Why Does Google Kill So Many Products?
matt-rickard.com
matt-rickard.com
Their desperate desire to find another money printing machine to replace their soon to be regulated to death ads monopoly. If it’s not instantly printing money, they don’t want to continue to support it.
Ruth and Sundar need to be fired or else Google/Alphabet will suffer the same exact fate as every other company that focuses on cost cutting over innovation and supporting future innovations.
Not every company needs to innovate. Some companies do the opposite, they are late movers with outstanding marketing.
We grew up with Google being the cutting edge, but that is far more expensive and risky than hiring a bunch of psychologists and marketers to figure out how to monetize old product at bigger profits.
They are doing this with their Pixel series as a prime example. They originally made the best budget smart phone with ~0 marketing, word of mouth made the Nexus great. Today, they have removed some hardware, charged literally 7x, and have ads running.
There is a happy fantasy of capitalism where each company innovates and competes for the best product. The reality is that plenty of companies are fine with producing a low quality product, selling it at a high price, and having the marketers find some niche that makes the product appear high quality so customers can rationalize their decision.
I genuinely hope that Marketing/Psychology is 'canceled' before I die. Its a horrible thing for humans.
> Ruth and Sundar need to be fired or else Google/Alphabet will suffer the same exact fate as every other company that focuses on cost cutting over innovation and supporting future innovations.
Although I wouldn't be sad to see them go, I'm not sure it will make a difference anyway. The vast majority of companies, especially those that built themselves on the foundation of being "innovative", will see an inevitable decline. The greater the rise, the sooner the fall. The reason is that successful businesses reach what I call the "cash cow" phase where enough individuals depend on the business's cash flow that risk-tolerance goes out the window. It's nearly impossible to do a rewind from the cash cow phase without firing almost everybody, and especially those at the C-level; even then, the chances of replicating former glory are not good if all the original founders are long gone and not coming back. The Google will of course always be in business, but it is on the fast track of being the IBM of the Zoomer generation.
Is bizarre that this and how it's tight to the way they do promotions has even evident to people outside Google for years and nothing changes.
The Apple business model is near impossible to replicate, but it is proof that it is possible to ride a cash cow indefinitely.
> indefinitely
Right. Indefinitely, just not infinitely. We've yet to see a company like Apple remain on top for hundreds of years.
Urs leads the SWE charter and pushes down technical complexity for promos. Sundar for the lackluster vision and continual not delivering
During some exploratory exercises on the future of entertainment, they kept pushing him to come up with "planet-scale solutions".
After some hesitation, he started pitching them on "Space Basketball" and the opportunities for a league on Mars.
He got the on-site interview after that.
Yes, making things people want to buy costs money. Most successful companies spend a certain fraction of their money on making what they sell. Making it cheaper just makes it cheaper. Which is definitely a choice one can make but few people respect the bargain brand.
Oddly the people who understand this the least also seem to care a lot about prestige, so I don’t understand how their brains work.
I think the whole project killing long precedes Sundar or Ruth. In fact chrome (which Sundar handled product for I believe) is something that has endured.
If you reward new products and don't reward continuing support on products without enough revenue/growth, that looks a lot like Google.
As an example, if I was CEO…
Killing any public facing product requires CEO approval. Boom instant oversight and change in behavior. Quickly raises the cost of killing something and the CEO can decide the brand damage and financials.
If I was Ruth…
All promotions should show the employee has an understanding of long term value, and understands how to help maintain a big system.
At places I’ve worked in the past promos expect to see on-call success wins and bug fixes and similar as part of a well rounded packet. Seems like an easy add.
Amazon maybe because AWS is not retail/logistics etc. Maybe Sony because they have (had) a decent sized film studio? Pretty much no one else.
No one wants to admit it, but economically a company with a decent position (like google) SHOULD ruthlessly cost cut, pump out dividends and leave research/innovation to nimble start ups. From a purely Dollar point of view, that is the correct approach...
I agree though that returning cash to shareholders should be more common when there are no good investment opportunities for the company. The book The Outsiders goes into several super successful CEOs who have the common thread of treating their business like an investment portfolio. I'd strongly suggest stock buybacks over dividends though because that gives control to the shareholders about when the taxable event occurs and will typically lead to lower taxes. It's somehow frowned upon by the general public, but it's very similar in what it does to dividends but more efficient at it.
Greed is the ultimate killer of innovation and sustainable business practices. But the board/c-suite doesn’t care unless it’s the one holding the bag at the end.
OpenAI was supposed to be a hedge against Google's monopoly of artificial intelligence. It's clear now that even if Google is sitting on the most powerful AI model on the planet, they have no idea what to do with it. They call it caution, but in fact it's indolence.
So fun fact - at FAANGs (at least) introduction of new vernacular is very important to show that you are the objective (tm). Eg we will now reward landing instead of just launching. All well and good.
