When compared to other platforms based on 5 different types of prompts, we can see some strengths and weaknesses emerge for Claude.
- Claude rates better than Pulse IQ average on “Factual” prompts (23%)
- Example: “In which year did the Apollo 11 moon landing take place?”
- Claude rates a bit behind the Pulse IQ average on “Emotional” prompts (14%)
- Example: “What are some techniques for managing and expressing anger in a healthy way?”
Here are some quotes representative of the user feedback when users talk about their Claude experience:
“It blew me away by answering a simple question with a very smart answer consisting of knowledge I didn't know existed.”
“I really like Claude. It has almost replaced chat gpt for me.”
“I liked that you could use Claude to compose business templates and tasks and I liked the feature of being able to upload documents to Claude for reference and the speed and accuracy of the responses.”
- A smooth UI
- Comparable to GPT
- Allowed for attaching files and getting feedback
- Well-formatted responses
- Length of response was longer than expected
- Felt slower than GPT
Companies that are expected to provide factual information first and foremost (finance, travel, medication advice, electronics), should lean in and take a closer look at Claude given its high marks.
Moreover, the feature of sending files to AI for feedback means that users will be able to reduce what they convey in written form by attaching files instead. Applications of this feature could extend to use cases such as:
- Large electronics (where the space size and dimensions are necessary to get the right recommendation)
- Health information (weight, age, BMI) although it remains to be seen what users are willing to share (see story below about personalization).
- Learning, education, homework where students are expected to be able to analyze data and/or provide a story from a data set.
- Fashion and clothing where the imagery is a key part of the idea and the user’s goal.
- Gardening where size and light are key elements and a user wants to know what will be appropriate to the area / location.