Hacker News Reader

Comment by skhr0680 | Hacker News Reader

Twitter is already improved IMHO: The "sign in or sign up" nag screen is gone for me!

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,

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

Interesting commentary on this from someone who used to work at Twitter:

- 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

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?

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.

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.

But that's also reduce ads for unlogged visitors.

Are ads for logged-in users so more profitable that it's worth it?

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).

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.

> 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.

This may be true in the short term but eventually more of these decisions begin to accumulate and cause more dissatisfaction. Each one on its own has barely any impact on the bottom line, it's just one more cut among a hundred cuts. In the end you end up with a product that is overall annoying to use. And the whole time you could be going along, measuring each decision and thinking that you're doing okay and no mistakes are being made. But measuring today with yesterday is not the same as measuring today with 3 weeks ago + 13 decisions reversed, 5 decisions reversed, 1 decision reversed.

Agreed, just following data can be a dangerous path. You have to build things because they’re the right thing to build. You need a cohesive vision not scattershot A-B tests. Unless you only care about short term profit, of course.

Play stupid games, win stupid prizes. That’s why the Facebooks die and the Apples and Googles last.

So what kind of experiment do you run to capture that dissatisfaction? Doesn’t have to be a/b. Just anything qualitative that doesn’t boil down to making decisions purely based on feelings.

Talking to people.

Lol, best of luck.

Generally when running these kinds of tests you measure for further down the funnel metrics than sign ups. It's obvious sign ups would increase. In all the tests I've done around this, you look either at revenue or engagement over a longer term.

This is a technique called "getting the foot in the door".

You see videogames employing these psychological tactics constantly nowadays to completely min-max human behavior into spending money.

I think when you're really good at using a computer, you expect things to be easy. When you get a sign up form instead, it's extremely frustrating. We're used to it being easy to use a computer and can see that someone is making it hard for their own benefit. And then we think, if it's hard for us, it's even harder for someone without a lot of skills.

And I think that's probably true! But the person without much computer skills comes in with an expectation that stuff is gonna be hard. The egregious signup page isn't a surprise in that sense, it's just one of the many things they have to struggle with to use a computer.

An A/B then doesn't show any benefit for removing the signup page, only downside, because an A/B test only tests what happens if one feature changes. Our intuition can still be right; if we could do all the work to make it so a website is actually easy to use for someone without a lot of skills, like how Apple radically simplified the state of affairs with the iPod, that will positively defy peoples' expectations and make the app more accessible -- even if just simplifying any individual component doesn't do that.

I do like the change, i can't even sign up with my email to twitter - most likely because it is a custom domain.

I've had no issues making accounts with my custom domains, fwiw.

Did get a phone-number nag that I was able to ticket myself out of.

Oh he ran A/B 12 years ago. I guess it's settled then.

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.

> 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".

Thank god, I stopped clicking twitter links due to this crap

Reddit and Instagram is even worse. Besides cookie boxes (which all have different UIs and rarely a good no button) it's probably the most annoying thing on an ad-blocked internet.

How do the designers who work at these tech companies not understand that repeatedly annoying non-user visitors doesn't make them want to sign up? Nor incentivize others to link to it?

I think these things come about because of A/B testing. The metric being optimized is probably "app installs" or "account creations" and the negative side effects of driving a bunch of users off are not considered/measured.

Indeed, short term metrics vs long term hatred.

I know how it happens, I've designed many a KPI. Something so fundamental and important as the default non-user first-exerpience (not just a 'signup flow') should be evaluated more holistically than just looking at numbers.

But I guess when you have layers of middle managers it's safer to point to a spreadsheet than make hard choices.

This is why I never touch new reddit or visit the site on mobile. It's old reddit + ublock origin on desktop and a third-party app on mobile every time.

Try using i.reddit.com on mobile, it's a bit more tolerable.

Considering where their revenue comes from, have you considered that they may not want people who run ad-blockers to sign up?

They'll tolerate you if you produce content to help keep the paying users engaged, but plenty of businesses choose not to pursue customers who are unlikely to generate profit.

What do ad blockers have to do with invasive sign up modals? They show up regardless. I think you missed the point.

Agreed. I've noticed instagram and reddit over the last year have really tried to squeeze the last bit of growth it can by blocking non registered visitors from using the site past a couple of clicks. I get that it's their server resources I'm using up but for some reason the way they do it just aggravates me.

