Hacker News Reader

Comment by imgabe | Hacker News Reader

I think a lot of it came down to their "invite-only" launch. I remember a lot of my friends were excited to try it. Some of us were able to get invites, some weren't. Eventually the people who couldn't get on right away just said "screw it" and went back to Facebook. The rest of us followed because that's where we could interact with everyone and not just the people who happened to get on Google+.

By the time it opened up to everyone, nobody cared anymore. I can't believe somebody thought it would be a good idea to restrict access to a product that depends on network effects to succeed.

View on HN

Sounds like they were trying to emulate the early days of FB some people pine for where you had to have a .edu email address to sign up. I know a few friend were hopeful that Google+ meant they would no longer see posts from their racist uncle or gullible aunt. Start out with that same kind of exclusivity and hopefully it builds this exclusive mystique so when you open it up to everybody you get a flood. Perhaps the big difference they didn't factor in is FB did this when relatively few people were on social media so they couldn't help but grow whereas Google+ needed to recognize they were trying to enter an already saturated market.

> Sounds like they were trying to emulate the early days of FB some people pine for where you had to have a .edu email address to sign up.

It wasn't just that. At the beginning, you had to be at an Ivy League. And that was the problem, G+ was invite only among a bunch of nerds, whereas FB was invite only among people who network for a living.

I don't think the nerd/Ivy distinction you're making is significant. Most students at Ivy League schools are nerds, at least by the standards of broader society (you need good grades to get in). Few of them "network for a living."

If anything, the real distinction is that a network for college students is appealing to teenagers, who are the perfect audience for social media. A network for adults doesn't draw in this crowd.

Facebook had what turned out to be an accidentally brilliant way of bootstrapping networks and trust.

Fundamental problem: social networks with no-one on them are not fun and not sticky.

Fundamental problem 2: if anyone can read what you're writing, it's not a social network, it's a blog. So you can't just let anyone see what people are writing.

The way FB handled that in the early days was that by default, people in the same university network could see your status etc. I don't remember exactly what they could and couldn't do but the point is that they created a semi-privileged circle of people with some access based on a pre-made group that people would be familiar with - fellow students at the same university.

That gave people a network from day 1 in what felt like a controlled way. Also, as noted, they started with a desirable closed group that people wanted to belong to.

Crucially, they had a strategy that worked for that time. Same strategy applied today would simply fail.

Facebook started out as a social network where you could get what we today regard as PII (e.g phone numbers of the opposite gender and such) for Harvard students, and its engagement numbers were really good (based on Zuck’s previous experience with the Harvard-only FaceMash).

They would later expand to other Ivy league schools — aka networks — one by one, rather than simultaneously, as a way to slow their growth.

I imagine an unintended side effect of this growth strategy was the sense of FOMO that it engendered among university students that had heard about but couldn’t join the exclusive/elusive Facebook, unlike the other social networks like Friendster, Hi5 etc.

For each network on their expansion list, Zuck & team would furiously scrape the student directory to seed the new network with plausible student info.

In the early days, you could only friend people at your school because that's who was colocated on your MySQL shard. Fully general addressing of sharded objects was added quite a bit later, about when they opened to non-.edu addresses.

Problem is even though Facebook was restricted to a very small set of people, each of them could interact with most/all of the people they wanted to (other college students). Same was not true for Google+, people didn't have enough invites to give to all of their friends.

It's an exclusivity play, to pump up demand. People like things that make them feel exclusive, so for launch someone at Google tried to increase demand by making G+ more exclusive with invite-only.

Unfortunately this increased the barrier to entry at launch too, which for social networks whose value is tied to users generating media is a cardinal sin.

I think they did that because of how well it worked for GMail. And Facebook was restricted too in the beginning. Obviously, Google+ didn't benefit from following the precedents set by other successful launches.

GMail's time was different, there was a lot less competition among web services back then.

At the time Gmail launched there were a bunch of mail services.

GMail was different in a number of ways.

1. Exponentially more space in a market where single or double digit megabytes was the norm. GMail's original pitch was that you never had to delete an email, and that you could always find them with search.

2. It was one of the first web apps. It was dynamic in a way few sites were then. Asynchronous JavaScript (Ajax) and Web 2.0 were still new concepts.

No, the difference with Gmail is that email doesn't require a network effect to be useful. If I'm the only one on G+, it is pointless. If I'm the only one on Gmail, it's just as useful as if everyone I know is on Gmail.

Gmail's invite only approach made sense because in 2004 it was absurd for anybody to offer 1GB of free storage. Hotmail offered 200MB I think.

Compared to that, nobody cared about a Facebook ripoff with a different UI and none of the network effects.

I think it was way less than that. Wasn't it like ... 2MB?

