Consumer, Productivity, and Enterprise GenAI

What Are Google's Three GenAI Offering Tiers?

When a business asks "How do we start using Gemini?", the honest answer is a question back: "Who is going to use it, and what data will they touch?" Google does not sell Generative AI as a single product. Instead, it packages the same underlying Gemini model family into three distinct offering tiers, each aimed at a different audience, a different risk profile, and a different billing model. For the Generative AI Leader exam, knowing these three tiers apart is one of the most heavily tested foundational skills — many scenario questions are simply "pick the right offering" in disguise.

The three tiers are: the Consumer tier (the free Gemini app and Google AI Studio), the Productivity tier (Gemini for Google Workspace), and the Enterprise / Developer tier (Vertex AI). They share the same brand and often the same model names, which is exactly why candidates confuse them. The exam exploits that confusion deliberately. The tier is not chosen by which model is "best" — all three can reach the most capable Gemini models — it is chosen by who the user is, where the data lives, who governs it, and how it is billed.

The Generative AI Leader exam does not ask you to write prompts or call APIs. It asks business questions: "An employee wants AI help drafting emails inside Gmail — which offering?" or "A bank wants to build a customer-facing chatbot with data-residency guarantees — which offering?" or "A marketing intern wants to brainstorm slogans for free at home — which offering?" This study note builds that decision framework using concrete Google product names, the data-governance boundaries of each tier, and Taiwan-friendly analogies.

By the end of this chapter you will be able to map any business scenario to the correct GenAI tier, explain who controls the data at each tier, articulate the cost model differences, and avoid the single most dangerous mistake on the exam — assuming the free consumer app carries the same enterprise data protections as the paid offerings.

The Three-Tier Spectrum at a Glance

Google's GenAI offerings sit on a spectrum of governance and customization. As you move from the Consumer tier toward the Enterprise tier, the organization gains more control, more guarantees, and more responsibility — while the individual user gives up casual convenience in exchange for accountability.

The Consumer Tier — Individuals and Experimentation

The Consumer tier is for individual people, not organizations. It covers the Gemini app (available at gemini.google.com and as a mobile app) and Google AI Studio (a free, browser-based prototyping environment for developers). It is designed for personal use, learning, and experimentation. There is a free version, and there is a paid consumer subscription (the Google AI Premium / Google One AI tier) — but even the paid consumer plan is a personal product, not an enterprise contract. There is no organizational admin, no IT-managed data boundary, and no enterprise data-protection guarantee unless the user is operating under a separate enterprise agreement.

The Productivity Tier — Employees Inside Workspace

The Productivity tier is Gemini for Google Workspace. This is AI embedded directly inside the apps employees already use: Gmail, Docs, Sheets, Slides, Meet, and Drive. It is sold per seat (per user, per month) as part of a Workspace subscription. Crucially, it is admin-governed: a Workspace administrator turns it on, controls who gets it, and the data stays inside the Workspace data boundary under the organization's existing Workspace contract and compliance terms.

The Enterprise / Developer Tier — Building Custom Solutions

The Enterprise / Developer tier is Vertex AI, Google Cloud's managed AI platform. This is for engineering teams building custom applications and AI agents — a customer-facing chatbot, a document-summarization pipeline, a retrieval-augmented search tool. Vertex AI gives full enterprise controls: IAM for access, VPC Service Controls (VPC-SC) for network isolation, data-residency options, CMEK encryption, and audit logging. It is billed pay-as-you-go based on consumption (tokens, requests, compute), like any other Google Cloud service.

白話文解釋（Plain English Explanation）

The three-tier model can feel abstract when you are staring at marketing pages that all say "Gemini." The best way to internalize it is to map each tier to a real-world equivalent that everyone in Taiwan already knows. The three analogies below all show the same idea — moving from "casual personal use" to "managed organizational use" to "fully custom enterprise build" — but each highlights a different facet of the decision.

Analogy 1 — Personal Bike vs Company Car vs Building Your Own Fleet

Imagine your organization needs to handle transportation.

Riding your own personal YouBike or bicycle (the Consumer tier) is what an individual does on their own. You grab a shared bike near the MRT station, ride it, return it, and pay a small personal fee — or none at all during the free window. It is fast, frictionless, and great for getting around quickly. But it is your personal account. The company has no record of where you rode, no policy over it, and no liability for it. If you carry a confidential company document in the bike's basket, that is entirely on you — there is no corporate insurance, no fleet manager, and no audit trail. This is the Gemini app and Google AI Studio: brilliant for an individual experimenting or learning, but it is a personal product with no enterprise data guarantee.

