I was initially drawn to the simplicity of this tool which shows a lot of promise for using chatbots through text and video interactions. I also like its unique aspect of using speech recognition to respond to chatbots. However, it misses the mark in several key areas:

1) Video chatbots are muted if set to open by default. While this is due to browser restrictions that have been around since 2018, a very simple fix used by similar tools is to allow the learner to click on a button over the video before it plays which would also unmute it. No one wants to see an avatar speaking without audio upon first visiting a site and then clicking unmute midway through the message.

2) It does not allow speech recognition for basic pre-programmed video and text chatbots, only more complex AI-based conversations that access knowledge files but also quickly use up your monthly credits. This is a huge limitation in my opinion since it prevents end-users from replying to basic prompts using their voice without wasting AI credits in the process. In short, the voice recognition option should also be made available for basic non-AI interactions.

3) The transition between each video chatbot response is a black loading screen which is annoying after a few video scenes. Ideally, the previous video scene should remain visible and paused until the next one loads so the transition from one video response to the next is more seamless and natural for the visitor.

4) When viewing a video chatbot on a mobile device that is rotated to landscape mode, the video gets oddly zoomed in and cropped rather than being automatically adjusted to properly fit the screen. This makes the video chatbot’s head appear cut off during the interaction. However, this does not occur in portrait mode.

5) Links in the chatbot cannot be opened in the same tab which means each selected link opens a new tab and leaves the previous chatbot tab open in the background. Most major browsers can easily handle links in iframes opening in the same tab so this was disappointing to see. So, if you intend to direct visitors from the chatbot to another site that will eventually provide a link back to the chatbot (as is my use case), your users will end up with a clutter of previously open chatbot tabs which can slow down their desktop browser.

6) There is no universal setting to remove the LouiseBot branding from all interactions at once. This means you must do this manually for each individual chatbot which can be quite tedious if you are managing a lot of them. What is the benefit of allowing us to use LouiseBot branding on some chatbots but not others?

7) There are no chatbot analytics available and no way to view users’ chat responses which is a big disappointment. The tool also does not offer a lead generation form, though they state that they are currently working on this feature.

8) For some reason, the profile section of the app requires you to enter your birthdate (month, day, and year) which is inconsistent with other GDPR-compliant tools I've used. Interestingly, if you don't visit the profile section, then it won't ask you to submit this information. But once you access this section, you are required to fill this out which concerns me from a privacy standpoint especially when no explanation is provided anywhere on why or how this information is being used.

Overall, I think with a little more time and refinement, LouiseBot could be quite powerful. Therefore, I am happy to revisit this review and update it in the future once the necessary improvements have been made.