Well, a business decision can be driven by all product decisions, and if we look at the weird consumer/business crossover with Google the empirical answer is "probably just use slack."
A business decision can be driven by all product decisions, but that would be a bad way to make a business decision.
Look at that discontinued list and look at the business services that have been discontinued, can you find any? Can you find more than 5? Look at G-Suite + GCP (Google's business offerings) and tell me how many there are. There are dozens. Many of which have been around for 10+ years (Email, Docs, Sheets, Analytics). So if I'm a business, making a business decision about a new business service from Google it's an incredibly safe bet that this service will be around for the long hall.
You can point to consumer services to try and discount that... but you'd be wrong to, and I think you know that.
Why does it have to be marketed as a business service to be considered a business service? We used Hangouts at my last job until it started shitting the bed in Firefox and Google told us to move to Allo or Duo or whichever. Yea we weren't a 50+ enterprise account, we were just a little English school in Taiwan, but we were a business using a google service that got shafted.
Business service = you pay for it, either directly or as part of your G-Suite subscription. I don't believe you that they told you to move to Allo/Duo. Hangouts works, still works, and will continue to work for the near future. It has hundreds of millions of users.
If you spent 100k to build integrations though and are all in on google per se... you get screwed out of 100k if they decide to change the way things work.
Much better imho for a growing business to use their own in-house tools that they can mold how they want and that won't change UNLESS they want them to.
"In-house tools", do you mean something built in house?
If you're spending 100K in integrations you think building an entire service, or suite of services is going to be LESS than that? Now take into consideration opportunity costs. This is insane.
I'll be generous though and believe that you're talking about using open source solutions. That's a more viable option, but you still have to do integrations so the cost is effectively the same, probably more because Open Source solutions are typically far less polished.
And even Open Source projects change dramatically over their life. Everything changes. That's technology. Better to get better at dealing with change than try and hold the world still.
This is the thing that really boils me. I mean, we can debate whether XMPP is a great protocol, but for a brief window it seemed like we’d have an open chat standard with at least one big stalwart to bring in the normals.
But then Google flipped us the bird and took their ball and went home. Their AMPification plans for email and my residual anger over screwing us over with XMPP was the straw that finally made me switch my vanity domain from Google to Fastmail.
EDIT: Every tech company seems to want to re-Compuserve us :-(
In fairness, Facebook pulled a similar stunt with messenger. It's marginally redeeming that their Android messenger app isn't terrible, but really annoying to lose the xmpp client connect abilityability, and any useful desktop client.
Apparently Google did allow broken (no tls) server side federation too - so they screwed us over twice - both on client and server.