The case for consolidation(benn.substack.com)
benn.substack.com
The case for consolidation
https://benn.substack.com/p/case-for-consolidation
11 comments
There's nothing wrong with the Microsoft stack. There's also nothing wrong with any of a dozen other stacks. A small team that knows their stack well will perform well.
Microservices, on the other hand, are an effective way to burn engineering resources on something that feels like productive work but delivers no business value. I wouldn't infer much about the merits of tech stacks based on that experience.
Microservices, on the other hand, are an effective way to burn engineering resources on something that feels like productive work but delivers no business value. I wouldn't infer much about the merits of tech stacks based on that experience.
> There's nothing wrong with the Microsoft stack.
Except for the security mono-culture, where a single vulnerability can take out your entire operation. Just ask Maersk about NotPetya:
* https://www.i-cio.com/management/insight/item/maersk-springi...
* https://gvnshtn.com/posts/maersk-me-notpetya/
Except for the security mono-culture, where a single vulnerability can take out your entire operation. Just ask Maersk about NotPetya:
* https://www.i-cio.com/management/insight/item/maersk-springi...
* https://gvnshtn.com/posts/maersk-me-notpetya/
Yeah I was going to say the same. They likely failed at micro services because it isn’t their wheelhouse and it’s a skill set you need to master.
I’ve seen lots of successful service oriented architectures
I’ve seen lots of successful service oriented architectures
Look at stack overflow. They run a very popular site as a monolithic app on a small number of IIS and SQLServer boxes (and presumably some amount of CDN/cache)
As usual it boils down to: open and modular is better except that people won’t pay for it, so we get closed monoliths and SaaS even if they are often inferior.
It really does seem that you have to make businesses pay. Otherwise they won’t, and if they don’t there will be no resources to actually build what people want.
But as I always say: people will spend $300 in a month at Starbucks. No problem. But paying a quarter or less of that for software is a hard no. So we get what we pay for, or rather don’t get what we won’t pay for.
It really does seem that you have to make businesses pay. Otherwise they won’t, and if they don’t there will be no resources to actually build what people want.
But as I always say: people will spend $300 in a month at Starbucks. No problem. But paying a quarter or less of that for software is a hard no. So we get what we pay for, or rather don’t get what we won’t pay for.
Contrarianism can be fun, but I draw the line at the idea that users don't care about product quality and therefore Slack is being bulldozed by Teams. Teams users are going up (nevermind this metric is probably heavily juiced and meaningless) but that hasn't caused Slack adoption to go down. There will always be a hard core of Slack users who will never use Teams because it would ruin their async collaboration culture.
Their growth is being bulldozed, and they are losing/have lost more than a few big enterprise customers. This doesn't mean that their revenue will drop to zero, but it drastically changed the fortunes of the company: Without teams doing what it did, slack might not have to have gotten acquired.
But really, this isn't really about users not caring about product quality: Users definitely do! The whole point of the Teams strategy is that enterprise sales is not about users at all.
I'll use as an example a large multinational, with a market cap over 50 billion.Many divisions had been using slack since at least 2015: integrations all over the place, engagement throughout the day and night.. a wonderful example of slack being a strong incumbent. But this winter, Slack was completely ripped out, despite the fact that the only group that prefers teams are counting money. Teams is for all intents and purposes, free with everything else that they are paying Microsoft for, and from the perspective of the people making the decision, Slack and Microsoft are the same: But we already have slack at home! Executives weren't using Slack much at all, they will not use Teams either, and for them, it's a matter of dollars. Losses from having to adapt? Communication losses from a tool that works badly? Invisible, just like when you outsource your cloud team to big outsourcing.
Some divisions wanted to just keep a slack that they'd be billed for directly. Hell, some teams would have paid for the monthly fee from their own pocket, given how much worse their work lives have become with teams... but this was a mandate from above, and with the head of IT behind it, you are going to have a lot of trouble connecting to any slack while on the VPN. Yes, I am told that their async collaboration culture, is severely damaged... but productivity losses are blamed on factors that have nothing to do with executive decisions.
