Summary: Full year 2021 revenue $257B, 41% year on year growth. Net income $76B, up from $40B yoy.
Quarterly: Q4 of 2021 they made $75B revenue, most of it came from Google Search, $43B up from $31.9B in Q4 2020. Youtube $8.6B of ads, Google Network $9.3B from ads.
Google cloud did $5B revenue for the quarter, up from $3.8B in Q4 2020. It lost $890m this quarter compared to $1.2B Q4 2020. "Other Services" did $8.1B revenue and "Other bets" did $181B
It's one thing for a midsize company to grow revenue 41% but for a massive global corporation with 156,000 employees to do it is nothing short of astounding.
And this giant, mature company isn't sacrificing margins for top-line growth, they're actually increasing margins at the same time.
It's like they have a knob that they can turn: turning it to the left means fewer ads and a better user experience, and turning it to the right means more ads and more revenue.
My own experience from Search and YouTube is that they have massively turned the knob to the right. I don't know why they are chasing revenue growth with such craze. It makes me worried that the leadership don't know what's Alphabet's next big thing so they are just milking the cow while they can.
I just don't see how 40% YoY growth is sustainable. Before they were doing ~25% which was already insane, but 40% will now be "expected" and anything less will be considered a failure probably. There's a limit to how much they can crank that knob...
> I just don't see how 40% YoY growth is sustainable
2021 was like catching lightning in a bottle[1]: because of lockdowns and WFH, 2021 was a monster of a year for online platforms at the expense of brick-and-mortar and traditional platforms. A lot of consumer and marketing dollars were redirected to online retailers/ad campaigns. All that drive-time- and ballgame placement ads went to Google/Facebook/Youtube and a thousands of other ad networks
1. Not that Google wasn't going to grow, but the ad growth was absolutely juiced. At the beginning of the pandemic, a bunch of Ads that had been filmed ended up not airing because they showed happy people outdoors, unmasked and having fun.
> All that drive-time- and ballgame placement ads went to Google/Facebook/Youtube ...
Somehow Google captured a lot more of that lightning than Facebook did.
"Facebook parent Meta Platforms Inc. startled investors with a sharper-than-expected decline in profits and a gloomy outlook in its first earnings report since Chief Executive Mark Zuckerberg outlined a pivot to the metaverse."
"Meta shares plunged after the results were announced, dropping more than 20%."
"The company said it expected revenue growth to slow because users were spending less time on its more lucrative services. Meta cited inflation as a weight on advertiser spending ..."
> There's a limit to how much they can crank that knob...
That limit is imposed by competitors who deliver a better user experience. So far most competitors are doing a much worse job.
Microsoft should be a much bigger threat, but they seem to rely on their ability to set defaults for Windows and throw obstacles in the path of users who want to change the defaults.
That's not a better user experience, it's abuse of their dominant market position.
Up ~10% after market. Should be around 2T Market Cap now. Apple is 2.8T while M$ is 2.3T.
I think I am very good at numbers, comparatively speaking. But these number is till mind boggling. It is hard to comprehend how and why they could continue to grow. With $140B cash while Android is still half polished. GCP not competing. And killing half a dozen random products and services from time to time.
I have used a PiHole ever since I learned that LittleSnitch resolves DNS queries (to IP) before the dialog prompts whether to Allow/Deny a connection to the unlisted host. It is an added bonus that I can route my entire subnet to the local PiHole, which prevents rogue software/OS/devices/phones from initiating undesired connections. If you know how to make a few simple IPfilter rules, you can even stop hard-coded devices (e.g. smart TVs) from phoning home with internal DNS IP addresses — all you have to do is capture all DNS queries, IPs included.
Simply blocking pagead2.google.com and googlesyndicate.com will remove 50%+ of website advertising. ReGex rules allow for ads.* (etc.), and these rules apply on your entire local network. For an added bonus, you can then use your local PiHole to resolve DNS queries remotely (e.g. from your phone) — just all around an incredible product!
Was thinking more in the vein of similar but different to PiHole on a cloud instance. Connect as a VPN. Not sure how much lag that would introduce, and if you could use it in place of pihole for local network at home as well. Set it up once, use it for everything.
Probably not as cheap as a Pi, but more beneficial. blah blah. decisions! yay!
Ublock will get you 99% of the way there tho unless for some reason you're stuck using chrome. Then usually brave or vivaldi will fill in the gap. Avoid chrome always.
I just tested "macbook pro for sale" without and without ublock and they disappeared when I turned it on. Are you able to produce an ad in a search result right now with ublock enabled?
I don't use voice at all other than "set an alarm for ____" but firefox search works just fine and ublock origin works great on it. It's a little slower than chrome but you would never notice it unless you're doing ms level tests on it.