Driving a company car (the Productivity tier) is different. The company assigns each employee a vehicle from a managed pool. The fleet manager (the Workspace administrator) decides who gets a car, sets the rules, and the car stays within company policy and insurance. You still drive it yourself for everyday tasks — commuting, client visits — but it is governed. Every car is the same standard model, integrated into how the company already operates. This is Gemini for Google Workspace: AI handed to employees per seat, embedded in the tools they already use, with the data staying inside the company's Workspace boundary.

Building your own logistics fleet (the Enterprise / Developer tier) is what you do when off-the-shelf transportation is not enough. You buy the vehicles, hire drivers, design custom delivery routes, install GPS tracking, and run the whole operation to your exact specification. It is the most work, but you control everything — security, routing, branding, residency. This is Vertex AI: you build custom AI applications and agents, with full IAM, VPC-SC, and data-residency control, paying only for what you consume.

Analogy 2 — Street-Food Customer vs Staff Canteen vs Central Kitchen

Think about how a person gets fed versus how a company feeds people.

Buying from a night-market stall as a walk-in customer (the Consumer tier) is the individual experience. Anyone can walk up to the 夜市 stall, order a bowl of noodles, pay cash, and eat. It is open to the public, cheap or free to sample, and requires no membership. But the stall keeps no account in your name, offers no corporate billing, and signs no service agreement with your employer. If you mention your company's secret recipe while ordering, the vendor is under no enterprise confidentiality contract with your firm. This is the Gemini app: a public, individual-facing service — superb for a quick personal task, but not bound by your organization's data agreements.

Eating at the company staff canteen (the Productivity tier) is the managed-employee experience. The company runs a canteen, every employee with a badge can eat there, the menu is standardized, and the food-safety rules are set by the company's facilities team. You still serve yourself and eat what you like, but it all happens inside company premises under company policy. This is Gemini for Google Workspace: AI served to badge-holding employees, inside the apps the company already runs, governed by the Workspace administrator.

Operating a central kitchen that supplies custom meals (the Enterprise / Developer tier) is what a catering business does. You design recipes from scratch, control every ingredient's sourcing, set the kitchen's hygiene certification, decide which region each kitchen operates in, and scale production up or down based on orders. This is Vertex AI: you build bespoke GenAI products, control data residency and access, and pay for exactly the throughput you use.

Analogy 3 — Household Tool vs Company-Issued Equipment vs Industrial Production Line

Consider how tools are used at three different scales.

A household power drill bought for personal use (the Consumer tier) is something an individual picks up at the hardware store for weekend projects. It is inexpensive, immediately useful, and needs no approval from anyone. But it is registered to you, maintained by you, and carries no workplace-safety certification for professional use. If you bring it onto a job site, the company's safety officer has no record of it. This is Google AI Studio and the Gemini app: an individual's tool for tinkering and learning, outside any organizational governance.

Company-issued equipment from the equipment room (the Productivity tier) is standardized gear the firm hands to every qualified employee — same model, logged in the asset register, maintained by the company, used within company safety rules. Employees use it daily for their normal work without thinking about procurement. This is Gemini for Google Workspace: a uniform AI capability issued per seat to employees, managed centrally by an administrator, used inside everyday work.

An industrial production line built to specification (the Enterprise / Developer tier) is engineered for a specific manufacturing goal. Engineers design the line, choose every component, install access controls and quality sensors, decide which factory it sits in, and pay for the materials and power it consumes. This is Vertex AI: a platform on which engineering teams design custom AI solutions, with full control over security, location, and cost, billed by consumption.

The Consumer Tier — Gemini App and Google AI Studio

The Consumer tier is the entry point most people meet first. It has two main products.

The Gemini App

The Gemini app is the chat-style assistant at gemini.google.com and on mobile. It is aimed at individuals for everyday tasks: answering questions, drafting personal text, brainstorming, summarizing, and creative work. There is a free version and a paid consumer subscription (Google AI Premium / Google One AI Premium) that unlocks more capable models and higher limits. Either way, it is a personal account product — the user signs in with a personal Google account (or, in some cases, a Workspace account where the admin has enabled it).

Google AI Studio

Google AI Studio (at aistudio.google.com) is a free, browser-based environment for developers to prototype with the Gemini API. A developer can write a prompt, test it, tune parameters, and grab an API key to start experimenting. It is the fastest way to get hands-on with Gemini models for a proof of concept. But Google AI Studio is explicitly a prototyping tool, not a production enterprise platform — when a prototype needs enterprise controls, the path forward is to graduate to Vertex AI.

