Gemini for SMEs: 10 concrete use cases in Google Workspace
Gemini is most useful to SMEs that already run on Google Workspace: it works directly inside Gmail, Docs, Sheets, Meet and Drive, on files your team is already allowed to open. Picking the right subscription tier matters more than the model itself, because several use cases are not available on the entry-level edition.
This article walks through 10 use cases, the Google app involved, the minimum Workspace edition and how much review each one needs. For the method behind writing good instructions, which applies to Gemini as much as any other assistant, see our prompt engineering guide for SMEs.
Which SMEs should look at Gemini?
Gemini makes the most sense for a company whose email, documents and calendar already live in Google. Its edge is not being smarter than ChatGPT or Claude on a one-off question. It is being present where the work already happens: in the email thread, in the spreadsheet, in the Meet call.
If your company runs on Microsoft 365, the logic flips. Copilot is the assistant built into Outlook, Word and Teams, and Gemini would just be one more browser tab. In that case the real choice is between Copilot, ChatGPT and Claude.
The French context makes the question practical. According to the France Num 2026 barometer published by the French Directorate General for Enterprise on 28 September 2026 (more than 9,000 companies surveyed), 40% of French micro, small and medium businesses use AI, and 53% of SMEs. Only 19% pay for a solution. Yet 85% of paid-solution users say AI has a positive impact, against 68% across all users. For a company already paying for Workspace, part of that paid solution is often sitting in the licence, unused.
What your Workspace subscription actually includes
Since 2025, Google no longer sells a separate Gemini add-on for Business editions: AI is bundled into the licence, but at different levels depending on the edition. Most SMEs find this out late, when someone looks for the Gemini panel in Sheets and cannot find it.
Here is the breakdown given by Google Workspace's help page on AI features in Business editions, checked in November 2026:
| Edition | Annual price listed by Google (per user per month) | What Gemini does |
|---|---|---|
| Business Starter | €6.80 | Help me write and side panel in Gmail, Gemini app with base models, NotebookLM (now Gemini Notebook) with base limits, Vids |
| Business Standard | €13.60 | Everything in Starter, plus the Gemini panel in Docs, Sheets, Slides, Drive and Chat, note taking in Meet, PDF analysis in Drive, Pro models in the app and a one million token context window |
| Business Plus | €21.10 | Same AI features as Standard, with more storage and security |
In practice, Business Standard is the minimum for most of the use cases below. Starter gives you writing help in Gmail and the Gemini app, which is enough to get started but not enough to equip an admin team.
10 Gemini use cases that work in SMEs
The most profitable use cases combine a repetitive task, an output that is easy to check and data that already sits in Workspace. For each one we give an example instruction, or prompt. Build your own using the anatomy of a good prompt: role, context, task, expected format.
1. Catch up on a 30-message email thread in two minutes
You are back from holiday and a client has exchanged 30 emails with three colleagues. The Gemini panel in Gmail summarises the thread, lists the decisions made and what is still open.
Prompt: "Summarise this thread in 5 points. Say who is waiting for a reply from us, on what topic, and since when."
Edition: Starter. Review: light, but check dates and amounts in the original emails before replying.
2. Draft client replies that match the history
"Help me write" in Gmail produces a draft. It gets more useful when you specify tone and constraints: "Politely explain that delivery is pushed back to the 18th, offer a 5% goodwill discount, and do not apologise three times."
Edition: Starter. Review: always, for any commercial commitment (deadline, discount, warranty). Gemini can rephrase a promise you never made.
3. Build a sales proposal from your Drive files
In Google Docs, Gemini can pull in Drive files you reference with "@": the proposal sent to a similar client, your price list, the meeting notes. You get a structured first draft in your usual format.
Prompt: "Using @Proposal_Client_A and @Meeting_Notes_Client_B, draft a proposal for client B. Keep our usual structure, adapt the scope to the meeting notes, and leave prices as [TO BE COMPLETED]."
Edition: Standard. Review: full. Leaving prices blank stops another client's rate from ending up in the document.
4. Analyse a sales export in Sheets
Open the export from your invoicing tool in Sheets and ask Gemini: "What is revenue per month and per salesperson? Which clients dropped more than 20% compared with last year?" Gemini can also write formulas, build a pivot table or a chart.
