5 ms·

https://www.theverge.com/2022/10/30/23430008/elon-musk-twitter-homepage-subscriptions-changes https://www.theverge.com/2022/10/30/23430008/elon-musk-twitt... >

https://www.theverge.com/2022/10/30/23430008/elon-musk-twitter-homepage-subscriptions-changes https://www.theverge.com/2022/10/30/23430008/elon-musk-twitt... >He requested that logged out users visiting Twitter.com be redirected to the Explore page that shows trending tweets and news stories, according to employees familiar with the matter who requested anonymity to speak without the company’s permission. Before, visiting Twitter’s homepage while logged out showed only a sign-up form,

whiplash451 4y ago
That change is welcome. The previous setup was really obnoxious.
simonw 4y ago
Interesting commentary on this from someone who used to work at Twitter: - https://twitter.com/joshelman/status/1586812933043933184 https://twitter.com/joshelman/status/1586812933043933184 > Changing the logged out home page of Twitter to show more content is likely one of those “easy fixes” that has symbolic effect but very little to no metrics effect. I ran A/B tests on this in 2010. Usually the best way to grow Twitter active users is make people sign up first
- shwoopdiwoop 4y ago
  If it has no effect on metrics but removes as major annoyance for users, why would you not have done that in the first place?
- cercatrova 4y ago
  Technical people, especially engineers and those on HN, hate sign up forms but having grown a product to X thousand users, the more you push the sign up form, the more people sign up. You can even plot a linear relationship between the two, which I've done before.
  - joegahona 4y ago
    I noticed the same thing in the world of publishing (both print and online). It's the reasons pop-ups have never gone away -- they work. It's the reason those annoying inserts in print magazines that fall in your lap have never gone away -- they work.
  - conradfr 4y ago
    But that's also reduce ads for unlogged visitors. Are ads for logged-in users so more profitable that it's worth it?
  - Reimersholme 4y ago
  - AlexandrB 4y ago
    This seems shortsighted. It's optimizing for what's easy to measure (sign ups) over what's hard to measure (brand, customer satisfaction, word of mouth).
    - cercatrova 4y ago
      Sign ups were directly correlated to revenue, at least in my case, and likely most others, such as in @joegahona's comment about publishing. That is to say, the customer satisfaction decrease in popups did not decrease revenue, and that is assuming there even was a satisfaction decrease at all; again, most technical users vastly overestimate how many people actually care about popups, most non-tech people simply click the X to make it go away, and a percentage of all visitors will sign up.
      - gunapologist99 4y ago
        > most non-tech people simply click the X to make it go away, and a percentage of all visitors will sign up. This is likely true, but only if there's an X. Facebook's growing sign-in box as you scroll down is not easily clicked away from.
      - Nathanba 4y ago
- warinukraine 4y ago
  Oh he ran A/B 12 years ago. I guess it's settled then.
- tootie 4y ago
  Yeah this is unlikely to be positive on revenue and very well might be negative. It's extremely unlikely that Twitter put that registration wall in place without extensive testing and micro-optimization. Musk is just throwing his weight around and declaring himself smarter than everyone else.
- grecy 4y ago
  > Usually the best way to grow Twitter active users is make people sign up first Call me crazy, but maybe a site should be designed so people actually enjoy using it, rather than always optimizing for "growth".

whiplash451 4y ago

That change is welcome. The previous setup was really obnoxious.

simonw 4y ago

Interesting commentary on this from someone who used to work at Twitter: - https://twitter.com/joshelman/status/1586812933043933184 https://twitter.com/joshelman/status/1586812933043933184 > Changing the logged out home page of Twitter to show more content is likely one of those “easy fixes” that has symbolic effect but very little to no metrics effect. I ran A/B tests on this in 2010. Usually the best way to grow Twitter active users is make people sign up first

shwoopdiwoop 4y ago
If it has no effect on metrics but removes as major annoyance for users, why would you not have done that in the first place?
cercatrova 4y ago
Technical people, especially engineers and those on HN, hate sign up forms but having grown a product to X thousand users, the more you push the sign up form, the more people sign up. You can even plot a linear relationship between the two, which I've done before.
- joegahona 4y ago
  I noticed the same thing in the world of publishing (both print and online). It's the reasons pop-ups have never gone away -- they work. It's the reason those annoying inserts in print magazines that fall in your lap have never gone away -- they work.
- conradfr 4y ago
  But that's also reduce ads for unlogged visitors. Are ads for logged-in users so more profitable that it's worth it?
- Reimersholme 4y ago
- AlexandrB 4y ago
  This seems shortsighted. It's optimizing for what's easy to measure (sign ups) over what's hard to measure (brand, customer satisfaction, word of mouth).
  - cercatrova 4y ago
    Sign ups were directly correlated to revenue, at least in my case, and likely most others, such as in @joegahona's comment about publishing. That is to say, the customer satisfaction decrease in popups did not decrease revenue, and that is assuming there even was a satisfaction decrease at all; again, most technical users vastly overestimate how many people actually care about popups, most non-tech people simply click the X to make it go away, and a percentage of all visitors will sign up.
    - gunapologist99 4y ago
      > most non-tech people simply click the X to make it go away, and a percentage of all visitors will sign up. This is likely true, but only if there's an X. Facebook's growing sign-in box as you scroll down is not easily clicked away from.
    - Nathanba 4y ago
warinukraine 4y ago
Oh he ran A/B 12 years ago. I guess it's settled then.
tootie 4y ago
Yeah this is unlikely to be positive on revenue and very well might be negative. It's extremely unlikely that Twitter put that registration wall in place without extensive testing and micro-optimization. Musk is just throwing his weight around and declaring himself smarter than everyone else.
grecy 4y ago
> Usually the best way to grow Twitter active users is make people sign up first Call me crazy, but maybe a site should be designed so people actually enjoy using it, rather than always optimizing for "growth".