Google Workspace Studio Tutorial: Auto-Organize Your Inbox with Smart Labels & Priority Notifications — DeepSeek Blog | Neura Market
    Neura MarketNeura Market/DeepSeek
    ChatGPTChatGPTClaudeClaudeGeminiGeminiCursorCursorGrokGrokPerplexityPerplexityDeepSeekDeepSeek
    CoPilotCoPilotStable DiffusionStable DiffusionMidjourneyMidjourney
    View All Directories
    OverviewRulesPromptsMCPsAgentsBlogVideosGuidesCoursesCommunityTrendingGenerate
    DeepSeekBlogGoogle Workspace Studio Tutorial: Auto-Organize Your Inbox with Smart Labels & Priority Notifications
    Back to Blog
    Google Workspace Studio Tutorial: Auto-Organize Your Inbox with Smart Labels & Priority Notifications
    workspacestudio

    Google Workspace Studio Tutorial: Auto-Organize Your Inbox with Smart Labels & Priority Notifications

    Aryan Irani April 2, 2026
    0 views

    The average professional spends 28% of their workday reading and answering emails. But here's the...

    The average professional spends 28% of their workday reading and answering emails. But here's the problem, not all emails are created equal. A Pizza Discount doesn't deserve the same mental space as a Project Delay notification from a CEO. So, I decided to fix it. I used [Google Workspace Studio](https://studio.workspace.google.com/) to build what I call an Email Organiser. It's an AI agent that actually reads my mail, labels it correctly, and - most importantly - only pings my Google Chat when something is actually on fire. Here is how I built it (and how you can too, without writing a single line of code). > If you missed my previous tutorial on the Meeting Prep Agent, start there for the [basics](https://medium.com/google-cloud/google-workspace-studio-tutorial-building-an-ai-meeting-prep-agent-6f27dcbcbdb5). Prefer to watch instead of read? Check out the video version of that guide [here](https://www.youtube.com/watch?si=jJqteoifcJRjJfFr&v=onNH6yJNtMs&feature=youtu.be). Prefer to watch instead of read? Check out the video version of the guide here. {% embed https://youtu.be/YLFYu_9dkw4 %} What Is Google Workspace Studio? (Quick Recap) **Google Workspace Studio** is Google's no-code AI agent builder (launched December 2025). It lets anyone describe a workflow in natural language, and Gemini builds the entire automation using your Gmail, Drive, Chat, Sheets, and more. **Key Features Used Here:** - New email starter - Gemini-powered classification - Conditional actions (Decide step) - Add labels + Send Chat message **Prerequisites** - Google Workspace Business, Enterprise, or Education plan with Gemini enabled - Access to studio.workspace.google.com - (Optional) Pre-create these Gmail labels: Urgent, Client, Internal, Marketing/Promo, Low Priority, Spam ## Step 1: Add the Starter (Trigger) Open **studio.workspace.google.com** and start a new flow. Name it something like **AI Email Labeller + Priority Sorter**.  - **Starter:** Select When I get an email. - **Settings:** Keep it on All emails. We want this agent to act as our first line of defense for every message. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/vnn9z2588mt7um9ybxkn.png) ## Step 2: The Logic Breakdown (The Extract Action) After adding the When I get an email starter, the next crucial step is to let Gemini analyze the incoming email. In Workspace Studio, we use the **Extract step (powered by Gemini)** to intelligently process the email content. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/13myo901ljtmew4754mp.png) In this step, we aren't just asking Gemini a question; we are using the **Extract** tool to pull specific **metadata** out of the raw email body. To make this reliable, we need to provide a **rubric** so the AI knows exactly how to judge your mail. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/vylukf7ob0v8u1pol25m.png) I've pointed it at the **Email Body** from Step 1 and created two custom fields. Think of this as giving Gemini a Rubric to follow. Instead of guessing, the AI is now acting as a professional assistant following a strict set of rules. Here is how I configured the two custom content fields to make them Production-Ready: **1. Custom Field: Is High Priority** - **The Goal:** A simple true/false result that we can use later to trigger a notification. - **Description for Gemini:** > Analyze the sentiment and urgency. Return 'true' ONLY if the email indicates a project-blocking issue, an immediate deadline (today/tomorrow), or a direct request for a deliverable from a known client. If it is a general update, a newsletter, or a 'thank you' note, return 'false'. Do not add any punctuation or extra text. **2. Custom Field: Classification** - **The Goal:** This will be used to apply actual labels in your Gmail sidebar. - **Description for Gemini:** Classify the email into exactly one of these categories: urgent, client, internal, marketing, newsletter or other. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/aheu3n14435ctd2m5fec.png) ## Step 3: The Organization (Physical Labelling) Now that Gemini has categorized our email in its mind, we need it to physically label the email in our Gmail inbox. This is where Step 3 comes in: the Add labels action. 1. **Mapping the ID:** First, we point the agent to the right email. In the **Email to label** field, use the **+ Variables **button to insert the**[Step 1: Email ID]**. This ensures the agent is acting on the exact message that triggered the flow. 2. **The AI-Powered Toggle:** Notice the **AI-powered labels** switch. When you turn this on, you aren't just picking a static folder; you are giving Gemini permission to apply labels based on a description. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/vu6eysqe3x7fp80h8xdd.png) While Google provides great templates like Receipts or Meeting updates, the real power lies in the **New AI-powered label** button. This is where you can create a label that perfectly matches your workflow. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/11j6b5dn98ap78mm2h14.png) For example, I created a custom label called To Respond. Instead of a simple keyword search, I gave it this description: > Prioritize emails that require my direct action. This includes messages that ask a question, request a deliverable, need my review or approval, or are pending my response. > Crucially, exclude the following: > 1. All automated notifications and system-generated emails (e.g., Drive file access requests, comment threads, or changes to shared documents). > 2. All Google Calendar updates (accepted, declined, rescheduled, canceled) and meeting confirmations. > 3. Broad announcements, newsletters, marketing, or promotional emails. > 4. Informational notes that do not require a response, such as "thank you" messages. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/hed9196j8wl2vnqig1zr.png) By using custom descriptions, your labels become much more accurate than traditional Gmail filters. You aren't just looking for "Meeting" - you're looking for the intent to meet. ## Step 4: Add the Decide Step (High Priority Check) After classifying the email and applying the label, we need to decide whether this email is important enough to notify us in Google Chat. This is where the Decide step comes in. 1. Click + Add step → Select Decide. 2. In the prompt field, paste the following: Is this email high priority? 3. Click the + Variables button and insert Step 2: Is High Priority. The full prompt should now read something like: ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/m0l8tgdmaeulw1olys39.png) ## Step 5: Check if the Decision is True The **Decide** step only makes a decision - it doesn't automatically route the flow. We need a **Check if** step to act on that decision. 1. Click on Step 5: Check if 2. Set the condition as follows: - **If** → Select **Step 4: Decision** - **is** → **true** ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/miwiloesewptxzttzeaq.png) This creates two paths: - **True branch:** High-priority emails (Urgent or Client) → We will send a Chat notification. - **False branch:** Normal emails → No notification (they only get labeled). This branching logic is what makes the agent smart - it only bothers you for important emails. ## Step 6: Add Notify Me in Chat Now we add the final action that only runs for high-priority emails. ### How to Configure Step 6: 1. Make sure you are on Check if step (Step 5). 2. Click **Add substep** → Select **Notify me in Chat**. 3. In the **Message** field, create a clean and useful notification using variables. You can add or remove variables as needed. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/vy3l6i0rmhl1t3mjtq2s.png) ## Testing and Activation Now that the full flow is built, it's time to test it with real emails. I sent two very different emails to my inbox. I wanted to see if Gemini could distinguish between an important email and a casual request. ### Case 1: High Priority I sent an email regarding **Project Quantum - Final Contract Feedback & Signature Required**. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/efsxid2ratlxzcckhsmb.png) Back in the studio, go ahead and click on **Test run** and select the email you would like to work on, which in this case is the **Project Quantum** email. Once selected, click on **Start**. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/mwddr4i6aimxpd0kpgwt.png) - **The Content:** It mentioned a London hub deployment and a deadline of today. - **The Result:** Gemini nailed it. In the **Test Run**, it identified that a signature was needed (Step 2 extraction: true) and classified it as a client request. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/d88vdd2fh6p6k51nf9e2.png) - **The Action:** Because the priority was true, the **Check if** logic passed. My phone immediately pinged with a Google Chat notification: 🚨 High Priority Email Received. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/ym7j210yp8xfofkmuy1e.png) ### Case 2: The Important but Quiet (Low Priority) Next, I sent a follow-up: **Question regarding the API Documentation**. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/18zmtkuerdywla9h23xb.png) Back in the studio, follow the same steps as we did previously to test the second email. - **The Content:** Sarah had a few questions but explicitly said, **No rush at all.** - **The Result:** Gemini recognized that Sarah was asking a question, so it applied the **To respond and Support** label in my Gmail sidebar . However, it also read the No rush sentiment and set priority to false. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/f2noeown6uvykuikjpbm.png) - **The Action:** The **Check if** logic stopped the flow. No Chat notification. My focus remained unbroken, but the email was waiting for me in the right folder when I was ready for it. ![Image description](https://dev-to-uploads.s3.amazonaws.com/uploads/articles/3l1pqbf6nufnicz3r8n3.png) **Going Live (Turn On)** Once you are happy with the test results, click the blue Turn on button. Within a few seconds, you'll see a confirmation snackbar at the bottom left: > Your flow is on and ready to work! From this point forward, the AI agent is my primary filter. It organizes the chaos in the background and only interrupts me when the work truly demands it. **Note:** The agent only acts on new emails received after it is turned on. Existing emails in your inbox will not be processed retroactively. ## Conclusion You now have a fully functional **AI Email Labeler + Priority Sorter** running in your Gmail. It intelligently classifies every incoming email, applies the right label, and only notifies you when something genuinely requires your attention - giving you back hours of focus every week. This is the real power of Google Workspace Studio: turning repetitive manual work into intelligent, automatic systems. ### What's Next? This is only the beginning of our journey with Google Workspace Studio. I have a full roadmap of content coming your way to help you master every corner of this platform: - 🚀 **More Workflows:** Stay tuned as we build dedicated AI agents for **HR Onboarding**, **Sales Lead Management**, and **Automated Expense Tracking**. - 🧠 **Deep-Dive Tutorials:** I'll be releasing specific guides on mastering Variables, optimizing your **Starter triggers**, and advanced **Prompt Engineering** to make Gemini even smarter. - 📺 **Follow the Youtube Playlist:** You can catch every click-by-click build and all upcoming tutorials in my [Google Workspace Studio Tutorial Series Playlist](https://youtube.com/playlist?list=PL_MCVBMm-9spp6kmOPkrJEgDVSyhmPWm7&si=D2aX8uev7pHyYe-C) on YouTube. Feel free to reach out if you have any issues/feedback at [email protected].

