PSA: If you use Google Domains to buy a domain for your company DO NOT USE IT TO ALSO BUY GOOGLE APPS (or whatever it's called this month).
If you ever want to change your company's primary domain... you can't. We're stuck now with a legacy domain (configured with an alias so it's invisible to others) that we can't move off of without basically burning the whole thing down and starting fresh. It reeks of technical debt and poor engineering design that buying Google Apps through a reseller (where that reseller is _also_ Google) totally ties your hands for properly managing your deployment.
Is this yet another service to be added to Google's graveyard in a few years time?
Domain registration is not Google's core business and I can't see it being massively profitable(double digits), so it will likely face the chop soon.
I am sure they will offer a way to transfer your domains, however, why go through this pain when the outcome is inevitable.
My rule of thumb is to only buy a service of a company if it is their core business and customer retention actually matters to them. If Google Domains gets shut down none of the share holders will suddenly panic, it probably won't even register. As a result customer retention and satisfaction will not be very high on the priority list.
Google delivers services that work wonderfully for 95% of customers until the day they arbitrarily decide to shutter the service with no warning. I use a whole bunch of google services these days but they're all the services that have clear exit strategies - unless Google pulls the plug without warning we've got migration strategies to get off them without too much damage.
I know that Google is notorious for shuttering services, but I can't imagine they would shutdown Google Domains without giving customers an opportunity to transfer domain ownership to another registrar.
What is ample time for a company that hired a webdev one time to build them a static website and never checks webmaster@domain emails? What is ample time when you've got ops scripts set up to automatically do domain stuff but don't employ that person anymore?
What is a reasonable money expenditure to migrate off a platform when you could look at the company record and see that you're much better off just choosing a different platform? Is a week or a month of dev hours worth putting your faith in a company notorious for randomly shuttering products?
I feel like Google marketing has no idea the absolute monster of a problem they've created by allowing their business to get a reputation for being extremely unreliable in the B2B space.
Yes they have a reputation for merciless shuttering of services, but come on.
What is ample time for a company that hired a webdev one time to build them a static website and never checks webmaster@domain emails?
That kind of outfit is much more likely to be a victim of missed renewal or a myriad other technical screwups than a Google-initiated exit of the registrar business.
I feel like Google marketing has no idea the absolute monster of a problem they've created by allowing their business to get a reputation for being extremely unreliable in the B2B space.
They can certainly count on a spate of hyperventilating HN posts whenever they announce any kind of service-related news.
More seriously, their most celebrated servicides are consumer-oriented, like Google Reader. "Extremely unreliable in the B2B space" sounds like an exaggeration to me.
I mean my company, which isn't small, jumped ship from G-suite based SSO for Okta a few years back out of business concern - so while single anecdotes don't make for proofs it's definitely not non-existent.
I'd also suggest just bringing your tone down, this isn't a fight and calling people on the other side of the argument "hyperventilating" doesn't really do anything to promote meaningful discussion.
Been there a half-dozen times with Google B2B products.
Best case was GSuite/Workspace, and not the recent one. I had a startup in the freezer (developed a lot of IP nights and weekends, but we needed to stop for about a year). Google sent two emails about the domain being idle, but filed its own emails as spam. If I don't log in within some time period, so I lost the domain. Google wiped out all the data of that startup.
Just because I don't check my gmail spam folder often enough.
It's getting hard to tell if you're witnessing astroturfing, Stockholm syndrome, or if we've simply become so calloused from all the past transgressions from these companies that we're now just going straight for blaming the customers for everything.
There are two kinds of people: Those who do backups, and those who will. What Google did in this case is inexcusable, but so is keeping all your own data in one place.
This experience helped move me from the first category into the second. I now do Google Takeout periodically, and have a full backup on external drives.
I'm also slowly degooglifying my life (and recommending the same to everyone around me). There is another grouping of two types of people: (1) Ones who don't do business with Google anymore (2) Ones who won't. That one took more incidents than backups.
My grandmother is now 82 years old, she has lived her entire life on a farm in rural New Zealand. She had not used a computer until she was in her early 70's, and she is now computer literate enough to know to backup her data to an external drive and does so regularly.
It's 2022. You dont need to "have a sysops background" to know or be able to backup your data.
The cost of backups also depends on your income. Stories like this led to more backups, but so did going from being a student, to first-job-out-of-college, to a senior employee.
There's a spectrum between a hard drive costing a full day's work at minimum wage to a few minutes work for a Googler, Amazonian, or a Metamate making $500k/year.
People said this about almost every service they ever surprise killed. Just let go of this idea that you cannot imagine this for whatever good reason, because other people have made this mistake before you :)
I agree, very easy to use. Prices are pretty good, but not the cheapest. But the most important question is: What happens when they ban your account? You will lose your domains.
