When you start a search on Perplexity, the platform groups your queries into a thread. Each thread keeps the context of your earlier questions, so you can ask follow-up questions without repeating the original topic. This feature saves time and produces more relevant answers because Perplexity remembers what you just discussed. In this article, you will learn how to ask follow-up questions, how threading works, and how to manage your threads for better results.
Key Takeaways: Managing Follow-Up Questions in Perplexity
- Thread context retention: Perplexity keeps the last 2,000 words of conversation in a thread to answer follow-ups accurately.
- Follow-up button or Enter key: Use the dedicated follow-up field below each answer or press Enter after typing your next question.
- Thread rename and delete: Click the thread name to rename it, or use the three-dot menu to delete a thread from the sidebar.
How Perplexity Threads Handle Follow-Up Questions
A Perplexity thread is a sequence of questions and answers on the same topic. When you ask a follow-up question, Perplexity uses the entire conversation history to generate the next answer. The platform stores the last 2,000 words of the thread as context. This means you can ask questions like “What are the main benefits?” or “Tell me more about the second point” and the AI will know what you are referring to.
Threads appear in the sidebar on the left side of the Perplexity web interface. Each thread is automatically named based on the first question you asked. You can have multiple threads open at the same time, and each thread keeps its own context separate from others. This is useful when you are researching different subjects simultaneously.
Follow-up questions work in any focus mode: Web, Academic, Writing, Math, or Video. The context retention applies across all modes. If you switch focus mode mid-thread, the AI still remembers the earlier discussion and adjusts its answer style accordingly.
Steps to Ask a Follow-Up Question in a Thread
The process for asking follow-up questions is identical on the web version and the mobile app. Follow these steps.
- Open or start a thread
Go to Perplexity.ai and log in. Type your first question in the search box at the top of the page. Press Enter or click the search icon. The answer appears, and a new thread is created in the sidebar. - Locate the follow-up field
Scroll to the bottom of the answer. You will see a text box labeled “Ask a follow-up” or simply an empty input field directly below the answer. On mobile, this field is at the bottom of the screen after the answer. - Type your next question
Enter a question that builds on the previous answer. For example, if your first question was “What is quantum computing?” your follow-up could be “What are the current limitations of quantum computers?” - Send the question
Press Enter or click the send arrow icon next to the input field. Perplexity processes the question with the thread context and displays a new answer below the previous one. - Review the thread history
All questions and answers remain visible in the thread. You can scroll up to see earlier exchanges. To return to a thread later, click its name in the sidebar.
Asking Follow-Ups on the Mobile App
The Perplexity mobile app for iOS and Android works the same way. After you get an answer, tap the text input field at the bottom of the screen. Type your follow-up question and tap the send button. The app appends the new answer to the thread. You can swipe left or right to switch between threads in the app.
Using Voice Input for Follow-Ups
On the mobile app, you can use voice input to ask follow-up questions. Tap the microphone icon in the input field. Speak your question. The app transcribes it and sends it as a follow-up. This works best in quiet environments for accurate transcription.
Common Mistakes and Limitations With Follow-Up Questions
Perplexity Forgets the Context After a Long Conversation
If your thread exceeds 2,000 words, Perplexity drops the oldest parts of the conversation. The AI may lose track of what you asked early on. To avoid this, start a new thread when the conversation becomes long. You can also summarize the key points in your next question to refresh the context.
Follow-Up Questions Produce Off-Topic Answers
When you ask a follow-up that is too vague, Perplexity may misinterpret your intent. For example, asking “How does that work?” after a long answer about climate change might confuse the AI. Be specific: “How does carbon capture technology work?” yields a better result.
Threads Disappear From the Sidebar
Perplexity saves threads only for logged-in users. If you are browsing without an account or if your session expires, threads may not persist. Always sign in before starting research. You can also export a thread by clicking the share icon and copying the link.
Cannot Ask Follow-Ups in Copilot Mode
Copilot mode, which provides more guided answers, does not support follow-up questions in the same thread. Copilot answers are standalone. To continue the conversation, exit Copilot mode and use the standard search or ask a new question in a fresh thread.
Perplexity Free vs Pro: Thread and Follow-Up Features
| Item | Perplexity Free | Perplexity Pro |
|---|---|---|
| Context retention | 2,000 words per thread | 2,000 words per thread |
| Number of threads | Unlimited | Unlimited |
| Follow-up field | Available | Available |
| Voice input for follow-ups | Available on mobile | Available on mobile |
| Thread export (share link) | Available | Available |
| Pro-only models for follow-ups | No | GPT-4, Claude 3, Gemini Pro |
The table shows that follow-up question features are identical between Free and Pro accounts. The main difference is which AI model answers those follow-ups. Pro users can select a more powerful model like GPT-4 or Claude 3 for deeper analysis. Free users get the standard Perplexity model, which is sufficient for most everyday questions.
You can now ask follow-up questions in any Perplexity thread by using the input field below each answer. Remember to keep your questions specific and start a fresh thread when the conversation grows long. For complex research topics, switch to a Pro model to get more detailed follow-up answers. A good next step is to try asking a three-question thread on a topic you are learning about to see how the context builds across answers.