How did that translate to in practice? Well X had impact by helping, supporting across many broad projects that needed legacy support. And X was a junior eng who went above and beyond in doing so - some times showing ownership and accountability at level + 1. So X should be ready for a promo yes? Yes but - "Focussing broadly meant low - more objective sound verbiage - focussed technical complexity is not enough". Translation - "you didn't build a complex shiny thing".
Google, however, is much more of a one-hit-wonder from a revenue perspective. While Google has had many societally impactful products, they mostly do not move the needle on revenue. Google's revenue is 80% ads and 75% of that ad revenue is from search ads.
I think the implication of this is that because Google doesn't make money from their interlocking services but rather from their near-monopoly on search ads, the incentives for them to actually behave like Microsoft and Apple (where you can trust long term support) just aren't there.
"Android has an incredible 97 percent market share in India".
https://arstechnica.com/gadgets/2023/02/androids-new-oem-rul...
Google and the phone vendors only take on half the responsibility of publishing an operating system as you are pretty lucky if you get software updates never mind the kind of major updates an iOS device gets.
It is exhausting to explain to people how to scan a QR code on Android because the built in camera app doesn’t do it so they have to go to the app store and download some free QR reader that looks like spamscam and have a slightly different name every time you try it. (And no giving them a QR code to get the app.)
That’s the inattention to detail that pervades the platform and it is made worse by the outright vandalism of Samsung or Verizon filling up your home screen with lots of twisty little app icons that all look alike and certainly don’t lead anywhere you’d want to go.
Google doesn’t get paid for Android, Asian companies in corrupt economies seem to lose money making Android phones but there is some social status to be had by being a leader in 21st century technology so they do it anyway. Apple probably profits more from Android because Android keeps them out of trouble with antitrust regulators but Android is a ‘spoiler’ that keeps any real competitive OS out of the market.
Hypothetically Android keeps mobile safe for advertising but it is possible Google could be selling as many ads on somebody else’s OS without the expense and moral hazard of keeping a zombie OS ‘alive’.
They may have had a very good handle on Android talent in the start, but now every facet of Android engineering has a non negligible community outside of Google.
I'm not saying it's like Google has nothing, but their position in the Android universe is one or two bad moves away from slipping from their grip at any moment. It's not a hard link like Apple's.
Also, arent 'ads' a stand-in for paying for a service? Google has search and youtube. They find niches where they can place ads in places that people typically won't be willing to pay for. I don't want to pay 1 cent to view some website that might be garbage, but I'd be willing to waste 5 seconds scrolling past an ad. However, they have shown that people will pay for streaming services, something people know is valuable.(And they also have free/ad based options)
It just seems that people are caught up on Ads, which is really just the payment method you use to access google services.
(Also 80% is a common pareto thing, Apple gets ~80% of their revenue through various iphone/ios channels. Apple's one hit wonder is the iphone. Its not exactly something to nitpick)
the revenue of google was 279.8 billion in 2022 [1]
GCP made 26.28 billion in 2022 [2] Google play made 42.3 billion revenue [3]
There are probably some other non-search add revenue streams besides these two, I am not too familiar with google - but 'one hit wonder' isn't entirely exact.
[1] https://www.statista.com/statistics/267606/quarterly-revenue...
[2] https://www.statista.com/statistics/478176/google-public-clo...
[3] https://www.businessofapps.com/data/google-play-statistics/
Google is more concentrated than Apple and Microsoft and their ancillary offerings are less synergistic with their main offering.
The differences between the two processes is a function of their environment. If you are working in a company with separate functional departments and a strong, existing brand, you need a step by step process to align everyone and get customer insight before launch. Stage-Gate answers this need and products that get killed die before they are formally launched (thus protecting the brand).
If you are a small team doing everything with an unknown brand, the most reliable way to get market feedback earlier is to just put something on the market. The Lean Startup comes from this perspective - if there isn’t enough traction, kill the product and invent a new brand.
(Of course, these processes can be adapted to other environments, but these are their native soil, so to speak.)
Google seems to be managing its product initiative like startups: they incentivize new product launches, and don’t hesitate to kill products that are already on the market. Perhaps we are better off adjust our expectation of the google brand: it’s just a VC brand (like a16z), not a product brand.
I don't see that those incentives even exist at Google. A well managed company would introduce new products that match the rest of the company's portfolio. When a product fails to gain interest, they would revise it and try to figure out how to provide customer value. There's nothing about Google's behavior in the past ten years that suggests that's what is going on there. From an outsider's perspective, there's nothing to indicate that Google is making any effort to follow through to make sure new products are successful.
It's like Google has the worst of both worlds: the upper management has the detachment of a VC firm, and the product teams have the detachment of being incentivized to start something new. At least in an actual startup, the team is committed to success.