Yeah, the change to 100 or 200mb came as a desperate attempt later to stop the gmail hemorrhage.

I kind of remember that was something like 10 or 20MB, but yes, was about 2 order of magnitude less.

No. It was 1 GB:

https://en.wikipedia.org/wiki/History_of_Gmail

"after considering alternatives such as 100 MB, the company finally settled upon 1 GB of space, a figure that was preposterous compared to the 2 to 4 MB that was the standard at the time.[3]"

"On 1 April 2005, exactly one year after the initial release, Gmail increased the mailbox size to 2 GB, advertising it as 2GB plus"

The sample of the news from these times:

https://betanews.com/2005/04/01/gmail-storage-to-reach-2gb-a...

I referred to what Hotmail had at the time.

True. My grandma was in fb early.. She never even knew what Google+ was. Its only a social network if there are ppl on.

Gmail could send and receive emails outside gmail.

FB wasn't restricted within your likely friends group (all students at particular schools were made eligible at the same time).

The main reason Google launches popular products by invitation only is to manage server load. Otherwise you get situations like the Pokemon Go launch, where the service was unreliable for the first few months.

Pokémon Go was mostly fine. I didn’t play it beyond a few months. Nor did most people I know.

Pokemon Go was an absolute nightmare at least for a few months.

I'd love to read some articles on what they had to do to fix it.

Unreliable because too many people want to use it is the goal :)

This is a frequently-repeated view, but I find it completely implausible.

A social network's initial frowth phase is completely predicated on it being a space people want to join, and that is accomplished by its delivering value and appeal.

That might be compelling content, compelling connections, or a mix of the two. A cultivated and exclusive club, as with the Haevard Facebook, or early Usenet, has appeal. It's the opening up to the world which kills that initial growth appeal.

Google+ would have done far better to extend the beta period, possibly by years, and shake out its fundamental architectural flaws.

> I find it completely implausible.

I disagree and think it's 100% the reason it failed.

> A social network's initial frowth phase is completely predicated on it being a space people want to join, and that is accomplished by its delivering value and appeal.

But people did want to join Google+, but they couldn't. And since a social network is only useful if all your friends are on it, people never got into the habit of using it. Why use a social network without your friends? It's not like GMail where it's still useful even if your friends aren't on it.

I can tell you personally that I was excited to get the invite to Google+, only to lose the excitement when only 3 of my friends were on it. By the time Google+ became available to everyone, it was too late. The excitement was gone.

> A cultivated and exclusive club, as with the Haevard Facebook, or early Usenet, has appeal.

The exclusivity of Harvard Facebook succeeded because social media was still in its infancy at the time. Most people didn't have access to the WWW in their pocket yet. Early Usenet had appeal because the Internet itself was still only being used by techies and college students.

> The excitement was gone.

Most people didn't hear about it. You can ask people on the street. They sure know facebook. Excitement is not a good way to motivate people to move platforms, in general (despite the niche cases like Fyre Festival which are relatively small). Maybe they should have gotten some celebrity support? Otherwise the excitement wasn't mainstream. G+ needed mainstream users.

> But people did want to join Google+, but they couldn't

That doesnt explain the lack of participation a year later. Nobody wanted G+ because they didn't need it. Even if it gave some benefit (which it couldn't) it didn't join with their existing Social Media data making a higher barrier.

> Most people didn't hear about it.

Lots of people did. But then they couldn't get an account. https://trends.google.com/trends/explore?date=2010-05-20%202...

I'd argue the problem was more one of the marketing / promotion than the closed beta.

The second spike in the trends graph corresponds with the open access public announcement. Interest spiked, but did not sustain.

It's the lack of sustain that hurt G+ in the end. And much of that was due to a very raw early release -- no search, little content, and very muddled concepts, many never really fixed.

The thing with Harvard Facebook is that it's a predefined social circle that came online as a unit. If you go to Harvard, most of your friends and people you regularly interact with are also going to Harvard, so you have a cohesive group altogether.

With Google+, invites were more or less random, so in a group of 5 friends maybe 3 got one and 2 didn't. Were the 3 people who got invites supposed to just drop their other 2 friends? Obviously people aren't going to do that.

People do like things that are exclusive, but they want to exclude outsiders, not their friends.

The early G+ was similar in regards to early Slashdot or Reddit: Tech / SV insiders, Googlers and Xooglers and their extended notworks.

Expanding beyond this, especially to Google's other natural affinity group, advertisers and marketers, did far more harm than good to the service IMO.

The actual invite-only period only lasted a few months -- it was general-access by late September 2011:

https://www.cnn.com/2011/09/20/tech/web/google-plus-public/i...

Facebook's closed period lasted years.

Reply on news.ycombinator.com