[nullIsAnObject]

The biggest problem with chatbots is that getting conversations right is hard... and expensive. To make one actually helpful, it requires extensive knowledge not just about the potential topics that customers could ask about, but also linguistics, NLU/P/Q, scripting, and the diverse technical environments that these tools will be interacting with. With all the drag-and-drop chatbot creation tools out there now, companies are being sold on the idea that ANYONE can make these chats, and as such don't see the benefit of hiring an experienced professional whose sole job it is to make these things work well. Instead, their creation is being left to existing help desk/HR/doc employees that lack said in-depth cross-functional knowledge, with predictable results. Sparse NLU models are poorly constructed, background tasks are hacky and frequently broken, and the resulting chats are stilted and disjointed at best. Which is why they're so frustrating.

The tools to make good chatbots are complex, but they exist, and in the hands of someone who is trained in all of the requisite areas can be wielded to create something useful. It just costs money most companies don't believe is worth spending.