IBM watsonx Assistant - Great NLP and NLU but so hard to understand
January 14, 2025
IBM watsonx Assistant - Great NLP and NLU but so hard to understand

Score 1 out of 10
Vetted Review
Verified User
Overall Satisfaction with IBM watsonx Assistant
We use IBM watsonx Assistant together with Twilio to have a complete chatbot that can understand our clients, and also be able to direct with using Twilio Functions. Our problem is that since we first implemented Twilio in our organization, the team that did it, wasn't able to use IBM watsonx Assistant at its full potential, and we have a "dumb" chatbot, that can only understand simple terms or buttons that we created.
Pros
- NLP and NLU
- Regex and simple functions
- STT and TTS
Cons
- Integration with other platforms (Meta WhatsApp or Telephone)
- Serverless
- User Experience
- Positive: NLP and NLU in Twilio
- Negative: Hard to understand and create using IBM watsonx Assistant (needs study)
- Negative: Few integrations with other platforms
- Negative: IBM Watson support team isn't much of a support (at least, in our experience)
The NLP and NLU capacity.
I do not have this knowledge. We pretty much use IBM watsonx Assistant as a way to translate what our clients are saying in order to redirect them to the right flow or direction in our chatbot. With that said, I do know it has a tremendous stack of tools, but, we do not use it.
I selected IBM watsonx Assistant so I could maintain the NLP/NLU processing from our last supplier (which was not Twilio). Since we switched, I have not liked IBM watsonx Assistant so much. We are sincerely thinking about changing it to a new one, the Kore.AI which seems easier to deal with, has a better ranking in Gartner and is cheaper.
Do you think IBM watsonx Orchestrate delivers good value for the price?
No
Are you happy with IBM watsonx Orchestrate's feature set?
No
Did IBM watsonx Orchestrate live up to sales and marketing promises?
No
Did implementation of IBM watsonx Orchestrate go as expected?
No
Would you buy IBM watsonx Orchestrate again?
No

Comments
Please log in to join the conversation