Congrats to the Google engineers who made this happen. But I transferred my domains out a few months ago because I will never win against the Google AI.
Until Google comes out and segregates actual and false positive ToS violations such that it doesn't kill your entire Google account, I can't in good conscience use any additional Google services.
This began TEN YEARS AGO when Google+ "real name" violations disabled your entire Google account. Gmail, Google Photos, AdWords, everything. This is completely unacceptable and has never been addressed or rectified.
If you use Gmail then your Google account is simply too large of a single point of failure to risk further exposure.
What happens when a false positive disables my account? Can I still transfer my domains out? I'm not convinced that's true so thanks but no thanks.
Even beyond the risk to the account (a non-zero risk), Google has other issues. Only this week I've had an order for some hardware cancelled by them for "sales terms" with no way to identify what sales terms I was in breach of even though I could contact a support person and was spending hundreds of $$$. It's not a huge loss for me to not be able to purchase some Nest equipment as there are a lot of competitors I can purchase equivalents from, but the experience underlines that Google support is terrible - support people are hard to contact and have no agency/autonomy/authority to solve anything.
The T&S risk is real enough that I wouldn't use Google Domains. But that they can't even support products that they 100% own and sell and which are very high margin (compared to domain names) really doesn't inspire confidence.
Tying your domain to a Google account — or any centralized megaprovider account — seems fraught with risk. What happens when Google bans your account for some misstep, or even a mistake on Google's part, in some other service? Does the domain just expire because you can't access your account?
At least if you separate your email provider and your domain, if the email provider decides to ban you you can move the domain to a different email provider and maintain continuity. But if the domain registrar bans you, what then?
It seems as though the wisest move is to keep your domains hitched to the most minimal account you can — avoiding centralized service providers such as Google, Amazon, etc.
Why jump through these extra hoops and deal with the risk? There are tons of reputable domain registrars out there, and I don't see Google Domains as providing some kind of killer feature to make it worth it.
I use Google Domains with an isolated single-purpose Google account. Emails from it are forwarded to my main account. It is easily the best domain registrar out there IMO.
I'm skeptical that accounts with megaproviders can effectively be isolated. If one account gets banned, and Google determines that another account is owned by the same person, they may well ban the other account too.
The solution is to register your domain (at least) with a provider that doesn't have so much insight into their users.
Again, this applies not only to Google, but also to Amazon and others. It's really a problem with centralized reputation.
Maybe you could share your experience with email that Cloudflare forwards?
I understand that they don't offer email hosting but do let you configure email forwarding. If I do that and forward my me@example.com email to something like GMail, Apple Mail, Hotmail, etc..., do my replies look like they come from me@example.com or do they come from the account the email was forwarded to?
Edit: I think I found the answer[1]. Replies come from the account the email was forwarded to, not me@example.com. That seems pretty useless for anything but receive-only accounts.
As far as I understand, this is usually the job of your mail server (I've done it with Exchange and GSuite). You add the domain to your mail server and then create aliases (or wildcards) in there and add them to your account, and then when you reply you can reply from the alias.
I only use Google Domains for any domains I can't get on CloudFlare. I definitely used to agree that they were the best, but CloudFlare is basically wholesale pricing and have the best DNS management, so I've moved over there for pretty much everything
A few weeks back I couldn't add/edit/remove domain names that start with a _ and contained an accent. That's when I found out how good their support really is, even for a bug on their side.
Probably best to use a service from a company primarily in the business of offering paid services.
Nothing about that setup is normal, feels quite risky... Every login you make, perhaps years apart, will be wildly different from the norm and could easily be flagged.
But...why? I mean, I get it -- domain registrars aren't exactly at the top of my list for easy-to-use, reliable, trustworthy partners that I want to build a core part of my business or identity on top of. But I'm damned sure that Google isn't either, after 15+ years of launching or acquiring fantastic tools and swiftly burying them in the Google Graveyard [1]. Say what you want about Namecheap or GoDaddy; at least being a registrar is core to their business and not just a lark.
Namecheap and Porkbun are actually very easy-to-use, reliable, and trustworthy. I don't think Google is trying to actually be better than other solutions. They're not competitive on price or support. I think their whole plan is to capture people using GCP since they're already there anyway, a la Route53 or GoDaddy. Why add complexity to your system with multiple providers to save a few bucks a year or help with theoretical issues?
Namecheap recently suspended an account because of a tweet thread speculating that a domain was maybe related to abusive behavior. Turns out it wasn't at all. It was so arbitrary even the people who were speculating were surprised by that decision.