Edition: Standard. Review: thorough. Simple formulas are reliable, but references across several tabs or complex array formulas deserve a manual check on a few rows. With numbers, the rule from our article on AI hallucinations applies: recalculate one total by hand before presenting it.
5. Process a whole column with the AI() function in Sheets
Less well known, the AI() function applies an instruction to every row of a table: sort 400 customer reviews into "price", "delay", "quality", summarise form comments, extract a city from a badly typed address. Manual sorting turns into a quality check.
Edition: Standard, with a monthly usage quota set by Google. Review: by sample. Check 20 random rows before using the output. If the classification drifts, add two or three examples to the prompt, the few-shot prompting technique.
6. Take meeting notes in Meet
Gemini writes up the Meet call, with decisions and action items, and stores the notes in Drive. For a team that runs weekly check-ins, the benefit shows up in the first week.
Edition: Standard. Review: by the organiser, before sharing. Let participants know the meeting is being transcribed, especially external guests.
7. Query a document library with Gemini Notebook
NotebookLM, which Google renamed Gemini Notebook in July 2026, answers only from the sources you upload: collective agreement, internal procedures, product sheets, regulations. Every answer links back to the original passage, which makes checking easy. It is a simple form of RAG with zero development, and our article RAG vs prompting explains when this kind of tool is enough.
Example: a "New starter onboarding" notebook with the welcome booklet, health insurance details and remote work policy. New hires ask the notebook instead of HR.
Edition: Starter with base limits, higher limits from Standard. Review: moderate, thanks to citations. The main risk is an outdated source left in the notebook.
8. Create Gems, specialised assistants for your team
A Gem is a Gemini assistant set up with standing instructions and, if you want, reference files. For example: "Job ad writer" with your style guide and three model ads, or "Quote checker" that verifies mandatory mentions are present. It is the user-side equivalent of a system prompt, which we explain in system prompt vs user prompt.
Edition: Gemini app; sharing between colleagues depends on your edition and admin settings. Review: test the Gem on 5 real cases before rolling it out.
9. Automate simple tasks with Workspace Studio
Workspace Studio lets you build automations by describing what you need in plain language: a daily summary of unread emails, an alert when a key client writes, saving incoming attachments to a Drive folder and logging a row in Sheets. Google also lists connectors to third-party tools such as Asana, Jira, Mailchimp and Salesforce.
Edition: Business and Enterprise, according to Google. Review: ongoing. An automation that runs on its own needs watching for the first few weeks. Google itself notes that data loss prevention does not yet extend to third-party services connected to Studio. For more complex flows across several pieces of software, a dedicated automation tool is still a better fit.
10. Run in-depth research before a decision
Deep Research in the Gemini app reads dozens of web pages and writes a report: overview of a local market, supplier comparison, summary of a regulatory change. Treat the report as a starting point, not a conclusion.
Edition: Starter with a small monthly quota, much higher on Standard. Review: thorough. Open the cited sources, especially for a figure or a legal rule. For an important decision, ask Gemini to spell out its reasoning step by step, the chain of thought approach.
Summary: which use case for which team
Start with two use cases, not ten. The table below helps you choose by department and risk level.
| Use case | Team | Minimum edition | Review level |
|---|---|---|---|
| Gmail thread summaries | Everyone | Starter | Light |
| Client replies | Sales, support | Starter | Always |
| Proposal from Drive | Sales | Standard | Full |
| Sales analysis in Sheets | Management, finance | Standard | Thorough |
| AI() function | Marketing, quality | Standard | By sample |
| Meet notes | Everyone | Standard | By organiser |
| Gemini Notebook | HR, support | Starter | Moderate |
| Gems | Everyone | Depends on admin settings | Test before rollout |
| Workspace Studio | Admin, sales ops | Business | Ongoing monitoring |
| Deep Research | Management | Starter | Thorough |
Limits to know before rolling out
Gemini makes mistakes like any assistant, and its access to your data creates specific risks. Three points deserve attention.
Drive permissions become an AI issue. According to Google, Gemini follows existing Workspace permissions: it only shows a user files they can already open. That is reassuring, but it also exposes overly broad sharing. If the salary file is shared with "everyone in the company", Gemini can discuss it with everyone in the company. A quick audit of Drive sharing is a good first step.