When to Pick the Consumer Tier

• Individual learning and experimentation: A person exploring what GenAI can do.
• Personal productivity outside a managed work context: Drafting a personal email, planning a trip, brainstorming ideas at home.
• Quick developer prototyping: Testing a prompt idea in Google AI Studio before deciding whether to build it for real.
• No organizational data involved: The user is not handling regulated, confidential, or customer data on the company's behalf.

The Productivity Tier — Gemini for Google Workspace

Gemini for Google Workspace is the Productivity tier: AI woven directly into the apps employees already live in.

Embedded in Everyday Apps

Instead of asking employees to switch to a separate chat window, Gemini for Workspace puts AI inside the workflow:

• Gmail: "Help me write" drafts and refines email replies; summarizes long threads.
• Docs: Generates drafts, rewrites paragraphs, and summarizes documents.
• Sheets: Helps create tables, organize data, and generate formulas.
• Slides: Generates images and helps build presentation content.
• Meet: Provides "take notes for me," translated captions, and meeting summaries.
• Gemini in the side panel: A chat assistant that can reason over the user's own Workspace content (their emails, files) with the user's existing permissions.

Per-Seat Licensing and Admin Governance

Gemini for Workspace is licensed per user, per month, added to a Workspace subscription. A Workspace administrator controls the rollout from the Admin console — deciding which organizational units get it, turning features on or off, and applying the organization's existing Workspace security and compliance settings. The AI inherits the same data boundary as the rest of Workspace: enterprise data-protection commitments apply, and content is not used to train Google's foundation models.

When to Pick the Productivity Tier

• Employee productivity inside Google Workspace: Drafting, summarizing, and meeting support for the whole workforce.
• Fast rollout with no engineering: The administrator flips it on; there is nothing to build.
• Governed use of company data: Employees can safely use AI on company emails and documents because it stays in the Workspace boundary.
• Predictable budgeting: A flat per-seat cost is easy for finance to forecast.

For a deeper look at the features, licensing tiers, and admin controls, see /en/certs/gcp/genai-leader/topics/gemini-for-google-workspace.

The Enterprise / Developer Tier — Vertex AI

Vertex AI is the Enterprise / Developer tier — Google Cloud's managed platform for building custom GenAI applications and agents.

A Platform for Builders

Where the Consumer and Productivity tiers deliver a finished experience, Vertex AI delivers building blocks. Engineering teams use it to:

• Call Gemini and other foundation models through an enterprise-grade API.
• Ground responses in the organization's own data with retrieval-augmented generation (RAG), using Vertex AI Search and grounding tools.
• Build and orchestrate AI agents with Vertex AI Agent Builder.
• Tune and evaluate models, manage prompts, and monitor production usage.
• Deploy customer-facing chatbots, internal copilots, and document-processing pipelines.

Full Enterprise Controls

Because Vertex AI is a Google Cloud service, it inherits the entire Google Cloud governance stack:

• IAM: Fine-grained roles control who can call models, deploy, and administer.
• VPC Service Controls (VPC-SC): Creates a network perimeter so AI requests and data cannot leave the trusted boundary.
• Data residency: Customers can choose the region where data is processed and stored, supporting regulatory requirements.
• CMEK (customer-managed encryption keys) and audit logging through Cloud Audit Logs.
• Enterprise data commitment: Prompts and outputs are not used to train Google's foundation models.

Pay-As-You-Go Billing

Vertex AI is billed by consumption — typically by tokens processed (input and output), requests, and any associated compute or storage. There is no per-seat fee; you pay for exactly the usage your application generates. This scales naturally from a small pilot to a high-traffic production service.

When to Pick the Enterprise / Developer Tier

• Building a custom application or agent: A customer-facing chatbot, an internal search assistant, an automated document pipeline.
• Strict governance requirements: VPC-SC isolation, data residency, CMEK, and audit logging are mandatory.
• Integration with Google Cloud data: Grounding AI in BigQuery data, Cloud Storage documents, or enterprise databases.
• Consumption-scaled cost: Usage is variable and pay-as-you-go fits better than fixed seats.

For full coverage of Vertex AI's GenAI capabilities, model garden, grounding, and agents, see /en/certs/gcp/genai-leader/topics/vertex-ai-for-generative-ai.

Decision Framework: Choosing the Right GenAI Tier

The Generative AI Leader exam expects you to reason through "pick the right offering" scenarios quickly. Here is the simplified framework.

Step 1 — Who Is the User?