Even Google isn't sk arbitrary as to base their ban decisions on random Twitter discussion
I have had problems with people using bots to spam abuse reports on my domains so that they can buy them after they're taken from me. Namecheap is one of two registrars that actually buckled.
Porkbun is amazing. I had a problem with my FIDO2 key (there was no password reset path for you if you still had your FIDO2 key), so I let support know. Less than a day later their engineers got back to me letting me know they had added the feature and deployed it. Everyone there is super nice as well :)
"We launched Google Domains in 2015 to be the easiest place to find, buy, and manage a domain."
I'm sure we're all painfully aware of the longevity of many Google services that are considered "in production", there's a long list of things shut down.
BUT is it common for something to be in BETA for 7 years? That seems like a REALLY long time for something to be beta anywhere. Is that common at Google?
Now seems like a good time to point out that Google is fully aware of the linkages between all of your various Google accounts (work, personal 1, personal 2, side project, other domain, etc). Both mobile platforms allow multiple Google account login and if you use one account for work slack and one account for personal gmail then thanks to your phone OS's giant gaping per-app sandbox hole, big daddy G can see all of your credentials are being used on the same device and connect them together as a single user.
This means that any itty bitty TOS violation on one account means you potentially lose all your accounts. Accidentally contradict the WHO (who, for example, said not to wear masks) in a YouTube comment and get your personal video watching account TOSed for posting fake news? Accidentally mention an ITAR/SDN country by name in a comment field in a Google Wallet/Pay transaction? Send a zipped malware sample to a colleague for analysis from your gmail? Say goodbye to your Play Store apps published under your work account, AND your domain names registered under your side project account, AND your Google Voice number, AND all your cloud instances on your GCE account, AND all your YouTube uploads on the video account, AND all your files in Google Drive, et c.
You also lose ability to reset all of your accounts that have your @gmail.com address as the login name, and you lose the ability to even log into any accounts for which you may have used that "Login with Google" button.
Google is not trustworthy. Don't give them your money or data. Register a domain with someone who isn't Google, and point it somewhere that isn't Gmail. There are step-by-step instructions on how to do this on my website.
Yea - they're an absolutely terrible service that has lost a lot of ground to more relevant comparisons in recent years. Personally I've had no issue with webnames.ca but I assume that's just a Canada thing.
I've been a Gandi customer since 2014 and my opinion of them is not as favorable, especially since they rolled out a new and buggy dashboard a few years ago. Historical problems I've had:
1. When attempting to update the email address on my domains and accounts in 2018, the form to do so was broken. Support merely linked me to the broken forms and did not comprehend the problem. After re-explaining the problem, they changed my address to something completely wrong, which was alarming because it wasn't an address I controlled. It was eventually resolved, but it was not an experience that inspired confidence.
2. After getting married and having my legal name changed in 2019, there was no easy process for updating Gandi's records. Support walked me through some convoluted and time-consuming process that involved creating a new account and transferring all my domains to it. I recognize this use case isn't as common, but it shouldn't have been as difficult as it was.
3. Because of the combined shenanigans of (1) and (2), I now get five duplicate emails from Gandi for everything. I have notified support of this a few times over the last couple years, but it's always met with, "sorry, it's a known bug", and I guess it will never be fixed.
The only reason I haven't moved off of them at this point is because of inertia and I don't know what other registrar would be better.
I've used Name.com for many years and have been very happy, but recently switched basic shared hosting from another provider to Gandi.net and got exposed to their backend and tools. I am now considering transferring my domains over as they come up for renewal; been very impressed!
Providing services to Russians =/= supporting a dictatorial regime. That's very reductionist in my view. Many Russians simply cannot up and leave their country at once, or worse, maybe they are running a website that broadcasts against the Russian government, and Namecheap was one of their only options. I don't think it's as clear cut as "provide them service" but it's more nuanced than that/
Meaning Namecheap treats those living under dictatorial regimes as second class people. As a citizen of a country headed in a similar direction, this only makes me less inclined to deal with the company.
Since complaining on Hacker News is the only way to effectively reach Google employees, I'll say that as a customer I am confused at the offerings of Google Domains and GCP Cloud Domains (via Cloud DNS). I recently registered a domain via Cloud Domains and it shows up under Google Domains.
> Since complaining on Hacker News is the only way to effectively reach Google employees,
That's an exaggeration. You can find a bunch of them on Twitter or LinkedIn. Joking aside, B2B services have support, and consumers can pay for it via Google One.
Given Google's habit of having some "AI" randomly kill someone's account, combined with a total lack of customer service, I can't think of a company I would trust less with my domains. It still makes me nervous that my email is hosted with them, but that's something I'll probably get sorted soon.
