How to use Perplexity for work comes down to searching with citations: you ask a question, Perplexity searches the web and adds numbered sources to every answer. This Perplexity guide explains how to get started, how Perplexity builds its answers, which features it has and how to check the sources.
Perplexity runs on models from OpenAI and Anthropic, among others, which places it among the tools built on the major AI models. Where Perplexity stands out is in searching and backing up answers with sources.
How to use Perplexity: getting started
To use Perplexity, open perplexity.ai in your browser or install the app, and type your question in plain language. You can ask questions without an account; with a free account you keep your search history and get a limited number of Pro Searches per day.
You can run your first research question in Perplexity in four steps.
- Ask a specific question with context, for example: “Which software vendors offer digital case file management for notaries in the Netherlands, with a price indication and a source?”
- Choose Pro Search or Deep Research when the question needs more sources.
- At the very least, open the sources behind the figures and names you want to use.
- Ask follow-up questions in the same conversation to refine the overview.
How Perplexity works
Perplexity works as an answer engine: it searches the web, summarises the relevant information and adds numbered source references to every answer. That lets you trace a claim back to the page it came from.
A source reference does not guarantee that the summary is correct. A summary with a source sometimes still differs from what is on the page, a form of AI hallucination you only find by opening the source yourself.
When Perplexity fits your work
Perplexity fits work where you quickly need a well-sourced overview, such as market research, comparing vendors or looking up current regulations. For longer texts or work in your own documents, ChatGPT, Claude or Gemini are the more obvious choice.
That balance between searching and writing is exactly where the difference between Perplexity and ChatGPT lies.
Perplexity’s features
For professional use, Perplexity consists of four parts.
Pro Search and model choice
Pro Search searches more sources for each question than a standard search. In the free version you have a limited number of Pro Searches per day; with Perplexity Pro you get wider access and choose the model yourself, or let Perplexity decide for each question.
Deep Research
Deep Research in Perplexity runs dozens of searches, reads hundreds of sources and delivers a report in two to four minutes, according to Perplexity’s help centre.
Comet
Comet is Perplexity’s browser for Windows, Mac, iOS and Android, with an assistant that carries out tasks in the browser. Perplexity’s own examples include answering email, comparing news stories and planning a trip.
Perplexity Computer
Perplexity Computer is a cloud-based agent, launched in February 2026, that splits a task into subtasks and distributes them across different AI models. At launch, Computer was available only to Perplexity Max subscribers.
What to watch for with work data in Perplexity
With work data in Perplexity, the same rule applies as with any AI assistant. Do not process client personal data in a personal account without a legal basis and a data processing agreement with the provider. So do not put client names or case details in a search query.
Comet and Computer act on your behalf on websites and in apps. Only give them access to accounts where you can oversee the consequences of a wrong action.
Learning more about Perplexity
You learn more about Perplexity one feature at a time, with the articles in this guide covering each one in detail. Perplexity appears with the other options in the overview of ChatGPT alternatives, with what each tool is good at.
For teams that use Perplexity for research, the AI Literacy Course covers which data you can enter into an AI tool, in light of the GDPR and duties of confidentiality. It is vendor-neutral and ends with a certificate.