This really is the secret of enterprise sales: Users and decision makers are different in almost every case, and large conglomerates understand this very well. And when dealing with said decision makers, bundling is always an advantage. To compete against a bundled product, you need to have something that is far better than the competition, in a way that is very clear to the people that pay for the software. So it can work for, say, zoom, as remote meetings are still a part of the executive suite. But selling, say, a better Jira? It better cost zero, or be far better at executive reports. Being better for the teams themselves is just not going to be very valuable.
But really, this isn't really about users not caring about product quality: Users definitely do! The whole point of the Teams strategy is that enterprise sales is not about users at all.
I'll use as an example a large multinational, with a market cap over 50 billion.Many divisions had been using slack since at least 2015: integrations all over the place, engagement throughout the day and night.. a wonderful example of slack being a strong incumbent. But this winter, Slack was completely ripped out, despite the fact that the only group that prefers teams are counting money. Teams is for all intents and purposes, free with everything else that they are paying Microsoft for, and from the perspective of the people making the decision, Slack and Microsoft are the same: But we already have slack at home! Executives weren't using Slack much at all, they will not use Teams either, and for them, it's a matter of dollars. Losses from having to adapt? Communication losses from a tool that works badly? Invisible, just like when you outsource your cloud team to big outsourcing.
Some divisions wanted to just keep a slack that they'd be billed for directly. Hell, some teams would have paid for the monthly fee from their own pocket, given how much worse their work lives have become with teams... but this was a mandate from above, and with the head of IT behind it, you are going to have a lot of trouble connecting to any slack while on the VPN. Yes, I am told that their async collaboration culture, is severely damaged... but productivity losses are blamed on factors that have nothing to do with executive decisions.
This really is the secret of enterprise sales: Users and decision makers are different in almost every case, and large conglomerates understand this very well. And when dealing with said decision makers, bundling is always an advantage. To compete against a bundled product, you need to have something that is far better than the competition, in a way that is very clear to the people that pay for the software. So it can work for, say, zoom, as remote meetings are still a part of the executive suite. But selling, say, a better Jira? It better cost zero, or be far better at executive reports. Being better for the teams themselves is just not going to be very valuable.
You are telling a story I can relate to. I am part of an org that uses both. As an employee from an acquisition we are still using slack internally. And teams as mandated by mothership.
It is okayish when talking to colleagues from different parts of the corporate. Or clients. But usage internally is near 0.
We are not forbidden to use it, but will be forced (due to "security") move to the most expensive tier on order of HQ. I know that there are talks if it is feasible with this price tag attached.
And I know the brain drain it would mean when loosing another symbol of not fully being the mothership.
It is okayish when talking to colleagues from different parts of the corporate. Or clients. But usage internally is near 0.
We are not forbidden to use it, but will be forced (due to "security") move to the most expensive tier on order of HQ. I know that there are talks if it is feasible with this price tag attached.
And I know the brain drain it would mean when loosing another symbol of not fully being the mothership.
The charts don't suggest to me that Teams stole Slack's growth. Slack's trajectory looks about the same before and after Teams was introduced. I'm sure there was some negative impact but it's impossible to read off what it was from the graph.
As for the rest, yes, this is an old story, but doesn't apply to all companies. Slack will always have its niche. I know of a much larger multinational that didn't ditch Slack, it just added Teams.
As for the rest, yes, this is an old story, but doesn't apply to all companies. Slack will always have its niche. I know of a much larger multinational that didn't ditch Slack, it just added Teams.
> Yammer was collecting revenues in the low eight figures and burning through a bonfire of cash every quarter. Microsoft, by contract, practically ran a printing press in their basement.
Funny typo (if it was indeed supposed to read “by contrast”).
Funny typo (if it was indeed supposed to read “by contrast”).
This is why we are a pure Microsoft shop today. Between Visual Studio, GitHub and .NET/C#, we have been able to construct a stable, long-term B2B product stack with a very small team. If we had selected a more diverse tech stack, it is almost certain that we would have failed. I know this because we tried microservices, simply on top of a pure Microsoft stack, and it almost killed us. I cannot imagine the depths of hell we would be in if we had tried to mix & match different tech on top of that horrible idea.
Are we taking some risk that we are missing out on some amazing new tech that Microsoft would frown upon? Absolutely. But, we know what works for us, the roadmaps are clear, and our customers are happy (they love that we only have ~1 vendor). The things that Microsoft does not like, we just invent in-house. Fortunately, there isn't a lot of stuff we disagree on these days.