Domains are incredibly cheap for what you get, but they're harder to use than they need to be. Someone shouldn't have to understand DNS records, IP addresses, ports, NAT, TLS certs, VPSes, the command line, etc in order to securely connect their domain to an app like Plex/Jellyfin/Nextcloud running on their own hardware.
You should be able to install a Nextcloud app on an old unused Android phone, go through a quick OAuth flow to connect it to a subdomain, plug the phone into power in a corner and be done with it.
Cloudflare is currently the closest to this ideal with their domains + Cloudflare Tunnel, but cloudflared is still CLI-only and all their products are built for developers. Plus they'll never offer end-to-end encryption.
I also maintain a list of more DIY tunneling services[0], but again CLI knowledge is currently a requirement. My main side project at the moment is seeing how far I can lower the barrier of entry.
> DNS records, IP addresses, ports, NAT, TLS certs
These and then some are things you will not have to care about if you instead expose it as a Tor hidden service - worth considering for the self-hoster.
May not be ideal for high-bitrate movie streaming but for many other use-cases like Nextcloud it's great and easy to set up.
My recent experience with Domains was trying to delete a Google Workspace account that used to have an attached domain. It took several weeks of back and forth with their support due to problems on their end, with them asking me various things like to help them debug the admin console front end, to pay them to re-register the domain so they could cancel the account, etc, etc. Lots of pass the buck "oh that is a different department" but no follow-through on connecting me to the right one. It was bizarre.
That's the point of brands! I don't understand how they don't understand the damage they've done to themselves in certain circles. Clearly their brand is still valuable to a wider audience.
I've long wondered why companies don't put their more experimental stuff under a different brand ("X by Google"-- no not that X), and promote them to full brand-name once they've gone reliable. Probably they want such experiments to get a boost from being associated with the better brand, and they figure the brand stain of stopping those experiments isn't so bad. But these things build up over time, and each individual product manager probably don't think they're the last straw...
Another thought is that Google comes from an era of much more experimental web. Now the Internet is a much more established market, and platform stability is more a premium. Google is having trouble transitioning from their wilder younger days (of just a few years ago) to a more reliable company.
> I've long wondered why companies don't put their more experimental stuff under a different brand ("X by Google"-- no not that X), and promote them to full brand-name once they've gone reliable.
They still have most of the Alphabet to do this too. I'm still surprised they didn't already do it years ago.
Absolutely, HN's echo chamber is not reflective of normal people. But there is a growing worldwide discontent with FAANG's domineering influence. Google is hammered almost as much as Meta by political/regulatory bodies, and news of that influences how regular people see those companies. The ever-encroaching ads also push people towards competitors. Wait a few years and most users will be looking for alternatives.
Most users on this site would cut off their own thumbs before saying something about Google that isn't negative; it's just one of the hallmarks of the HN community.
To everyone complaining about the possibility of you google account getting false flagged as TOS violation: do you actively use this reason to avoid all other domain providers like GoDaddy who can terminate your account?
Just make a separate email account that exists solely to manage your domains, and this advice goes for using ANY domain registrar/provider.
> To everyone complaining about the possibility of you google account getting false flagged as TOS violation: do you actively use this reason to avoid all other domain providers like GoDaddy who can terminate your account?
My domain provider is only used as a domain provider, everything else - email, hosting etc is elsewhere.
I won’t wake up one morning and find that I’ve been caught in some kind of youtube purge that disables my account.
It means your domains have been canceled by an algorithm that determined your account is in violation of Google's terms of service. Hopefully you know a Google engineer or have experience starting Twitter mobs.
'Have experience starting Twitter mobs' is something I'm excited to start seeing on resumés soon. Followed by the discussion over whether that shows key leadership/organization abilities or not, a la the "Wow Raiding Guild Leader" debates of the late 2000's.
Looking how much of a mess they made ending the free Gsuite accounts with the multi-users, mail routing etc. you'd be a fool to trust Google products for any free feature.
They even promised the free gsuite accounts would be free forever in writing and they aren't, so imagine relying on any feature of Google domains.
I'm wondering how long until certain aspects of gmail aren't free anymore.
As someone leery of tying my domain to my Google account, I'm curious what people recommend for this use case. I have a vanity domain currently hosted on a web hosting platform. These days it only gets used to forward the email of a few people to their contemporary email accounts, all gmail based.
As I no longer use the web hosting, storage, or anything else like that I'd love to find something cheaper that can just handle the email forwarding. I like the idea of using Google as for many things I trust them more than a smaller company. But there are too many stories out there of people having no idea what they did wrong and losing all access to their account, with no recourse.
Agree, cloudflare is ideal for parent's use case although you have to join a waiting list of several weeks for forwarding to be activated (my experience anyways).
