Seller Forums
Sign in
Sign in
imgSign in
imgSign in
user profile
Seller_HoYEgVhxHIIp6

A B split testing - multiple at the same time.

Hi everyone.

Would anyone know the answer to this question please:

I am about to run my first ever split test A/B using Amazon manager your experiment tool (testing the title for my product).

Will I be able to run another experiment in amazon testing main image at the same time?

Or do I have to wait 4-8 weeks to get the results for title and only then run my main image test?

Thank you very much.

Aya

104 views
7 replies
00
Reply
user profile
Seller_HoYEgVhxHIIp6

A B split testing - multiple at the same time.

Hi everyone.

Would anyone know the answer to this question please:

I am about to run my first ever split test A/B using Amazon manager your experiment tool (testing the title for my product).

Will I be able to run another experiment in amazon testing main image at the same time?

Or do I have to wait 4-8 weeks to get the results for title and only then run my main image test?

Thank you very much.

Aya

00
104 views
7 replies
Reply
7 replies
user profile
Seller_ZJhFeE3tNKzfh

unsure but even if you could, I wouldn’t recommend it as you’d have no idea which of the tests was actually working.

Generally with AB testing, you test one thing at a time.

20
user profile
Seller_Y5arOb7nEajZ6

In theory, you could potentially conduct four A/B tests.

However, the issue arises when using four variables, as the A/B test would essentially transform into an AxBxCxD test. This could make it challenging to extract meaningful data to pinpoint the best metrics.

Amazon would typically provides A/B results for 4 or rather (16 (4x4, technically) separate tests, requiring extensive mathematical and manual calculations.

A more practical approach would be to start with an A/B test for images, assess the results after a week or so, and then conclude the test. Subsequently, you can experiment with optimizing your product title.

20
Follow this discussion to be notified about new activity
user profile
Seller_HoYEgVhxHIIp6

A B split testing - multiple at the same time.

Hi everyone.

Would anyone know the answer to this question please:

I am about to run my first ever split test A/B using Amazon manager your experiment tool (testing the title for my product).

Will I be able to run another experiment in amazon testing main image at the same time?

Or do I have to wait 4-8 weeks to get the results for title and only then run my main image test?

Thank you very much.

Aya

104 views
7 replies
00
Reply
user profile
Seller_HoYEgVhxHIIp6

A B split testing - multiple at the same time.

Hi everyone.

Would anyone know the answer to this question please:

I am about to run my first ever split test A/B using Amazon manager your experiment tool (testing the title for my product).

Will I be able to run another experiment in amazon testing main image at the same time?

Or do I have to wait 4-8 weeks to get the results for title and only then run my main image test?

Thank you very much.

Aya

00
104 views
7 replies
Reply
user profile

A B split testing - multiple at the same time.

by Seller_HoYEgVhxHIIp6

Hi everyone.

Would anyone know the answer to this question please:

I am about to run my first ever split test A/B using Amazon manager your experiment tool (testing the title for my product).

Will I be able to run another experiment in amazon testing main image at the same time?

Or do I have to wait 4-8 weeks to get the results for title and only then run my main image test?

Thank you very much.

Aya

Tags:Listings
00
104 views
7 replies
Reply
7 replies
7 replies
Quick filters
Sort by
user profile
Seller_ZJhFeE3tNKzfh

unsure but even if you could, I wouldn’t recommend it as you’d have no idea which of the tests was actually working.

Generally with AB testing, you test one thing at a time.

20
user profile
Seller_Y5arOb7nEajZ6

In theory, you could potentially conduct four A/B tests.

However, the issue arises when using four variables, as the A/B test would essentially transform into an AxBxCxD test. This could make it challenging to extract meaningful data to pinpoint the best metrics.

Amazon would typically provides A/B results for 4 or rather (16 (4x4, technically) separate tests, requiring extensive mathematical and manual calculations.

A more practical approach would be to start with an A/B test for images, assess the results after a week or so, and then conclude the test. Subsequently, you can experiment with optimizing your product title.

20
Follow this discussion to be notified about new activity
user profile
Seller_ZJhFeE3tNKzfh

unsure but even if you could, I wouldn’t recommend it as you’d have no idea which of the tests was actually working.

Generally with AB testing, you test one thing at a time.

20
user profile
Seller_ZJhFeE3tNKzfh

unsure but even if you could, I wouldn’t recommend it as you’d have no idea which of the tests was actually working.

Generally with AB testing, you test one thing at a time.

20
Reply
user profile
Seller_Y5arOb7nEajZ6

In theory, you could potentially conduct four A/B tests.

However, the issue arises when using four variables, as the A/B test would essentially transform into an AxBxCxD test. This could make it challenging to extract meaningful data to pinpoint the best metrics.

Amazon would typically provides A/B results for 4 or rather (16 (4x4, technically) separate tests, requiring extensive mathematical and manual calculations.

A more practical approach would be to start with an A/B test for images, assess the results after a week or so, and then conclude the test. Subsequently, you can experiment with optimizing your product title.

20
user profile
Seller_Y5arOb7nEajZ6

In theory, you could potentially conduct four A/B tests.

However, the issue arises when using four variables, as the A/B test would essentially transform into an AxBxCxD test. This could make it challenging to extract meaningful data to pinpoint the best metrics.

Amazon would typically provides A/B results for 4 or rather (16 (4x4, technically) separate tests, requiring extensive mathematical and manual calculations.

A more practical approach would be to start with an A/B test for images, assess the results after a week or so, and then conclude the test. Subsequently, you can experiment with optimizing your product title.

20
Reply
Follow this discussion to be notified about new activity