- They lost $3 Billion in 9 months.

- For every $6 they spent, they got back $1 in revenue.

- The product is an app which is free to use.

- A large chunk of their system runs on Google App Engine; which is probably the most expensive and highest lock-in infrastructure solution that you could possibly use.

- The CEO, who was basically fresh out of university, turned down a $4 Billion offer buy the company.

So basically; they're a company which loses money at an incredible rate on a product which is nearly impossible to monetise and with an inexperienced and irrational person as a CEO. I'm surprised it still exists tbh.

No sane rational person can look at those numbers and think that it's a good investment.

> most expensive

Not really. Like any hosting platform, if you know how to tune it properly costs can be kept low. We cut our infra costs in half moving to GAE. YMMV.

They're already in Google Cloud. One could argue that's preventing them from keeping costs low. Google Cloud is tremendously expensive in op-ex at Snap's scale. You're paying Google instead of having a good ops team(s).

Dropbox did the math and found they could do better on their own. Only time will tell who is right.

I use Google Cloud. Love it's simplicity and design.

It's definitely cheaper on the lower to even medium end. But once you're spending six figures per month, it's probably only ~10% cheaper than buying the equipment and hiring three people to manage it. Not sure why anyone would be on Google Cloud (or any cloud really) and spend 7 figures a month, unless you got a sweet deal from them.

This is the first I've heard that Google Cloud is cheaper than doing it yourself. All of the conversations I've had (and some pricing I've seen) say otherwise. Do they actually give 85% discounts, because that's what it would take to make Google Cloud equal to the cost of running my business.

It's not for everyone, but I don't think you're including the cost of employing someone to manage the hardware. If you're a scrappy startup that is willing to do whatever it takes to keep costs down, or a giant company that already has an in-house hardware team, then it's more expensive.

Not for a very bandwidth-focused business like snap.

I'll also add that they've come quite a way making it more modular. It's clear GAE will eventually become more a collection of parts rather than a locked-in stack.

You can already run GAE in a Docker container using app engine flexible.

The ceo got almost 4 billion taking the company public and he still controls his baby. I think he’s made some mistakes but keeping his company isn’t one of them.

Yeah but then he'll have to deal with the lawsuits and the bad rep when it all collapses.

irrational Can you explain? Curious.

Assuming they meant irrational because no rational person could turn down 4bn offer as they stated. My belief is that is a very shallow way to judge one's rationality.

I'm usually quite bullish on Snap. They seem to be doing some things right. From the outside, they seem overstaffed so it's good to see them reducing headcount. The fact that they're rebuilding the android app completely sounds like a bad idea that will be a bit of a timesink.

Nope. Snapchat is almost unusable on Android right now.

Compared to instagram's stories, which loads up in an instant, Snapchat takes its own sweet time. They made the mistake of not paying enough attention to android in the beginning. And now they are trying to fix their mistake. It is a timesink; but they brought it upon themselves.

Ditto. They did a big upgrade a few months (years?) back and yet my Nexus 6P takes literally 10+ seconds before Snapchat responds to touches. It's horrendous

Agreed. They need to up their Android game. This is a time sink for sure but it is absolutely necessary.

The Android app crashes frequently and freezes up even more often. Have you tried using the app for an extended period of time?

Instead of fixing each individual problem, they appear to have declared bankruptcy and are throwing out their old code.