Happy customer of Cloudflare Email Routing - it works exceptionally well. I've paired it with Amazon SES for the occasional time I need to send emails from custom domain.
I feel sorry for anyone who actually use(s|d) this. Google have proven time and time again their services cannot be trusted. Why anyone would actually use this is beyond me.
I'm sure there will be some play at vendor lock-in and definitely holding your data to ransome; but worse is the uncertainty this service will exist for any meaningful length of time. How long before you're mysteriously locked out with no way to log back in and no one to contact to sort it out?
Aside from all the very good reasons people are giving for not using google, the question I'm asking is
What is google doing in the domain space that is different from any other provider? What's their unique USP?
Anyone technical can use another provider and connect to Google services if they'd like (or other service).
Anyone non-technical that is using SquareSpace (Canva now has a website builder as well - but no domain sign-up yet), can get a domain through SquareSpace.
Additionally: Are there any good registrars that support U2F/FIDO/WebAuthn (aka hardware tokens) for 2FA? Mine only does TOTP and I would like the security provided by hardware.
It's very easy to get started, and easy to progressively add services.
The most important distinction is between a domain registrar and a domain reseller. As far as I know, most start out as resellers (Namecheap started as an Enom reseller. AWS still resells Gandi). Being a legit registrar is actually fairly expensive.
Reselling complexity depends on the upstream registrar. Most of the big ones have reseller programs, including white label services. TakingNames.io uses Name.com, which doesn't require anything official. I just sell domains using Stripe for payments then use the Name.com API to buy the domain. Then I forward DNS changes using their API as well.
Currently I default to using Name.com's nameservers, but users can set custom nameservers and I may run my own eventually.
I use Porkbun because they're typically the cheapest, tbh. They have a really nice interface too, but my registrar isn't something I interact with more than once a year typically. As long as they aren't shady like GoDaddy, it doesn't matter that much. There's tons of good options.
I wanted a ".dev" domain for a project a few years ago. At the time, Google was the only one offering that TLD. However, it looks like you can get ".dev" domains anywhere. Was this change part of Google Domains leaving beta? Or is it completely unrelated?
All other domain registrars out there have probably started thinking about their "transfer back your domains from Google Domains" for when whoever at Google loses interest in this and shuts down.
With my experiences with Google in the past, I would never trust my domains with them. Not my personal ones, and especially not my business ones. I also don't trust their customer support.
> Google Domains is available to anyone whose billing address is in one of our supported countries: Australia, Belgium, Brazil, Canada, France, Germany, India, Indonesia, Italy, Japan, Mexico, Netherlands, Spain, Thailand, United Kingdom, United States, Vietnam, South Africa, Switzerland, Philippines, Poland, Malaysia, New Zealand, Singapore, and Sweden. We’re working hard to expand our availability.
I was using a Google Cloud database last year on a production system. It shut off without any emails or notifications or messages or phone calls. I won't be fooled a second time.
I used to be a Namecheap customer, I have been using Google Domains for a few years now. Namecheap isn't that much cheaper, maybe a dollar or two per year in my quick look just now. I've found Google Domains really easy to use for my personal domains, including for dynamic DNS. So far it's just worked and I've been happy with it.
Not to say that Namecheap was hard to use, just that Google is pretty easy and the cost difference isn't enough to make me want to consider switching back to Namecheap.
I do not recall why I stopped using Namecheap, it was a while ago.
Interesting when I clicked the link they offered a 20% discount. Two minutes later I get an email offering a 30% discount.
I used Network Solutions before moving to Google Domains and I will take no customer support over bad customer support any day. With network solutions, domains were over priced. To get the real price, you had to call customer service and ask for a lower price.
If I were you I would never do this, mostly because there’s already an insane amount of competition so it’s really, really not worth giving yet more power over your sanity to Google.
Someone marks your business @domain emails as spam, now your account is locked. Why would you put yourself in this situation? Just, don’t.
It looks like a lot of people here are unhappy with this. I'll add a voice as someone who recently used Google Domains and had a pretty good experience.
lol no thanks. Congrats to all the engineers that have worked on this and getting this out the door, but I don't think anyone should use or recommend these Google services, especially when they have been SO inconsistent about the life, longevity and future of their products.
I'm curious to be proven otherwise (and open to criticism of my claims), especially when other services have existed for a while that provide the same functionality, or offer better tie in with alternative products (hosting, pages, workers, lambdas, compute etc).
I think a larger worry would be running afoul of one of Google's capricious, difficult-to-reverse account suspensions. I wouldn't want my domain to end up expiring or becoming otherwise inaccessible because Google's algorithms decided to suspend my account over something innocuous I posted on YouTube or put in Google Drive.
