Gemini Workspace AI Opportunity Map
A PM opportunity map for Google Workspace's AI surface — public user feedback across Reddit, HN, Stack Overflow, App Store, and YouTube, analyzed for theme and competitor mentions.
The problem
Google ships Gemini across every Workspace surface — Gmail, Docs, Sheets, Meet — but the public conversation about whether it's working is scattered across forums Google doesn't own. I wanted to build the opportunity map a PM on Workspace AI would actually use on Monday morning: what are users saying, where do they say it, and what do they wish it did that ChatGPT already does.
My hypothesis
If I pulled curated public feedback across Reddit, Hacker News, Stack Overflow, the App Store, and YouTube, then tagged each post to a theme (accuracy, grounding, memory, tone, integration) and flagged competitor mentions, I'd end up with a dashboard that shows where Gemini loses — and to whom — in the users' own words.
What I built
A dashboard at schlacter.me/workspace-ai-gaps with three modules: theme distribution (which frictions dominate), competitor callouts (ChatGPT, Copilot, Notion AI — with the specific reasons users prefer each), and a raw-feedback viewer where every stat drills into the source posts. A methodology page documents sourcing, dedup, and theme mapping.
What broke
Competitor mentions are directional, not quantitative — Reddit/HN audiences skew technical and early-adopter, so ChatGPT and Copilot over-index. A real version of this analysis would need Workspace's internal CSAT and support ticket data alongside the public signal. The dashboard is honest about this; I'd rather be explicit than let a reader assume representativeness.
What I learned
The competitor reasons matter more than the counts. 'ChatGPT is more reliable on grounding' and 'Copilot has unified memory across M365' are specific enough to act on — they name a product gap, not just a sentiment. Those quotes are where a Workspace PM could start a roadmap conversation.
If I kept going
Refresh the corpus weekly via scheduled scrape. Add a fourth module — 'Shipped vs. Requested' — comparing the public feedback themes against Google's Workspace AI release notes to see what's been addressed. Also widen to include Twitter/X and LinkedIn signal for enterprise-buyer perspective.
Prototype
Dashboard hero: theme distribution across 5 public sources.
Competitor callouts: ChatGPT, Copilot, Notion AI — with the specific reasons users prefer each.
Raw feedback viewer: every stat drills into the source posts.