Gemini Deep Research
Visual Walkthrough
Step-by-step guide to running a Deep Research project inside Google Gemini

Google Gemini Dashboard: This is the main Gemini interface at gemini.google.com, where you can access different AI tools.

Opening the Drop Down Menu: Click the dropdown menu to reveal all available features and models for Gemini.

Selecting “Deep Research”: Locate and click Deep Research to launch Google’s agent-based research tool.

Deep Research Page: You’re now inside the Deep Research workspace, ready to enter your prompt.

Give Gemini a Deep Research Prompt: Type your detailed research request into the input field.

Starting the Research Process: Click Enter to begin — Gemini will prepare its research plan.

Gemini’s Research Plan Dialog: Gemini presents a plan showing how it will approach your request.

Reviewing the Research Plan: Read the plan carefully to confirm it
covers all your needs.

Editing the Plan if Needed: If something is missing, make changes
before proceeding.

Confirming “Start Research”: Once satisfied, click Start Research to
launch the full process.

Choosing to Show Thinking or Browsed Sites: Gemini can display its reasoning or the list of websites it browses.

Research Process in Action: The canvas begins filling with Gemini’s thoughts and findings.

Websites Researched Count: On the left, you’ll see the number of
websites Gemini has processed.

Live Website Feed: On the right, you can watch the actual sites Gemini is reviewing in real time.

Research Complete - Analysis Begins: When research is finished,
Gemini begins analyzing the collected data.

Final Research Report Ready: The completed findings are now
displayed in the workspace.

Copying the Report: Click Copy to paste the content into another document to modify, store, etc.

Exporting to Google Docs: Use Export to send the report directly into
your Google Docs account.

Viewing the Report in Docs: Open Google Docs to see the full
exported report.

Generating an Audio Overview: Click Generate Audio Overview to create a spoken summary of the research.

Audio Overview Completed: Gemini delivers a playable audio version
of your findings.

Deep Research Process Complete: Your text and audio reports are now ready to use or share.
A Guide to the Deep Research Feature in Google Gemini
The text outlines a comprehensive guide on utilizing the Deep Research feature within Google Gemini, which is specifically designed to enhance agent-based searches across various websites, synthesizing information based on user prompts.
Users begin by accessing the Gemini platform through a designated URL. They must locate the drop-down menu and select the "Deep Research" option. This feature enables users to input a specific prompt that Gemini will use to conduct its research. After entering the prompt, users are instructed to click "Enter" to start the research process.
Once the prompt is submitted, Gemini analyzes it and formulates a detailed plan to address the request. This plan is essential as it delineates the strategy Gemini will employ to gather relevant information. Users are encouraged to scrutinize this plan carefully, as it allows them to identify any potential omissions or areas needing further clarification. If the plan aligns with the user's expectations and encompasses all necessary elements, they can proceed by clicking "Start Research."
After initiating the research, a new window opens where Gemini reveals its thought process. Users have the option to view the specific websites that Gemini is accessing to collect information, although the default view displays Gemini's reasoning. The interface is organized with a left section indicating the number of websites being researched and a right section that lists these websites in real-time.
The research process is not instantaneous; users are informed that they can step away while Gemini conducts its analysis. Notifications will be sent to users to indicate that the analysis is in progress. During this phase, it is suggested that users pause any accompanying video content and return once the research is complete.
Upon completion of the research, users have the option to copy the report directly into a document or export it to their Google Docs account, which is linked to their Google Gemini email. The export function allows for a seamless transfer of the report into the user's Google Docs. Additionally, there is an option to generate an audio overview of the report, providing an auditory summary of the findings.
The guide concludes with a note of appreciation and an invitation for users to explore further instructional content in upcoming videos.