This is my feeling as well. When I left in 2010 I shared in my exit interview that they should seriously consider embracing actually investing in customer support and leaning into the idea of a "normal margins[1]" kinds of businesses. I speculated that as their advertising margin continued to deteriorate (it was already going down fairly rapidly by then) that without these sorts of 'base load' businesses to carry the operational costs Google would begin to struggle to maintain goodwill with their "free" services and "lavish" workplace perks.
To be fair though I was a lot more pessimistic (which is why I exercised and sold off all of my ISO options and RSU options) so left some money on the table. :-) That said, having gone on to be responsible for operations at Blekko I felt I got a pretty good understanding of the moving pieces associated with the operational costs of providing "web 2.0" services. And that informed a better sense of what level of margin was required. Certainly Amazon does that calculation all the time in pricing their AWS services.
So I read this announcement and wished the people who signed up a good experience that didn't leave them high and dry with no where to turn. I've got high hopes and low expectations.
[1] Search ads generating a billion dollars a quarter in free cash flow that can be run by a company 1/10th the size of Google is how profitable they were at the time.
> When I left in 2010 I shared in my exit interview that they should seriously consider embracing actually investing in customer support and leaning into the idea of a "normal margins[1]" kinds of businesses. I speculated that as their advertising margin continued to deteriorate (it was already going down fairly rapidly by then) that without these sorts of 'base load' businesses to carry the operational costs Google would begin to struggle to maintain goodwill with their "free" services and "lavish" workplace perks.
What would it have looked like if you were right about this? What would it look like if you were wrong?
I ask because later you mention that you left some money on the table, but it looks more like you left most of it (85%+) on the table (which is fine and impossible to know ahead of time!). It just seems to me like you made some predictions but every metric points to those predictions being incorrect. Have perks been removed? Has compensation fallen?
> What would it have looked like if you were right about this? What would it look like if you were wrong?
Google's per ad click revenue (CPC) started falling like a rock off a cliff during the mortgage crisis and has continued to fall since. Google also sold ads through third party sites (AdSense for Content) where they shared a meaningful and proportional chunk of that revenue with the site provider. As part of their employee onboarding I had added AdSense ads to my personal web site to get a feel for how that part of the business worked.
I also have a number of friends who ran web sites with AdSense ads and we all saw exactly the same pattern, for the same number of page views the "revenue" generated by the ads was tracking the decline in CPC.
What that said to me was that if Google maintained the level of advertising on its own sites and continued to share the same revenue with third party sites, then in roughly 5 years it wouldn't be enough money to cover costs. At that rate in 2013 or 2014 Google would have its first layoff, and then by now it would be a much smaller company or perhaps bought by someone (Apple comes to mind). (I did say I was pretty pessimistic right? :-))
But "making more inventory" (putting more ads on a given page) has zero marginal cost and if you add more inventory than the cost per ad decreases, statistically you increase your revenue.
Also if you're the #1 search engine you automatically get the best possible ad prices because you have the biggest audience. But what is #1? Well its the search engine with the most traffic going through it, and if you're not organically getting enough traffic you can pay people to send you traffic. When I started at Google they paid less and $100M a year to third parties to direct traffic to Google search, by 2015 they were paying a billion dollars a quarter for that traffic. Why? Because if they didn't pay that the traffic would go to other equally good search experiences (notably Bing).
I've watched Bing too and seen their CPC values go up while Google's were going down. The most likely explanation of that is their increase in market share and the fact that a lot of Windows people didn't bother to change search engines and of course they offered fairly cheap API access for third parties who wanted to create a custom search experience but didn't want to invest in owning and operating a web crawler, indexer, and ranker.
So if I had been right, Google would right now be a division of Apple or maybe Microsoft. Meta (nee Facebook) could have made an excellent play here and wiped out Google by integrating a search engine into the Facebook experience. They chose not to, largely because Mark just didn't "get it." He did not appear to understand search is social, after all when you're going to buy a new car who do you ask for recommendations? Your social network of friends. Blekko demonstrated a really killer implementation where we could return search results ranked by what you and your friends liked vs what random people like (which is what Google/Bing do) and it was pretty amazing (not great for privacy and had a bunch of other implications like finding out one of your friends likes web sites selling BDSM gear when you search for mask :-))
So as the search revenue declined, Google has compensated by adding inventory (using search for things like products is nearly all ads), buying ever more traffic through deals with folks like Apple and Mozilla, and cutting costs by killing off projects/products. Sometimes directly and sometimes by forking the people responsible into another group under Alphabet and then quietly selling or closing down that group. The mean time to live once you've been spun out into an Alphabet "other bet" seems to be pretty short these days.