• An individual, for personal or learning use: Consumer tier (Gemini app / Google AI Studio).
• The general workforce, doing everyday office tasks: Productivity tier (Gemini for Google Workspace).
• An engineering team building something: Enterprise / Developer tier (Vertex AI).

Step 2 — Does the Scenario Require Building a Custom Application?

• Yes — a chatbot, agent, or pipeline must be built: Vertex AI.
• No — people just need AI help in their normal tools: Gemini for Google Workspace.
• No — a person just wants to chat or prototype: Consumer tier.

Step 3 — What Data-Governance Level Is Required?

• Regulated data, residency, VPC-SC, CMEK, audit logging: Vertex AI.
• Company data inside Workspace, admin-governed: Gemini for Google Workspace.
• No organizational data, personal use only: Consumer tier.

Step 4 — What Cost Model Fits?

• Predictable per-seat budgeting for a workforce: Gemini for Google Workspace.
• Consumption-based pay-as-you-go scaling with usage: Vertex AI.
• Free or low-cost personal subscription: Consumer tier.

Data Governance Differences Across the Three Tiers

Data governance is where the three tiers diverge most sharply, and it is the highest-value exam concept in this topic.

Consumer Tier Governance

The Consumer tier operates under consumer privacy terms. There is no organizational admin, no enterprise contract, and no enterprise data-protection guarantee. Consumer-tier conversations may be reviewed and used to improve Google's services unless the user changes their personal settings. There is no audit trail an organization can inspect. This is appropriate for personal, non-sensitive use only.

Productivity Tier Governance

Gemini for Google Workspace inherits the Workspace enterprise data boundary. The organization already has a Workspace contract with data-protection commitments, and Gemini operates inside it. Content is not used to train Google's foundation models, the Workspace administrator governs access, and existing Workspace compliance certifications and controls extend to the AI features.

Enterprise Tier Governance

Vertex AI offers the deepest controls: IAM, VPC Service Controls, data residency by region, CMEK, and Cloud Audit Logs. Prompts and outputs are not used to train foundation models. This is the tier for regulated industries — finance, healthcare, government — that must prove where data is processed and who accessed it.

The governance gap between the Consumer tier and the two enterprise-grade tiers is the single most exam-relevant distinction in this topic. For a full treatment of data residency, training-data policy, and isolation controls, see /en/certs/gcp/genai-leader/topics/data-governance-for-genai.

Cost Models Compared

Each tier is priced for its audience, and the exam may ask which billing model fits a stated business need.

• Consumer tier: Free for the base Gemini app and Google AI Studio prototyping; a flat personal subscription (Google AI Premium) for the upgraded experience. Billed to an individual, not an organization.
• Productivity tier (Gemini for Workspace): Per seat, per month, added to the Workspace subscription. Predictable and easy for finance to forecast across a workforce. You pay whether or not every employee uses it heavily.
• Enterprise / Developer tier (Vertex AI): Pay-as-you-go by consumption — input and output tokens, requests, and associated compute. Costs scale up and down with actual usage, which suits variable workloads and lets a pilot start cheap.

A useful rule for exam scenarios: a fixed headcount of employees points to per-seat Workspace; variable, usage-driven demand from an application points to consumption-based Vertex AI; an individual's personal spend points to the Consumer tier.

How the Three Tiers Relate to Each Other

The tiers are not rivals — they form a continuum that an organization moves along as its needs mature.

• Discovery starts in the Consumer tier. Employees and developers first meet Gemini through the free app or Google AI Studio. This builds intuition and surfaces ideas.
• Workforce enablement happens in the Productivity tier. Once an organization wants its whole staff to benefit safely, it adopts Gemini for Google Workspace — governed, per seat, embedded.
• Differentiated products are built in the Enterprise tier. When a use case becomes a custom application — a branded customer chatbot, an internal knowledge agent — the team builds it on Vertex AI.

A prototype that begins in Google AI Studio has a natural graduation path to Vertex AI: the same Gemini API concepts carry over, but Vertex AI adds the IAM, VPC-SC, residency, and audit controls that production demands. Recognizing this prototype-to-production journey is a common exam theme.

Common Business Scenarios Mapped to Tiers

To cement the framework, here are scenarios in the exact style the exam uses.

Scenario — A Marketing Intern Brainstorming Slogans at Home

An intern, on a personal laptop, wants to brainstorm slogan ideas for a non-confidential public campaign. No company data, individual use. Correct tier: Consumer (the Gemini app). It is free and frictionless, and nothing sensitive is involved.