Between the expectation that products won't last, and the possibility of getting randomly locked out of one's account with no recourse (other than screaming loudly on Twitter or HN), Google services seem to be considered a major liability.
I've also said this on before on HN, but I typically don't hear about new Google services until they're shuttering - for an advertising I've always found it ironic how poorly they advertise most of their products.
Yes, this. I'm great with experimentation, but please give us an indication of what services are experiments vs. what we can rely on as critical infrastructure. A lot of us were blindsided by Google Domains because of this!
I'm personally amazed Pixel phones haven't been killed yet and can't shake the feeling that will happen any day now.
The Nexus line followed the JDM model you describe, ending with the 2015 Nexus 6P designed by Huawei. Starting with the Pixel in 2016, Google has been the designer and seller of record for all their phones, accelerating with the acquisition of HTC's phone team in 2017. The 2021 Pixel 6, moreover, used Google's in-house SoC ("Tensor"), which had been under development for years, rather than a Qualcomm chipset. That was a significant milestone.
Google is no special snowflake in this regard, but they hit it big early, and they still have the resources to keep trying things (and hoping for another big win). Lots of consumer-facing useful things that sometimes get popular, but rarely give Google whatever is it they want. And unless it really matters to their bottom line, they probably aren't going to stand behind it long-term.
That being said, it's more than fair as a consumer not to trust their (little) bets.
They ship little bets and see what sticks.
Even with sizeable discounts and professional services funds for migration, I wouldn't consider a move to Google Cloud until they calm this down. This is coming from someone that once worked for a Google Cloud Partner.
Can someone help me understand how mail and docs are the roads that lead to search?
They're just to big (and expensive) to be doing mass incubation.
It's much more cost effective for them to simply buy-up and buy-out others.
The pros of Google Workspace/Cloud:
- Everything (for us) can be Google: from general business stuff like email/calendar/etc to storage (Drive and/or Cloud Storage) to automating tons of it via their Cloud API (or Google Apps Script, when that's the right tool for the job).
- Everyone at my company, especially the non-technical folks (i.e. the majority of people in my company) "gets" Gmail, Calendar, Meet, Drive, etc. It's ubiquitous, and so selling administration, billing etc on "let's just go all-in on Google" is easy.
- We don't need S3 storage, but we do need a lot of (mostly hot) storage, with easy collaboration. Because user accounts pool Drive storage company-wide, we mostly use that for our cloud storage needs. So, 99% of our storage needs are effectively free, because we're paying for those user accounts anyway. This saves us thousands upon thousands of dollars per year, and that's not even accounting for the employee time needed to integrate and work with S3(-compatible) storage.
- The Cloud API is pretty solid, the documentation is generally good, and it's language-agnostic, making hiring that much easier as we expand the team and automate more processes.
- It's all good enough, and rightly or wrongly, it doesn't have the Microsoft "smell" -- a lot of my non-technical peers are in my experience quite resistant to Microsoft-anything, even though they arguably have a much better track record of not killing products.
The cons:
- I don't trust Google anymore. I expect them to kill or radically alter any of these core products at a moment's notice.
That one con completely obliterates all of the above very compelling pros.
There's honestly nothing else quite like Google Workspace. AWS is massively beyond our needs or budget. Microsoft 365 is perfectly fine for mail/calendar/Office/etc but not for the Google Cloud side of things (I sure as hell am not touching Azure, and their 365 API documentation is a mess).
The only other option is to roll our own set of modular services, which is something we've looked into quite a bit, but there's always a show-stopper somewhere, not least of which is the thought of someone else eventually inheriting this stuff if/when I move on. At the end of the day, rolling it yourself and maintaining it simply requires more chops and time on the part of the employee(s), and I shudder at the thought of essentially building a cathedral that others allow to rot once I leave, one big reason being that the middle of the venn diagram of "the skillset of people who would find themselves applying for my position" and "Linux sysadmin chops" don't really overlap at all.
Philosophically, I prefer moving the monetary cost from closed-source platforms to the salaries of developers collaborating on open-source products; but for our company's particular use-case, we're ultimately not a tech company, we're a media company, with a lot of tech-focus, yes, but not in terms of building products -- rather, we simply do as all companies in 2023 probably should do, and leverage as much automation as possible to keep things efficient, organized, scalable...generally, keep it on the up-and-up.
I'm sure that I'm not alone in feeling this way. There is a vast, vast I-dare-say "silent majority" of companies out there that are in this weird middle ground -- too small for AWS, not tech-focused-enough to build it themselves, but still need a solid suite of...everything required to run a business, and the tools to automate processes as needed. I know that Google Workspace is in the "enterprise" side of things and thus is less likely to be killed off, but...also, who knows? Google is so obviously completely aimless that I truly do not trust them anymore, and yet I don't see a truly viable alternative.