swyxio / swyxdotio

This is the repo for swyx's blog - Blog content is created in github issues, then posted on swyx.io as blog pages! Comment/watch to follow along my blog within GitHub
https://swyx.io
MIT License
342 stars 45 forks source link

Slack is Fumbling Developers #214

Closed swyxio closed 2 years ago

swyxio commented 2 years ago

source: devto devToUrl: "https://dev.to/swyx/slack-is-fumbling-developers-142f" devToReactions: 52 devToReadingTime: 7 devToPublishedAt: "2020-05-17T16:43:08.223Z" devToViewsCount: 2181 title: Slack is Fumbling Developers subtitle: and the Rise of Developer Discords published: true description: Reflecting on the sudden switch away from Slack for Developer communities tags: Reflections canonical_url: https://www.swyx.io/writing/slack-fumble slug: slack-fumble

Reddit | HN | Dev.to

A few days ago I had a sudden realization: I hadn't been active in Slack in the 2 months since I left Netlify. Those of you who live and work in Slack know how big this is - I personally went from opening Slack practically every day from 2015 - 2020, to zero meaningful usage whatsoever.

In its place, I am now active in a ton of Developer Discord channels. I feel like this is a meaningful shift this year, and based on responses, I'm not alone.

But They Didn't Want Developers?

The obvious proximate cause of course, is that I am professionally and personally highly attuned to developer communities, and Slack has actively pushed away developer communities. Quote Harry Hedger:

Slack lost Reactiflux bc they wanted $70k/month to support them.

Sure, I get that this was a deliberate decision. Storage costs money. But how much are we talking? Further, every company understands the logic of running "loss leaders" in order to seed stickiness and future growth. While literally every other company on Earth is throwing all sorts of free benefits to attract developers, Slack found developer communities growing like a weed on it, and flicked them away like so much dandruff. Stewart Butterfield is like the anti-Ballmer.

I'm not here to argue that Slack lost developers because it didn't want developers. That's a tautology. I'm here to argue that this is a strategic fumble that opened up the field to whatever will eventually replace Slack for startups (Microsoft Teams' success is difficult to assess so I will ignore it here).

Slack Losing Its Way

Developers are your canary in the coalmine for user experience - because they create the damn things!

Slack's original appeal was that it had a much better user experience than prior work chat/email/collaboration tools including the now-dead HipChat and Campfire. Reams of VC pitch decks were made about the "Consumerization of the Workplace" thanks to Slack's approachable design (as MetaLab never fails to remind you) and early touches like emoji reactions and bots. The theory was that we demand the UX polish that we see in our personal lives, in our professional lives as well.

Fast forward to today: "I’ve been using Discord a ton lately. Slack for work and Discord for hobbies." That's from Marc Grabanski, who knows a thing or two about developer trends.

I'll raise you another point: In 2016, Slack was originally backronym'ed into the Searchable Log of All Conversation and Knowledge.

Fast forward to today and advice from the curiously high amount of "how to work from home" content you've undoubtedly been receiving is overwhelmingly Communicate Asynchronously and Use Anything But Slack for Long Lived Knowledge.

The Game is Changing

Meanwhile the UX bar has risen in the past 5 years:

The past year or so in Office Productivity has been a free-for-all as other companies moved into Slack's open flanks. Notion has torn it up as the Searchable Knowledge Base to beat today. Zoom has overwhelmingly captured video chat, despite Slack also offering it natively. Even a bootstrapped startup like Tuple has become successful despite Slack buying the literal category maker in the space. And of course Discord and even Telegram have made inroads in real time chat.

My intuition is that the Great Unbundling has come for Slack. Whether or not it survives will probably depend on this: Is text chat really a feature or a product? Either way, having to answer that question is an undesirable strategic position for Slack to be in. There is no victory in this fight.

Note: if you feel lost right about here, I released the Developer's Guide to Tech Strategy chapter of my upcoming book recently, feel free to pause and check it out.

Not Over Yet

Slack has fumbled a ball, but it still has a strong lead. Two features still make Slack a strong fit for realtime workplace communication: threaded messages, which Discord refuses to build, and Shared Channels, which I've gone on record as saying will add years' worth and billions of dollars of revenue. Both were introduced in 2017, probably the peak year of innovation for Slack before creating the Slack fund in 2018 and then its 2019 IPO.

From my limited anecdata, the no-code Workflow Builder hasn't yet had significant adoption, but I would be happy to see this change over time. All great platforms eventually add low/no-code automation.

Slack has also made heavy investments for Enterprise Adoption, with things like Enterprise Grid (also in 2017! hmm...). I of course don't have any insight into Slack's growth in the Enterprise. I'm fully aware of the economic incentives of going upmarket and forsaking low end customers.

Communities Over Teams

The net result of all this is that Slack is now very much not the operating system for your team that it set out to be. However its entire user experience is tied to teams.

More specifically - having one single home team, interacting with that team on one device, and the assumption that you don't join or change teams very often. Perhaps this is the disconnect we are sensing.

Slack's focus on teams may be becoming outdated in this new world we live in, where Deep Work and organized knowledge bases come at a premium, and our professional work crosses professional and personal boundaries, and spans across multiple devices, modalities and even multiple communities we both lead and participate in. Discord's focus on communities may win hearts and eventually wallets. Here's Kurt Kemple on Discord's community features, for thought:

Discord is more community friendly IMO. Things like moderation, roles, boosting, group A/V, all make it a great place for a more democratized and scaleable platform.

By focusing on communities instead of siloed teams, Discord becomes a friend of community leaders. In other words, Discord aggregates aggregators. In a world of Aggregation Theory, this is a very good thing.

I thought it was Paul Graham or Benedict Evans who said this, but am unable to find the source: "What hackers do for fun today, we will do at work tomorrow".

Edit: It was Chris Dixon: "What the smartest people do on the weekend is what everyone else will do during the week in ten years"

Hackers are deserting Slack in droves. Slack should be on high alert.


Recommended reads:

Postscript

Here's a full inventory of emails you opt in to EVERY DAMN TIME you sign up for Yet Another Slack:

Just f right off with this engagement spam.