A/B testing for titles/blurbs

Reisinling

Well-known member
Joined
Feb 5, 2021
Messages
357
Points
63
We want more readers. @Tony wants us to have more readers. You know what we could do to get more readers?

Perform A/B testing with blurbs/covers/titles

What is A/B testing? Let's take an example of title. Page would split users into two groups (they wouldn't know which group they are). You could set up 2 versions of the title, and then, the site would show some people version A of the title and some version B on the title. Then, it could tell you which version had more engagement.
 

Zirrboy

Fueled by anger
Joined
Jan 25, 2021
Messages
1,135
Points
153
Nice for authors perhaps, but shit for readers.

A: "You read xxx?"
B: "Read what now?"
The same goes for content, if you want to include that.

And once you've determined the better one?
Force one group to readjust to whatever you have decided on?
If you allow the same for content, they'd need to straight up re-read the "better" version.

Sure, at the end you're left with the more popular, but at the cost of inconveniencing half of your readership up until that point.
 

Reisinling

Well-known member
Joined
Feb 5, 2021
Messages
357
Points
63
Nice for authors perhaps, but shit for readers.

A: "You read xxx?"
B: "Read what now?"
The same goes for content, if you want to include that.

And once you've determined the better one?
Force one group to readjust to whatever you have decided on?
If you allow the same for content, they'd need to straight up re-read the "better" version.

Sure, at the end you're left with the more popular, but at the cost of inconveniencing half of your readership up until that point.
That's why I'm only suggesting the A/B testing for marketing stuffs. And you can already change the title, and many people do (including myself)
 

Agentt

Thighs
Joined
Oct 8, 2020
Messages
3,384
Points
183
So, its basically like volunteering proofreaders?
 

Ddraig

<First Dragon of SHF> <Pokemon Goddess of NuF>
Joined
Dec 23, 2018
Messages
558
Points
133
It should be alright with covers - maybe even summaries, not too sold on titles.
 

Piisfun

Playful Spacetime Dragon
Joined
Jun 2, 2020
Messages
211
Points
83
Honestly, I can't see this working. That isn't because of system limitations; rather the reader base is just too small for it to be practical. It isn't uncommon for a chapter on a story to not get a single comment, and even polls often get less than 50 votes. the value of the data would end up being marginal at best.
 

Reisinling

Well-known member
Joined
Feb 5, 2021
Messages
357
Points
63
Honestly, I can't see this working. That isn't because of system limitations; rather the reader base is just too small for it to be practical. It isn't uncommon for a chapter on a story to not get a single comment, and even polls often get less than 50 votes. the value of the data would end up being marginal at best.
It would be useful mostly to the "power users" so to speak. But this data would work with far more users then these that take part in polls, or interact with the story in any way.
 
Top