I've got lots of ideas and anecdotes about how they got there, but basically their behavior has now put them into a very difficult position.
Starting a new service costs money, the market doesn't trust you'll stand behind your service, so adoption (during which you lose money on the service) is slower. As a result you lose more money than you might have with rapid traction and growth and this hesitancy masks the "is it good?" signal. So you run with the service, which is losing money, unable to distinguish between it losing money because nobody likes it, and losing money because people like it but are afraid to adopt it. If you then cancel the service from a lack of adoption, you increase the hesitancy and you don't really know if the offering would have been successful had it not carried the 'baggage' of people not trusting you.
Obviously, even though I was wrong about Google "being dead" within 10 years I'm still not bullish on their future prospects.
True true. Also, this is a story about the curse of success. Many of their cancelled products were really things that nobody cared about, but since google itself was so high profile, that a relatively few disgruntled users had such a high resonance because reporting and echoing their woes generated clicks elsewhere.
I believe this is in reference to the legacy Google Apps for your Domain being sunset and the free accounts being moved over to paid accounts on Google Workspaces.
I did the same after having one of the free legacy accounts for about a decade. Migrated over to Fastmail and de-googled myself.
Not the .app domain but the google apps service(now called google workspace (I think... they change the name at the drop of a hat)).
They basically removed a free tier that they explicitly guaranteed would be free "forever". Because of this people set up friends and family as users using the service. Google then told people no you have to pay ridiculous amounts per month for them to keep any apps they have already bought tied to your account and there's no way to seperate your account once it's tied in. They did this with a 2 month notice period.
They have since walked back some of that due to backlash. They're now saying that you have a 6 month notice period (July 1st) and that at some inderterminate time before july 1st they'll let you transfer your account to a normal free google account. You will however lose your email address as part of the process.
A service that was supposedly free forever has been shut off with only a few months notice, an aggressive price structure, and no help to migrate data. Poorly communicated and badly executed - par for the course for Google these days.
Google Apps for Domains, basically GMail & other services for your own domain. It was free intially, in 2012 I believe they stopped the free version but grandfathered the existing free accounts. Now in 2022 Feb they announced that those free ones are going to cease to exist.
> I don't think anyone should use or recommend these Google services, especially when they have been SO inconsistent about the life, longevity and future of their products.
Can we also say the same thing for GKE and Google Cloud since it is also a Google service?
Google's behavior has convinced me that it's perfectly OK to use their services, as long as you don't rely on them. They may terminate your account at no notice, or announce that a service is shutting down or adopting an unworkable pricing scheme at what would be reasonable notice to an individual, but is not acceptable to a business.
The worst things seem to happen randomly. They happen to a small percentage, perhaps even a small absolute number of their customers -- but they do happen, and there is no recourse.
Google Cloud is a whole different game. If they have shut down any service there I can't remember.
But it's generally a bad idea to rely in any capacity on products marketed to consumers (not B2B) by Google, even Workspace I feel iffy on using, as you can be locked out on everything at any time.
To clarify the context though, it wasn't a comment, but what was supposed to be an internal memo. However, since I can't find a single trace of this either here or elsewhere, I retract my comment and admit I may just have had too much crack one night.
I don't see a problem. It's a domain. You can always transfer it somewhere else. It's not like other Google services that hold your data and which you might not be able to get out anymore
A registrar cannot prevent you from transferring for something that’s not directly related to a domain dispute, that’s an ICANN rule, and a rule of most ccTLDs registries. No registrar can hold you hostage, or that could backfire horribly for them. Meaning they could get their ICANN accreditation revoked and basically get shutdown, and/or paying huge fees.
There are email addresses, a phone number, a physical address, and an online form to contact them.
Have you ever tried to contact Google? I have, using a button they designed to contact them about Site migration problems. You think I ever heard back from them? Hell no.
So if your gmail account gets hacked, sure - you are supposed to still have control of your domains according to ICANN's rules. But if Google only supports domain changes via email, and you can only contact them via email, and your email doesn't work, I think you'd be completely fucked. And complaining to ICANN about Google would just be pissing up a rope.
What I need from Google before using their service for something important is a promise along these lines:
> We will keep running this product, with existing functionality substantially unmodified and backwards compatible, for at least 10 years, with pricing not increasing by more than inflation. Or, if we discontinue the product, we will refund 10x the amount you have paid for the product, plus $1000.
And you can't contact us. Unless you use that one trick we don't want you to know: post it on twitter and go viral. Then it'll all be fixed within the day.
Yeah IMO this is the kicker. A block-first process is obnoxious but livable with if there's a clear understanding of why you were blocked and a quick way to fix and notify of remediation actions. Without that the policy is just stupid and anger-inducing.
