know how a potential customer hit the landing page. He saw ads on another site? Did he switch to it after receiving your newsletter? Or through a search in Google? In each case, the user's behavior will be different, so you should clearly know how a potential client hits the landing page.
The question to which you need to get an answer: "Where was the potential client before he went to my site?"
You need to understand the problems and motivations of your target audience. If you know, based on what factors a potential client makes his decision, you can get it.
Questions to get answers to:
Now you know who the potential client is, where he came from and for what reason, what factors influence his decision. At this stage it is necessary to determine the content that should be on the landing page.
Questions to get answers to:
After you have received answers to all previous questions, you can go to the most important thing.
Final question: “How should I change the page to increase the conversion?”
Having received answers to all previous questions, you can easily answer the last one. And, based on this answer, you can formulate the idea of ​​a successful A / B experiment.
Only 5 questions. Some are lighter. Others are harder. To get some answers, you will have to try and delve into web analytics, user analysis, polls, etc. It all depends on the information already available. But this is precisely the most important advantage of this technique.
If you cannot answer any of the above questions, then you do not fully understand your target audience.
If you do not fully understand your target audience, you cannot offer exactly what you need.
If you cannot do this, then it will be difficult to create a really good A / B experiment that will allow an increase in conversion and sales.
Therefore, just try using this technique in your work. I am sure that you will be satisfied. At a minimum, this will give you a better understanding of your target audience. And in the long run, this is never over.
By the way, it would be very interesting to learn how you generate ideas for A / B testing? Leave your comments on this.
Translation and adaptation: http://unbounce.com/ab-testing/ab-testing-sharpening-the-ax/
Prepared by the blog Boosta.ru and the project Changeagain.me
Read also our other popular articles:
1. They will help you find all the “killers” of the conversion - 10 reports in Google Analytics (5000+ views)
2. 100 ideas for A / B testing. Part One (2000+ views)
3. 100 ideas for A / B testing: Part Two (1000+ views)
4. How to find an idea for an A / B test: heat maps and polls (1000+ views)
Michael Aagaard (one of the most famous experts in the field of conversion optimization and copywriting), decided to make a list of the most serious errors in A / B testing. Based on his many years of experience, he came to the conclusion that his main mistake was to conduct too many experiments.
')
It sounds very strange. What, then, should an expert concentrate on in A / B testing , if not on conducting experiments? The answer is inside the article.
The story of one successful A / B test
Let's start with a short story. Working on increasing the conversion of one site, Michael launched an experiment in which the test version of the landing page was very different from the original. As a result of the experiment, the conversion increased so much that the client earned an additional amount with five zeros in just two weeks after the implementation of the test version on the site.
This result was not achieved due to the large number of experiments or ideas for testing . No, it has nothing to do with success at all.
It was possible to conduct a successful A / B test due to thorough preparation. Most of the time was spent on:
- web analytics
- user surveys
- sales funnel analysis
- heat map analysis
- user behavior analysis
- view recorded sessions
- survey of company employees who are engaged in sales and customer service.
Only after collecting all the information did Michael start formulating ideas for increasing conversion and sales.
Surely, many of you think that A / B testing is the simplest way to increase conversion. And you are mistaken. Really successful experiments that allow you to increase sales and make a good big score with five zeros are based on what is listed above.
That is why, if Abraham Lincoln was an internet marketer of the 21st century, his quote would have something like this: “Give me 6 hours to increase the conversion of the landing page, and 4 of them I will spend on analyzing information”
Doing A / B testing without web analytics, understanding users is the same as trying to cut a tree with a blunt ax.
The success of A / B testing depends directly on how good the hypothesis is for the experiment.
It is to this conclusion that Michael came after running hundreds of tests, most of which were unsuccessful. And all just because he always chased the quantity, not the quality. And in cases where emphasis was placed on the preparation, the experiments were most often successful.
So why don't you learn this lesson. Or is it better to step on the same rake?
Most A / B experiments fail either because of a weak hypothesis for testing, or because of its absence. No matter how funny it sounds ...
The article in the topic - the 10 most common errors in A / B testing
The A / B testing hypothesis includes information about what you want to change on the page and how this change can affect the final result.
To simplify the process of generating good ideas for testing, we suggest that you use special techniques that Michael uses in his work.
5 questions to help understand the target audience and run a successful A / B test
This technique is very simple. Its essence is getting the right answers to the right questions in the right order. Who! Where? Why? What? How?
It all starts and ends with an understanding of the target audience. You must know your potential customer.
The question for which you need to get an answer: "Who visits my site (the page on which you are going to conduct an A / B test)?"
You need to know how a potential customer hit the landing page. He saw ads on another site? Did he switch to it after receiving your newsletter? Or through a search in Google? In each case, the user's behavior will be different, so you should clearly know how a potential client hits the landing page.
The question to which you need to get an answer: "Where was the potential client before he went to my site?"
You need to understand the problems and motivations of your target audience. If you know, based on what factors a potential client makes his decision, you can get it.
Questions to get answers to:
- Why did he visit my site?
- Why does he need what I offer?
- Why does he agree to pay for it?
- Why can he refuse?
Now you know who the potential client is, where he came from and for what reason, what factors influence his decision. At this stage it is necessary to determine the content that should be on the landing page.
Questions to get answers to:
- What should a potential customer know to make a purchase?
- What should I focus on to convey the value of my proposal to a potential client?
- What should I focus on in order to overcome all objections (to answer all the questions of concern) of a potential client?
- What should happen after he agrees to pay for my offer?
- What is missing on this landing page and what should I add to it?
- What should be removed from the page?
After you have received answers to all previous questions, you can go to the most important thing.
Final question: “How should I change the page to increase the conversion?”
Having received answers to all previous questions, you can easily answer the last one. And, based on this answer, you can formulate the idea of ​​a successful A / B experiment.
Only 5 questions. Some are lighter. Others are harder. To get some answers, you will have to try and delve into web analytics, user analysis, polls, etc. It all depends on the information already available. But this is precisely the most important advantage of this technique.
If you cannot answer any of the above questions, then you do not fully understand your target audience.
If you do not fully understand your target audience, you cannot offer exactly what you need.
If you cannot do this, then it will be difficult to create a really good A / B experiment that will allow an increase in conversion and sales.
Therefore, just try using this technique in your work. I am sure that you will be satisfied. At a minimum, this will give you a better understanding of your target audience. And in the long run, this is never over.
By the way, it would be very interesting to learn how you generate ideas for A / B testing? Leave your comments on this.
Translation and adaptation: http://unbounce.com/ab-testing/ab-testing-sharpening-the-ax/
Prepared by the blog Boosta.ru and the project Changeagain.me
Read also our other popular articles:
1. They will help you find all the “killers” of the conversion - 10 reports in Google Analytics (5000+ views)
2. 100 ideas for A / B testing. Part One (2000+ views)
3. 100 ideas for A / B testing: Part Two (1000+ views)
4. How to find an idea for an A / B test: heat maps and polls (1000+ views)
Source: https://habr.com/ru/post/295234/
All Articles