After the Super’s previous articleon Perplexity’s lengthy article, it was officially announced that they completed a Series B funding of $73.6 million, with over 10 million monthly active users, providing over 500 million query services in 2023!

Perplexity has achieved a significant disruption in a market long dominated by Google using AI. This article will take you to explore the high standards of AI user experience behind Perplexity, revealing the design philosophy hidden behind the code.

The inspiration came from seeing a blog post abroad that analyzed Perplexity’s interaction design using Jakob Nielsen’s 10 general principles for interaction design from 1994, which was fantastic.

Following its content, I further found Jakob Nielsen’s 10 user experience design principles. These principles transcend the so-called CUI, GUI, LUI debate, all of which are linked at the end of the article, and I recommend everyone to take a look:

Each chapter of this article consists of three parts:

1. Translation of the original blog content
2. Specific principles by Jakob Nielsen
3. Supplementary thoughts by Super Huang

01 Visibility of system status (Let users see what you are doing)

Design should always inform users of the current progress through appropriate feedback within a reasonable time.

When users perceive performance delays, it is crucial to show them that the system is still working. Ideally, you can shorten the waiting time by improving system performance. However, this is not always feasible, so progress indicators become a basic expectation.

Perplexity does this better; it not only shows users the exact working status of the system while searching for the requested information but also displays text like “Considering 8 sources” or “Researched and summarized”, allowing users to see not only that the system is working but also what it is doing and how this new technology operates. As a result, users develop more trust in the technology itself. Users gain a deeper understanding of the technology, making it more familiar and likely to be reused.

Jakob Nielsen:

When users understand the current system status, they will know the results of their previous interactions and determine the next course of action. Predictable interactions can build trust in the product and brand.

Tips:

• Clearly convey the system’s status to users – actions that affect users should not be taken without notifying them.
• Provide feedback to users as soon as possible (preferably immediately).
• Build trust through open and continuous communication.

Super Huang:

For example, an elevator: when you press the up button, it lights up, as if saying,