The number of products, Google actually "killed" and not just rename, merge into another product, or develop further under a new name, is quite low. And frankly, most people, complaining about product lifetime, probably never even used any of those products at all.
Hey - I'm still personally sore over iGoogle. They killed off an absolutely trivial to use RSS reader - I made heavy use of iGoogle while it still existed... and they killed it off to try and save Google+ which they then proceeded to also abandon.
I've definitely felt the burn of Google just randomly killing off a useful thing for some opaque marketing decision that turned out to be extremely ill-advised.
I'm not worried about them killing the service, I am worried about them blocking the domains of my business, for some vague reason, and me having no ability to talk to anyone to have it solved.
I am sure it doesn't happen often, but if it does it sure is not going to be to me.
So you're saying all I have to worry about is my account being nuked by an opaque algorithm or the price suddenly being jacked up 20x, like say the Maps API?! Great, sign me up!
> The number of products, Google actually "killed"... is quite low.
That is an extraordinary claim, but I'm listening.
https://killedbygoogle.com has 264 listings. Are you saying that if we excluded those that were actually renamed or folded into other products, that it would reduce 264 products to some number that is "quite low"?
The killed products that I personally actually used were:
YouTube Originals
Google Chrome Apps
G Suite (Legacy Free Edition)
Material Gallery
Google Sites (Classic)
Google My Maps
Google Play Music (replaced by YouTube Music, which is a degraded experience I pay for but do not use as much)
Google+
Google URL Shortener
Google Goggles
Google Nexus
Picasa
Google Reader
I personally think 13 is a lot, but understand that's a matter of opinion. The quantity is less important than the quality, though. I relied on Google Reader, G Suite, Google My Maps, Google Play Music and Picasa. I know of businesses built on Google Chrome Apps, and purchased services from one of them.
Those particularly personally inconvenient ones made me understand that I was too reliant on Google products, any one of which could be yanked at any time for any reason or no reason: Gmail, Voice, Google Drive, Google Cloud Platform, YouTube, and now Google Domains.
I don't have insight into why some products get yanked and others do not, and came over time to understand that my internal model about it does not correspond to actual Google reality. For instance, I might think that Gmail is way too valuable a property for Google to ever kill, but I would have thought that about many of those others, too.
Sure, if a specific feature of Classic Google Sites was used and is now different or was not merged to the New Google Sites, it's a matter of opinion.
But that's what I mean. I cannot account for all of the services you mentioned, but there is still Google Sites, there is still Google Workspace as the G Suite successor, many of the My Maps features were just moved to Google Maps. The hardware line of phones is now Pixel instead of Nexus. Obviously it evolved from Nexus and there is stuff from Nexus that didn't make it to Pixel. But if this is what we're talking about, every single tech company on the planet is "killing" stuff every time they release a new version of soft- or hardware.
Those are just examples off the top of my head. There are things that Google "killed" & I loved Google+, I used the URL shortener regularly, and the end of Google Reader was a tragedy (probably the one Google product I personally miss), but this huge list isn't a very fair depiction, IMHO.
Google's "customer support" is so bad, I think it would be a huge mistake to let them control your domains. Dealing with Google's "support" has been a nightmare over and over on many unrelated issues.
They also abandon many projects or hold you hostage in extremely inconvenient ways.[1]
Which customer support? I've heard nothing but good things for the actual customer support you pay for ( Google One) as a consumer, and a mixed bag for the not-enterprise GCP support you kinda pay for.
I’ve had simple issues with Google Ads that took MONTHS to resolve. Basically their algorithm marked one of my sites as having an image that contained malware. There was nothing I could do about it. That’s right, somehow they think a PNG hosted on AWS Cloudfront had malware. It was absurd, and they kept canceling my ads and never fixing the problem. Their customer service is atrocious.
People have been complaining about this for years. Besides Google Cloud, have they addressed this or do they keep ignoring it? Does anyone know if Google Cloud has customer service that's equivalent to AWS?
It isn't just Google Cloud. I've had absolutely terrible experiences with several Google support departments. It's often impossible or difficult to resolve the problems.
A domain isn't the kind of thing that you can quickly move somewhere else when things go wrong.
I could be convinced to put up with the lack of customer support if the price is right, but the price is higher than competitors. I'll stick with Cloudflare Registrar.
If you ever want to change your company's primary domain... you can't. We're stuck now with a legacy domain (configured with an alias so it's invisible to others) that we can't move off of without basically burning the whole thing down and starting fresh. It reeks of technical debt and poor engineering design that buying Google Apps through a reseller (where that reseller is _also_ Google) totally ties your hands for properly managing your deployment.