A/B test in the form of a survey of random people?

Question

Marketing

Hi all.

I have 2 landing pages, and I would like to evaluate which one people like more before launching the project.

Question: Are there similar services where you can order (preferably cheaper of course) a simple survey on different groups of people, which of two or four pictures a person likes more. And based on the result of the study, launch this version.

It only comes to mind to create a form in Google and run an ad on it asking you to take a survey, for some kind of promotional code. But something seems to me there are some services for this, which already have their own audience.

Answer 1

Yandex.Vzglyad, while I found this one, but I would like to hear more options.

Answer 2

There are two main testing methods:
1) A/B testing .
We divide users into two groups and show the new functionality to only one of them. At the same time, both groups must be homogeneous (from the same structure of traffic sources or even distribution, if the experiment is conducted on old users).
2) Version testing .
We release a product update with one new feature, and then we compare the metrics of this version with the metrics of the previous one. (In your case, these will be landing versions)
This approach has its drawbacks. The key is the heterogeneity of traffic and audience. When very different users come to the product and react to new features in completely different ways, it can be difficult to determine the reason for the change in metrics unambiguously.
I also had a case (may be useful to someone) with testing new functionality .
I did not know if it was worth doing at all (the development was quite labor-intensive). Therefore, I hung a pseudo-button of this functionality on the site, when pressed, there was an alert that everything is still under development.
Of course, they measured who, when and how much clicked on it. Realizing that the statistics are quite high (that is, people are interested in such functionality), I sat down to develop and, in principle, did not regret it - it turned out to be a fairly frequently used button later.