Personal accounts remain the real risk. Google's commitments (no model training on customer data without permission, no human review of Gemini app conversations) apply to eligible Workspace editions. An employee who pastes a contract into Gemini on their personal Gmail account steps outside that framework. That is exactly the pattern described in our article on shadow AI.
Some new features reach France later. On 20 October 2026 Google presented a Gemini space for small businesses, with connections to Google Business Profile, Google Ads and third-party tools. Google states that these connections were not available in France at the time of the announcement. Before building a process on a feature you read about in a US article, check that it is active in your admin console.
On the regulatory side, Article 4 of the EU AI Act (Regulation (EU) 2024/1689) requires companies using AI systems to ensure a sufficient level of AI literacy among their staff. Rolling out Gemini without training the team does not meet that expectation. Details are in our article on AI Act Article 4.
Gemini, ChatGPT or Claude: do you have to choose?
For an SME on Google Workspace, Gemini is the default choice for work inside Google apps, and nothing stops you from using a second assistant for other tasks. The real issue is setting rules, not finding the perfect tool.
A few pointers to decide:
- Gemini wins when the task happens in Gmail, Sheets or Meet, and when you want to avoid paying for another licence.
- ChatGPT is still the tool your staff know best, with a wide ecosystem. Our article on 10 ChatGPT use cases for SMEs follows the same structure as this one.
- Claude stands out on long document analysis and structured deliverables, as shown in 10 Claude use cases for SMEs.
Stacking tools has a cost: several subscriptions, several privacy rules to explain, several habits to train. A short company AI policy that says which tool to use for which data keeps things under control.
How to get started in four weeks
A realistic rollout takes one month, with a small pilot group and two measurable use cases.
- Week 1: check each user's Workspace edition, enable Gemini in the admin console, review overly open Drive sharing.
- Week 2: pick two use cases from the table above and measure the time currently spent on those tasks.
- Week 3: train the pilot group (3 to 6 people) on those two cases, using their own files, and write two or three reference prompts together.
- Week 4: compare time spent, list the errors caught during review, then decide whether to extend or adjust.
Frequently asked questions
Is Gemini included in Google Workspace?
Yes, in Business and Enterprise editions, with no paid add-on since 2025. The feature level depends on the edition: Business Starter mainly covers Gmail and the Gemini app, while Business Standard adds Gemini in Docs, Sheets, Slides, Drive and Meet.
How much does Gemini cost for an SME?
The cost is the Workspace licence. Google's help page lists, on an annual plan, €6.80 per user per month for Business Starter, €13.60 for Business Standard and €21.10 for Business Plus. Check the price that applies to your contract in your admin console, as it can vary by reseller and renewal date.
Is my data used to train Gemini?
Not without your permission on eligible Workspace editions, according to Google's published commitments. Google also states that Gemini app conversations are not reviewed by humans in that context. These guarantees do not cover a personal Gmail account.
Does Gemini work well in French?
Yes for writing, summarising and everyday analysis. Some recent features launch in English first or are not yet available in France. Test each feature in your own environment before turning it into a team procedure.
Can we use Gemini if the company runs on Microsoft 365?
Technically yes, through the Gemini app, but you lose most of its value. Without Gmail, Drive and Sheets, Gemini becomes one chat assistant among others. In a Microsoft environment, Copilot fits your tools better, as we explain in our article on Microsoft 365 Copilot training.
Should teams be trained before Gemini is switched on?
Yes, that is what turns an unused licence into a measurable gain. A short training session built around your files, your privacy rules and how to check outputs is enough to start. It also addresses the AI literacy obligation set by Article 4 of the AI Act.
How GrowthPerf helps
GrowthPerf is a Qualiopi-certified training provider specialising in AI, no-code and automation for SMEs and non-profits. Our Operational AI for SMEs course starts from the tools you actually use, Google Workspace included: choosing your first two use cases, reference prompts, data rules and a method for checking outputs.
To revisit the basics of writing prompts, whatever the assistant, go back to our complete prompt engineering guide for SMEs. Our courses can be funded depending on your situation, for example through your OPCO: details are in our article on funding AI training through your OPCO.
Using Google Workspace but Gemini sits unused? Get in touch to discuss your use cases and the training format that fits your team.