Scenario — Rolling AI Out to 5,000 Office Staff

A company wants all 5,000 employees to get AI help drafting emails and summarizing meetings inside the tools they already use, governed centrally, with predictable cost. Correct tier: Productivity (Gemini for Google Workspace). Per-seat licensing, admin-controlled, embedded — no engineering needed.

Scenario — A Bank Building a Customer-Facing Chatbot

A bank needs a customer-facing support chatbot, grounded in its own policy documents, with data residency in a specific region, VPC-SC isolation, and full audit logging. Correct tier: Enterprise / Developer (Vertex AI). Only Vertex AI provides the governance and the building blocks to construct and deploy the application.

Scenario — A Developer Testing a Prompt Idea

A developer wants to quickly test whether a summarization prompt works before proposing a project. No production traffic, no enterprise data. Correct tier: Consumer (Google AI Studio) for the prototype, with a planned graduation to Vertex AI if the idea is approved.

Frequently Asked Questions

Q: Is the Gemini in my Gmail the same product as the Gemini app I use at home?

A: No — they are different tiers. The Gemini features inside Gmail and Docs are Gemini for Google Workspace (the Productivity tier): admin-governed, per-seat, and operating inside your organization's Workspace data boundary. The standalone Gemini app at gemini.google.com that you use with a personal account is the Consumer tier. They may use the same underlying model, but the governance, billing, and data terms are completely different.

Q: Can I just use the free Gemini app for work to avoid paying for a license?

A: You should not put confidential or regulated company data into the personal Gemini app. The free Consumer tier operates under consumer privacy terms and carries no enterprise data-protection guarantee — interactions may be used to improve Google's products, and there is no admin oversight or audit trail. For work involving company data, the organization should use Gemini for Google Workspace or Vertex AI, both of which carry the enterprise commitment that data is not used to train foundation models.

Q: When should a business choose Vertex AI instead of Gemini for Google Workspace?

A: Choose Vertex AI when the goal is to build something custom — a chatbot, an AI agent, a document-processing pipeline — or when the scenario demands deep controls like VPC Service Controls, data residency, CMEK, and audit logging. Choose Gemini for Google Workspace when employees simply need AI assistance inside Gmail, Docs, Sheets, Slides, and Meet with no development effort. "Build / develop / integrate" points to Vertex AI; "use / draft / summarize" points to Workspace.

Q: What is Google AI Studio, and how is it different from Vertex AI?

A: Google AI Studio is a free, browser-based environment for developers to prototype quickly with the Gemini API — part of the Consumer tier. Vertex AI is the enterprise platform for production GenAI applications, with IAM, VPC-SC, data residency, and pay-as-you-go billing. The typical journey is to prototype an idea in Google AI Studio and then graduate to Vertex AI when it needs enterprise governance and is ready for real users.

Q: Do all three tiers use the same Gemini models?

A: They can reach the same Gemini model family — that is exactly why the tiers are easy to confuse. The model is not what distinguishes them. What differs is the packaging, audience, governance, and billing: who can use it, where the data lives, who administers it, and how it is paid for. On the exam, never choose a tier based on "which has the best model" — choose based on the user, the data, and the governance requirement.

Q: How is each tier billed?

A: The Consumer tier is free for the base experience, with an optional flat personal subscription for upgrades. The Productivity tier (Gemini for Google Workspace) is billed per seat, per month as part of a Workspace subscription — predictable for a fixed workforce. The Enterprise / Developer tier (Vertex AI) is pay-as-you-go by consumption (tokens, requests, compute) — it scales with actual usage and suits variable demand.

Summary

Google packages its Gemini models into three GenAI offering tiers, and telling them apart is a core Generative AI Leader exam skill:

• Consumer tier — Gemini app and Google AI Studio: For individuals, learning, and prototyping. Free or personal subscription. No enterprise data guarantee — never for confidential company data.
• Productivity tier — Gemini for Google Workspace: For employees, embedded in Gmail / Docs / Sheets / Slides / Meet. Per-seat licensing, admin-governed, inside the Workspace data boundary.
• Enterprise / Developer tier — Vertex AI: For engineering teams building custom applications and agents. Pay-as-you-go, with full IAM, VPC Service Controls, data residency, and audit logging.

The tier is chosen by who the user is, what they need to build, what data governance is required, and which cost model fits — not by which model is "best," because all three reach the same Gemini family. Master this decision framework, anchor on the verb in each scenario, and remember the governance trap — the free consumer app is never the answer for sensitive enterprise data — and you will confidently handle every "pick the right offering" question on the exam.