Beyond the prompt: Mastering the “deep end” with generative AI

Google Workspace with Gemini isn’t just a tool I use at work — it’s the product I evangelize as a product marketing manager at Google. When I recently took this role, I saw an opportunity to use Gemini to learn a new job and prepare for one of the largest tech conferences of the year, Google Cloud Next. To do this, I had to quickly embrace the “deep end” and find smart ways to amplify my knowledge and impact using Gemini. 

Overall preparation

I started by creating my own Gem, which is a custom version of Gemini that I can use for repetitive tasks such as building content, summarizing, copywriting, brainstorming, and more. The Gem — “Product Marketing Manager helper” — became my daily assistant, checking my work along the way. I uploaded our positioning and target audience and preferred formats to the Gem. Then, submitted panel descriptions and scripts and it offered up rewrites, making my content more compelling and targeted so that we could land our key messages at Next. 

Research 

To prepare for Next, I had to craft a compelling product demo for our business customers. To create a relatable demo, I kicked off my research in the Gemini app with the following prompt: “Help me develop 10-15 Gemini use cases for various industries, including manufacturing, telecom, software, and professional services. Consider various professional roles including sales, marketing, operations, financial management, and customer service.” Then I continued the conversation with Gemini to refine the list.                                                                                  

Create and refine

I also had to create a breakout session focused on the value customers get from using Workspace with Gemini. I used Help me create in Google Docs to develop my session presentation outline, session title, and description. 

First, I created a prompt that included:

• Session goals and audience, as well as the title and description we’d already developed.

• Supporting materials (sales decks, marketing positioning frameworks).

• Content planning documents (such as guidelines for presentations, length, and time).

• The panelists we were going to work with and their titles, to help brainstorm potential panel questions.

Gemini drafted an outline that was 90% of the way there. I could link to all the files in my Google Drive to keep the response grounded in trusted sources. Having that context helped eliminate a lot of back and forth to refine the content, even though it took a bit more time on the front end.               

Then I used Help me write in Google Docs to refine my outline. I requested specific changes to polish the writing and Gemini elaborated or shortened the content depending on my request.  Gemini helped me create a relevant and tailored outline with a bank of panel questions ready to share with the team.

Present                                            

Next, I created a Google Slides deck using the refined outline. Gemini in Slides helped me develop talking points. I referenced my document outline and asked Gemini to build out a script to accompany the slides I pulled together, then used Gemini to create speaker notes. Gemini even offered pacing suggestions, including how much time to spend on each slide. This was a great start to share collaborating with my colleague who was delivering the session. Now he could take it and make it his own.

Share

After the conference, I used NotebookLM to help consolidate insights around what happened during all of the breakout sessions and what was announced. This AI-powered tool builds unique information hubs by drawing on sources uploaded to a notebook.

I loaded all the conference content, including:

• Product announcements

• Keynote session summaries

• Panel transcripts

• YouTube videos

• Press and news reactions

NotebookLM was the perfect way to consolidate the materials. Now my team can get the information they need without searching through multiple platforms, documents, or recordings. They can quickly get answers to questions like: “What were the main announcements around Security at Next?” This mind map shows an example breakdown if you upload information on Google Cloud Next 25 then generate a Mind Map.

Shift your approach

Using gen AI tools across Google Workspace helped me make the most of my onboarding period and Next 2025, saving me a lot of hours across the whole process as I got ready for the event. It’s also transformed the way I approach most complex, new projects. Workspace with Gemini helps me tackle complex problems and get additional insights, so I can work smarter.