I think the $100 per month limit makes this comparison more interesting because the cheapest option isn’t necessarily going to be the most cost effective one in the long run. I’d be looking beyond the monthly subscription and checking things like usage limits, integrations, how much maintenance is required, and what happens when the chatbot can’t answer a question.For me, the biggest test would be what happens after the initial setup. It’s easy for a chatbot to look impressive when you’re testing a few common questions, but real customers will ask things you didn’t anticipate. Being able to review those conversations, fix the knowledge, test changes, and improve the agent without rebuilding everything is really important.I also think human handoff deserves more attention. The goal shouldn’t necessarily be to automate every conversation.

• This reply was modified 1 month, 2 weeks ago by  Gilbert.