Launch HN: Fogbender (YC W22) – B2B support software designed for customer teams (fogbender.com)

The problem that we solve is a painful one for both B2B customers and vendors, so I’ll describe it from each point of view.

A B2B customer buys a product and has an ever-changing group of people using it over time. Those users receive customer support as individuals, not as a team. This gets painful when the vendor can’t answer a question because part of the answer is only known by somebody else at the customer. This delays solutions, creates confusion, and hinders the spread of information.

A B2B vendor sells to teams and wants to support them effectively. But existing customer support products don’t do that—they only work with users one-by-one. To get around this, many vendors open a shared Slack channel with each customer team. Now they have two problems, because not only is Slack too unstructured for processing complex support requests, inevitably not everyone on the customer’s end knows about the Slack channel. The vendor ends up running two separate support systems, neither of which fully works, and now has the painful task of trying to synchronize the two.

To sum that up: support products don’t work with teams, and team products don’t work for support.

My first startup was a team messaging service (launched on HN as LeChat in early 2013! https://news.ycombinator.com/item?id=5202381), where we created a shared support channel for every customer—this is how we found out about the power of doing support this way. When competing against Slack got too difficult in 2015, we pivoted to sameroom.io, which bridges rooms/channels between different messaging systems. We got acquired by 8x8 in 2017, then built their messaging infrastructure.

While at 8x8, my team began using Segment Analytics, and usage soon spread to other departments. I struggled to keep the deployment sane because I had no way of knowing which of our developers needed help. Segment supported us with Zendesk, which was aggressively unhelpful. I knew the problem was important when I started getting emails from Segment Analytics with PDF attachments outlining how much money we owed them in overage fees—that was literally my only exposure to the work other developers were doing with our Segment setup. A lot of money and people's time was getting flushed down the drain, all thanks to improper tooling.

I decided to start a startup to address this because (a) I thought we had the technical expertise to do it; (b) it feels like the market is enormous; (c) it's clear who the buyer is, and the price point can be relatively high, and (d) I really want to use it myself.

Our product is similar to Intercom or Drift in that our customer (the B2B vendor) installs a messaging widget on their customer dashboard. However, if the dashboard is accessible to a team of end users, all users see the same data in the widget. They can use it to talk to each other and with the vendor, just like with shared channels. We also make it easy to link these conversations to support tickets in vendor’s issue tracker.

Fogbender uses a pubsub-inspired messaging protocol (currently the only transport is websocket, but we may add others), powered by Elixir and Postgres. The frontend is TypeScript, React, SolidJS, and Tailwind. We're fans of monorepos and prefer to add supervisors to our BEAM supervision tree instead of spinning up microservices.

Our philosophy regarding the messaging product itself is closer to Telegram than Slack. We like to have the ability ...