Gemini Enterprise also includes:
- A no-code workbench for process automation and agent orchestration, alongside pre-built Google agents for specific functions such as research and data analysis, as well as custom agents from partners and those the customers build themselves.
- Out-of-the-box connectivity to enterprise data sources, including Google Workspace, Microsoft 365, Salesforce, and SAP. The Gemini plans included in Google Workspace only connect to Google Workspace.
- A central governance framework for agent visibility, security, and auditing, and supports an open ecosystem for third-party integrations.
Why it Matters
Google’s move to embed Gemini, previously a USD20 per user per month add-on, into Workspaces in 2024 was seen as a measure to drive AI adoption in a market that was more resistant than expected. Microsoft was, and to some extent still is, suffering this resistance with Copilot for 365. The underlying challenge is that prompt-based AI services either fail to deliver the expected productivity gains or cannot easily demonstrate them for businesses. While there are examples of productivity gains, many organisations struggle to measure them.
However, organisations are increasingly investigating the practical application of agentic AI – where generative AI is linked to business data and leveraged to create semi-autonomous services. This is where Google hopes its AI efforts will start making money. Hence, Gemini Enterprise is a positioning of AI that enables business value to be more easily materialised.
What’s the difference?
When comparing Gemini Enterprise with current Google Workspace Gemini offerings, a primary distinction is the scope and customisation.
Current Gemini integrations within Google Workspace typically provide AI assistance within standard productivity applications, such as generating content or summarising documents.
Gemini Enterprise, by contrast, emphasises workforce agentic services. The ‘no-code workbench’ and the ability to orchestrate agents is aimed at capturing general workers’ desire to automate daily activities. The secure connection to diverse enterprise data sources (Workspace, Microsoft 365, Salesforce, SAP) is crucial for real-world applications, as AI agents require access to proprietary data to be effective. However, such connections need to be carefully monitored and governance put in place, so Gemini Enterprise offers a central governance framework to visualise, secure, and audit all agents from one place.
Compared with Microsoft’s Copilot for Microsoft 365, which also leverages AI agents within the Microsoft ecosystem, there are architectural and integration differences. Microsoft Copilot for Microsoft 365 is priced at USD30 per user per month and supports a wide range of applications within the Microsoft 365 suite. Its strength lies in its deep integration with Microsoft applications, allowing users to interact with documents, emails, and presentations using natural language.
While Microsoft’s offering primarily focuses on augmenting Microsoft 365 applications, Google’s Gemini Enterprise emphasises broader data source connectivity and custom agent development through its no-code workbench. Google is opting for a more horizontal approach, aiming to integrate AI across a range of enterprise systems, not just its productivity suite. It needs that horizon. Google Workspace has far less market share compared to Microsoft 365 in Australia. Without domestic data centres for its Workspaces and related Gemini services, it is only likely to go backwards in the market as more organisations seek to meet their own, or their customers’, data sovereignty demands. However, IBRS notes that Google has a strong global position and a strong presence in Australian K-12 education. The inclusion of broad Gemini services within the Workspaces suite will stand it in good stead.
Who’s Impacted?
- Heads of Digital Transformation: Focused on leveraging new technologies to automate processes and drive operational efficiencies across various departments.
- Data Governance and Security Teams: Need to ensure compliance, data privacy, and secure access for AI agents across diverse data sources.
- Business Unit Leads (e.g., Marketing, Finance, Operations): Seeking to utilise AI for targeted applications, from content generation to data analysis, without requiring extensive technical expertise.
Next Steps
- Organisations should conduct a thorough assessment of their specific AI requirements, considering whether a dedicated platform for custom agent development (like Gemini Enterprise) or a more integrated suite extension (like Microsoft Copilot) better aligns with their strategic objectives. It should be noted that Gemini Enterprise is far more than an agent service and provides a broad range of AI services and capabilities beyond those offered in Google Workspaces.
- Investigate the total cost of ownership, including licensing, implementation, and ongoing maintenance, for both Gemini Enterprise and Microsoft Copilot for Microsoft 365.
- Engage with Google Cloud, AWS and Microsoft Azure representatives to understand the specific capabilities for data integration and governance relevant to their existing enterprise architecture for AI.