Non-technical non-users are not dissuaded, they sign up. Technical non-users using adblock aren't profitable conversions anyway, so "who cares?" is an entirely reasonable decision for the PO/PM who made it.

When a sign-in-to-see-the-thing-you-just-searched-for shows up, I take it as a sign that the company isn't doing very well. If you've ran out of good ideas for driving growth, you start considering the bad ideas.

Do you have numbers for the claim that this behavior doesn’t incentivize sign up? I’d guess that for most folks it indeed does — at some point you get annoyed enough to go through with the sign up process. These sites never log you out so it’s a one time thing. I personally am not inspired to sign up in these situations but I’d guess the numbers show that I’m in the minority.

Even fucking Google keeps putting a half-page pop-up on my iPhone saying "WANT TO SIGN IN, OR JUST KEEP SEARCHING 'ANONYMOUSLY' LIKE A CHUMP?" or something just as fucking annoying.

The Guardian does something similar.

Yea, and the annoying part about Google is that you know that they know you're user X, just not logged in. They know I am logged in to several different accounts from this IP, they know I search for similar stuff under those accounts as I do when not logged in, etc. I feel like it would be more honest for Google to write something like, "Hey, ok_dad, we know it's you, just log in so we can save your porn searches to your account properly, rather than having to do data mining on your nasty habits."

> repeatedly annoying non-user visitors doesn't make them want to sign up?

It absolutely does, and that is precisely why they keep doing it. It doesn't work for the kinds of people that browse hackernews, but that's not their core audience anyway.

It's like saying ads don't work because you and me never click them because we use an adblock. Clearly they do work, and quite well...

https://addons.mozilla.org/en-US/firefox/addon/breakthrough-...

Enjoy :)

I looked at Musk's feed and it did not appear, but just now clicked on one of the other Twitter links in this thread and got hit with it almost immediately. Maybe the change is still being rolled out? Maybe it was just a fluke? Maybe it doesn't appear on certain people's feeds for one reason or another?

Prolly not pushed to your local edge location, my version of twitter updated and it no longer shows the pop up screen yet it still shows the blue bar at the bottom of the page (which is fine imo)

I am still seeing it too, looked at Musk’s feed while being logged out. I am in Europe.

I switched to the nitter client long ago, but that's down today due to a cert error.

https://nitter.net

Other instances here:

https://xnaas.github.io/nitter-instances/

Wow, best change in years.

This is a nice change, I refused to click on any twitter links before. Feels weird scrolling twitter without the terrible log in CTA every action.

That 'feature' never made sense to me. "You always miss the advertising shots you don't take" when you force people to sign up first.

I’m sure they crunched the numbers and found that the gain from targeted ads toward the people that sign up is bigger than the loss of people who don’t use Twitter because of the annoyance. Instagram is even worse for what it’s worth.

It seems inconsistent now. Sometimes I can scroll for minutes with nothing and other times I get the sign-up pop up right away. Or was it always this way?

Its been that way at least 6 months. If you click Login then a close button finally appears in the top left. Garbage piece of shit UX.

it's gotten worse for me. i can't access it. in a browser that's always logged in, i try to visit any page on the site and get a page that says "Something went wrong, but don't fret — let's give it another shot."

the site works whenever i login via a fresh, private browser session. for a minute i thought i was blocked after muting the words "elon" and "musk"

I had that yesterday in Safari - clearing all the caches and closing every Twitter tab (I somehow had 5 open) meant that it worked when I opened a new one. Maybe something Service Worker-y? Have had that issue with the Twitter before.

I still see the annoying sign up/login pop up if I scroll down while viewing someone's twitter feed on the web.

I am still getting it

I’m still getting it…

Maybe OP signed up for Twitter :)

It just appeared for me clicking a link in this post.

It is for me until I scroll down.

Same... doesn't really help if you can't scroll down and read tweets without being prompted to sign in.

Love seeing the confirmation bias from people who expect Musk to improve Twitter on day one.

What? Confirmation bias comes when interpreting some random thing as evidence. This is intention change by him.

As you can see from the many other replies, this change predates his purchase.

I wish.

Doesnt work for me

Reply on news.ycombinator.com