For searching in Google? I use Google with the Firefox mobile browser, yes. I also don't use voice commands on any platform, just transcription GBoard and I've had no issues.
Growth is primarily due to demand - by advertisers, not the users. Online ad spending increased during the pandemic and is expected to continue to grow throughout this decade. Advertisers flush with cash are looking to spend it online. And google, fb, etc are more than happy to oblige. Google is simply increasing the supply ( ads ) to meet the demand ( advertisers ). This sadly means we'll be seeing more and more ads online...
This is one of significant misconception for digital marketing. More ads doesn't necessarily mean more revenue since it's bound by total advertising budget. Let's assume an optimal ads market and you put more ads there without further optimization or budget, then there will be less auction pressure per ads (thus less CPM) and roughly similar revenue. For something like 40% YoY growth at Google's scale, you need to convince big advertisers to allocate more budgets.
They don't even need growth. Their earnings and profit margins are high enough that they can deliver billions of dollars of value every quarter to shareholders absent of growth.
It's gives a better understanding of what's the bottom line after expenses.
You could have a ton of revenue, Walmart for example, but if you have razor thin margins that revenue won't necessarily translate into giant piles of money in the bank, which in the end is what most businesses are after.
I think Amazon was different from Apple and Google in that on the way up, they intentionally managed their cash flows to sort of keep their margins small (always investing extra cash into growth). Walmart is that, but to an extreme.
They could sort of just make money on the interest of their cash flow (I'm not an economist or accountant, so that is probably not true)
If Walmart could be making more money on interest from their cash flow, they would be already be making more money. It is not a novel or difficult idea.
The profit margins of the business of retail is not in the MAMAA league. Being a retail middleman is simply not as profitable as what the tech companies can do because they can scale at extremely low marginal costs, and barrier to entry is very high.
Amazon’s value lies in AWS and prime video/music/logistics/platform, not the part where you buy something shipped and sold by Amazon.com. Hence why Amazon removed the button to filter searches to only items shipped and sold by amazon, and why Walmart introduced their own version of a platform where they can make money as a platform, rather than a retailer.
Perspective is weird with large amounts of money. I think human brains are just not good at big numbers. You need something to compare it to, and the big numbers of other companies still leave it completely abstract.
Perhaps it might work better with a comparison to a person, like "Alphabet made so much revenue this quarter that, assuming they had no expenses, after a full year, Alphabet would have roughly as much money as Elon Musk." But maybe not, because now you've just moved the problem to understanding how rich Elon Musk is.
It's much harder to understand how rich Elon Musk is, because if he tried to sell even a small percentage of his Tesla shares, the value would evaporate. The revenue figures of these companies are more understandable because that is cold hard cash coming in the door
He actually sold a boatload of TSLA last year [1]. But yes, his wealth is not really expressible in USD.
If you tried to acquire a similar stake in TSLA, the price would shoot up and it would cost you more. If he tried to sell all his shares, it would plummet and he’d get less out.
Just recently I saw a headline about former Bezos wife selling $8B of Amazon stock. Does not seem to have affected the price all that much. She is probably going to have to pay $2B in taxes for that transaction.
>adjusted the estimated useful life of our servers from three years to four years and the estimated useful life of certain network equipment from three years to five years
Another sign of the slowing advancement in CPU power/performance?
The 3-year figure was always totally unhinged from the actual useful life of cloud servers. As evidence, please refer to the fact that on EC2 you can still provision a C4 instance with a Haswell CPU made in 2014. In GCE you can still provision an N1 instance with a Sandy Bridge CPU from 2012.
Fair, I was thinking of this in the non-cloud perspective where efficiency improvements can push you to upgrade even if the hardware still works. In cloud provider mode it makes sense to keep it around as long as it still works and it's not too annoying to run. It doesn't really matter how (in)efficient the hardware is because you set the pricing to keep it profitable as long as someone's willing to buy it.
It's not that simple. Every rack that is sitting there is an opportunity cost for another efficient or more profitable rack. So, there has to be smart calculation to make the keep/upgrade decision
Disclaimer: Previously worked at Amazon but not AWS.
You would typically be right. However, the saying as of 2013 was "At any given second there is always at-least one new computer being plugged in, to support S3's growth."
> Every rack that is sitting there is an opportunity cost for another efficient or more profitable rack.
If AWS wasn't already supply side constrained on new hardware to fill new AWS data centers, you would be correct. However, they do not yet need to re-use those old racks, instead they are accelerating how many new racks they are building.
Any large user of datacenters could have a shortage of datacenter space if they didn't plan ahead enough. It might take two years to build a new data center? Space crunches have happened before.