    Tags

    workspacestudiogoogleworkspaceaiemailogranisergooglecloud

    Comments

    More Blog

    View all
    How I'm using ASTs and Gemini to solve the "Codebase Onboarding" problem 🧠ai

    How I'm using ASTs and Gemini to solve the "Codebase Onboarding" problem 🧠

    Hi everyone! 👋 I’m Tara, a Senior Software Engineer and Consultant. Over the years, I've jumped...

    T
    tworrell
    Local AI Will Save Us All (The Math Says So, Trust Me)ai

    Local AI Will Save Us All (The Math Says So, Trust Me)

    Every few weeks a take goes viral in tech circles making the case for ditching cloud AI and running...

    S
    Sebastian Schürmann
    Lost in the AI Hype, I Started Smallai

    Lost in the AI Hype, I Started Small

    And it helped me get back into tech without drowning TL;DR at the end Coming back to...

    R
    Rohini Gaonkar
    Building a Replay-Tested Interactive Brokers Client in Gogo

    Building a Replay-Tested Interactive Brokers Client in Go

    I wanted an IBKR library that felt like Go and had testing I could trust. So I wrote one.

    T
    Thomas Marcelis
    Playwright in Pictures: Fully Parallel Modeplaywright

    Playwright in Pictures: Fully Parallel Mode

    Playwright’s fullyParallel mode is often treated as a simple performance switch. In practice, it...

    V
    Vitaliy Potapov
    Designing a CLI for Both Humans and Agentscli

    Designing a CLI for Both Humans and Agents

    Learn how Alpic designed its CLI for both human developers and AI agents — covering tradeoffs like polling, context windows, interactivity, and statelessness.

    J
    Julien Vallini

    Stay up to date

    Get the latest DeepSeek prompts, rules, and resources delivered to your inbox weekly.

    Neura Market LogoNeura Market

    Discover the best AI prompts, plugins, and resources for DeepSeek and more.

    Content Types

    • Rules
    • Prompts
    • MCPs
    • Agents
    • Guides

    Platforms

    • ChatGPT Directory
    • Claude Directory
    • Gemini Directory
    • Cursor Directory
    • Grok Directory
    • Perplexity Directory
    • DeepSeek Directory
    • CoPilot Directory
    • Stable Diffusion Directory
    • Midjourney Directory
    • All Directories

    Resources

    • Blog
    • Documentation
    • Help Center
    • Marketplace

    Legal

    • Privacy Policy
    • Terms of Service

    © 2026 Neura Market. All rights reserved.

    |

    Not affiliated with any AI platform vendors.