Google domains seemed like a natural tool in the GCP and Google ecosystem. It provided one-click access to setup website, Google workspace, provided direct verification in Google Search Console and Google Analytics, and just made sense.
Unlike some of the other investments like Stadia, domains is also not capital or resource intensive. They never offered discount on domains, they sold them at standard prices. The team working on domains must be small. It’s also a low risk project. Why would they kill it?
What next? They’ll sell Google Fi to T-mobile and abandon Google voice?
Domain registrars end up with a long tail of customers holding small numbers of domains. If Google sells a .com for $12, most of that goes to Verisign, and they're stuck with a customer who will get agitated when "their domain doesn't work." Many registrars use domains as a way to upsell hosting or premium domains (e.g. GoDaddy, Namecheap, Gandi) but the median Google Domains user is unlikely to be attracted to GCP.
Google dislikes having to deal with customers at scale (free users are fine as you can ignore them) and domain registration is high on the list of businesses where you end up with large amounts of customers expecting a certain level of service coupled and almost no margin to provide it.
The section "Q: Won't Big Company X just clone your product and steal all your customers?" in https://apenwarr.ca/log/20180724 also described this phenomenon
I wonder if this is true. Technology-wise, maybe, but I wonder if domains just requires a lot of negotiation and contact with registrars. I suspect it might be quite labour intensive in that way.
Additionally, to sell to consumers you need to provide services like helping people port domains for renewal, helping with DNS setup, and other consumer-level customer support, and that's expensive. If you've ever heard a Hover advert on a podcast, it sounds like an expensive business to run.
That sort of business doesn't fit with the cloud business. For cloud, sure customers will want a domain, but any business using cloud will have the ability to buy a domain, it's unlikely to be a deal breaker for using GCP, and the only up-side is $10-$100 a year on the account, that's less than a single VM for a month in many cases.
Disclaimer: I work at Google but haven't looked into this at all, this is all personal speculation as an outsider and previous customer of GCP.
One is managed at domains.google.com, the other as part of a GCP project either in the Console or via API. See https://cloud.google.com/domains/docs/register-domain
They’re selling the former.
My domains are the starting point for my other infra/sites/web apps. And cloudflare has become that product that already sits there, dns, cdn and workers.
If google can’t sort that out, they seem dysfunctional. Cloudflare seems to understand the value and execute accordingly.
And their positioning as “at cost domains” makes it a no brainer. Only a matter of time before I move them all over.
Everything else is a hobby. Holding this removes from the focus of bringing up ad revenue. We will see them get rid of a lot of these things.
Speaking as a long time user of Google Voice I'd say that they all but abandon it quite some time ago. Voice still mostly works but it's barely limping along. I'm slowly, one by one, getting my contacts to switch to my carrier number because I don't hold out much hope for Voice being usable product for much longer.
I imagine they've underestimated what it takes to operate as a domain registrar, and presumably there's too much politics and conflict of interest in that space.
For a company like Google that pays top tier salary, there is no way that selling domains (especially not in bulk, their targets are mostly consumers and businesses, domain traders use other services) make enough money. If you take a look at the commiters on https://github.com/google/nomulus
That's at least 10 engineers maintaining one component. Add in SRE and product executives, you are looking at expenses in the two digit millions minimum.
The decision is definitely short sighted but somebody high up no longer wants it as a loss leader.
More seriously, Google's culture seems fundamentally incapable of longterm investment in anything that isn't directly related to their core advertising business, even if that thing is an onramp to their core offerings (as you pointed out). It's sad and frustrating to observe.
https://cloud.google.com/dns (Cloud Domains)
Google Domains competes with hosting registrars like GoDaddy.
The most logical thing that would make sense to me is that someone at SquareSpace has a close relationship with someone at Google that is able to influence such decisions?!
Why not?
For anyone that missed it, is there a link to the story you're talking about? on here or otherwise
And if there is a story on here (which obviously there would be) why aren't you just in there asking your question with everyone else? :-\