sigh I remember arguing with other engineers, managers, and executives, even though I was never part of the team, that IM had huge market potential. But they

sigh I remember arguing with other engineers, managers, and executives, even though I was never part of the team, that IM had huge market potential. But they had zero interest in exploring the space to find the right recipe. After the server and client acquisitions their one and only goal was to implement a long checklist of enterprise features, like LDAP integration with every conceivable variation, as if that was the only thing holding back the market. Those features were a bottomless resource sink, and the kind of thing that only makes sense to focus on after market adoption breakout. The IM space was too immature for that, but it was all they knew--grab some pre-existing solution, preferably open source-based, and refine it according to the demands of IT departments. It was also a waste of talent as your team had some of the best talent in the company and perhaps the only ones with the foresight and restraint capable of exploring the space and implementing big, novel solutions. I mean, a rich client GUI and a backend that permitted strong value-add data processing and transformation were what they acquired. But they never continued down the path that was being blazed prior to the acquisitions. Plus, other than spam it was the only project that saw (or even could see) significant internal usage, which means you guys could iterate quickly with live feedback, as opposed to being guided by the [poor] imaginations of the product and sales managers. Indeed, AFAIU it was only a couple of years ago that Barracuda moved off the internal IM product. Barracuda's anti-spam product, the only one that was even remotely trail blazing, succeeded precisely because they ate their own dog food; it was originally built to scratch their own itch while the founders were pursuing another venture. Everything required for success was right under their noses.... :( To be fair, by the time Slack came along expectations for enterprise integration were limited to web-based SSO, and you could throw together a multimedia rich GUI with Electron. Scaling simply meant using AWS, and the economics of cloud infrastructure meant enterprises were already used to higher per-user costs (presuming an attempt at profitability). There were fewer distractions. But at least Slack saw the potential.