Integrating Chat Platforms

We integrate a number of third party APIs and SasS platforms to make the Get Well Cities platform work. Some of them are very, very good. Others? Well, not so much.

The biggest challenge we've had so far is with chat and messenger vendors. We spent months trying to make Zendesk's Zopim product work. In the end, it was just too buggy and management had no commitment to fixing things. In a Hail Mary play, we switched to Intercom.IO. Their platform at first glance appeared promising and support is really good, but the longer we worked with their dashboard, APIs and iOS SDK, the more apparent it became to us that the rule at Intercom is, "Our platform doesn't do that." Now the "doesn't do that" just appears to be a function of thoughtless or meandering application design and architecture and not rational product road map.

All platforms start as an idea and grow organically from the beginning. The net result is an implementation that over a few versions things becomes brittle and fragile. As features are added that were not contemplated in the garage design stage, enhancements and bug fixes just becomes harder. The code debt becomes enormous and new feature development takes a long time. We call it churn and turbulence.

This appears to be the case at both Zopim and Intercom. What makes the situation intolerable to continue on those platforms is the lack of commitment on the part of executives at either company to fix the problems by either hearing (and acting on) customer feedback with a promise to fix issues or biting the bullet and starting over. Starting over takes courage. Great startups like Basecamp bit the bullet and rewrote everything in version 2.0. We did the same thing with both our iOS app and our CMS at Get Well.

Anyway we tried really hard with both Intercom and Zopim. The farthest we got with both companies was middle management. We never did manage to engage their executives or product managers in a meaningful conversation. We'd think that the recent Facebook Messenger announcement would wake them up. No such luck. Chatbots and invisible apps are the hot new thing in tech. While Zopim and Intercom were early, they're not going to last, even with $50 million in the bank.

At this point, we need to move on. While we can and will suffer with Intercom for a while, we will either develop our own chat platform on Open Fire or figure out how to use Facebook Messenger.