But maybe the chip crunch is a current bottleneck, or results in higher prices for new hardware? If so, there's a reason to delay upgrades and run older hardware a bit longer.
This is just guesswork. Supply crunches aren't predictable from first principles.
In terms of investments, I like Google the most out of the monopoly techs. Microsoft is my least favorite because they’re just a fund of loosely related technologies at this point.
This makes the stock price in line with other dow jones stocks, something that Apple did right before it was included in the Dow average. The Dow unlike other indices does not weight by market cap or anything like that it literally uses the stock prices.
Increased liquidity. Also option trading is usually in 100 share units, meaning retail investors can now afford more options (e.g. $15,000 vs $300,000 a unit).
Anecdote warning:
This is one reason I have not been able to buy/sell google options. (I did around Brexit when google was 650 or so, but not since). Even a single contract is a massive risk for a layman portfolio like mine. 1/20 will imply I will definitely attempt to trade this.
It's not only psychological -- for example, a lot of brokerages don't offer fractional shares, so a split allows some retail investors to invest who couldn't afford to before. (Although that almost certainly has a smaller effect than the options contracts etc mentioned in sibling threads.)
This is the closest thing to investing in Renaissance Technologies. It's too bad there is not a way to get 2-3x leverage version of google. The CAGR since its IPO in 2004 is the same as Renaissance too. Rather than executing millions of trades, the profit is from millions of clicks. Like renaissance , massive statistical analysis is involved, such as tracking click patterns and optimizing the ads.
There is no limit to how high this will go. More and more economic activity is being funneled into these dozen or so mega-tech companies, which is how they are able to grow annual revenues at 20-40%/year despite 3% GDP growth.
I remember reading a paper from LSE that showed that advertising tended to crowd out investment in R&D and that in the long term increased advertising has a detrimental effect on the whole economy.
not as good. margin has high borrow fees and path dependency. options are better but still has a lot of fees, slippage, and needs to be adjusted to target appropriate delta. a 3x fund would make it so much easier and cheaper.
At least with IB you're looking at margin rates in the 1-2% neighborhood (which is potentially more like 0.6-1% after you take the tax deduction). Depending on your tax bracket, that's actually cheaper than the expense ratios of most (all?) leveraged funds, and let's you avoid the wonkiness that comes with holding leveraged ETFs over time
Azure doesn't mint money. If you look closely at any numbers MSFT reports whenever Azure is part of the mix, the margins drop considerably though revenue growth jumps. I had looked closely at their reports about a year ago.
Azure is usually bundled with some very profitable if stagnant pieces like Windows Server or SQL Server when reported on. You never see pure Azure numbers.
Amazon and Microsoft uses AWS and Azure themselves. Google doesn't use GCP, the service you buy is not the service Google engineers use internally. This both means that GCP cannot bill the rest of Google to inflate their own numbers and that GCP likely are less battle tested since they don't have a huge internal customer.
Also there seems to be a widespread perception among developers that Google will drop services on a whim. Whether it's true or not that has to have an impact on people signing up to their services.
To put that into perspective - if the rate continues for 4 years (doesn't seem that unbelievable) - revenue would be ~$24Bn - which is close to FB currently...
I am more surprised by having such stunning growth by many big tech without actually having any new businesses. It used to be that 10% growth was a lot and now people scoff at even 30%.
This is not 100% new business. It displaces existing one - on-premises, company owned datacenters and colocations. Just as electric cars are growing fast because they are eating into ICE market.
I don't know how many others think like I do, but I'd never use GCP. No one will ever convince me that it'll be around in 5 years. I just don't trust them to keep anything they made in the past decade or so going long term. Call me paranoid, call me whatever you want, but I just do not trust their ability to focus on anything anymore.
Not sure if sarcasm or not. These results demonstrate that the HN zeitgeist about the utility of Google is not related to its actual utility to the public at large.
This is very noticeable. I remember everyone here saying Atlassian was a sinking ship when they switched to cloud-only. Ever since, its stock has almost doubled. I can probably think of a pretty big list of other things where HN hivemind completely misaligns with how 90% of the normal population thinks.
Doesn't really follow. a) Google is a lot more than search and b) it's not like people noticing "crap, search results are getting worse" suddenly stop using it entirely, given there is no clearly superior competitor. Similarly with many other products.
Quarterly: Q4 of 2021 they made $75B revenue, most of it came from Google Search, $43B up from $31.9B in Q4 2020. Youtube $8.6B of ads, Google Network $9.3B from ads.
Google cloud did $5B revenue for the quarter, up from $3.8B in Q4 2020. It lost $890m this quarter compared to $1.2B Q4 2020. "Other Services" did $8.1B revenue and "Other bets" did $181B
They have ~$140B in cash and liquid investments.