> For the complete documentation index, see [llms.txt](https://help.bizmachine.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.bizmachine.com/en/mcp-connector/mcp-connector-prompts.md).

# Examples of using the BizMachine MCP connector

Specific prompts for the BizMachine MCP connector: from preparing for a meeting to prioritizing a whole segment by your use case. Ready to try.

*Updated: 7 October 2026*

The BizMachine MCP connector can handle many different tasks. That is why we have put together a set of examples of how to use it in practice and where to start. For each prompt you fill in your own values, and the assistant reads the data directly from your BizMachine workspace.

The prompts are grouped by level: start with the basic ones and add more complex ones as you go. At the end you will find prompts for people who lead a team.

***

### Before you start

* **The connector must be connected.** We have guides for [Claude](/en/mcp-connector/connecting-claude.md), [ChatGPT](/en/mcp-connector/connecting-chatgpt.md) and [Microsoft Copilot Studio](/en/mcp-connector/connecting-copilot-studio.md).
* **Replace the text in square brackets with your own values.** `[industry]`, `[region]`, `[company]`, `[amount]`, `[label name]`.
* **Be specific.** The more precise the request, the better the answer. Add the industry, region, size or revenue.
* **You can chain prompts.** Use the result of one prompt as the input for the next.
* The prompts are templates, not exact wording. Feel free to rephrase them in your own words.

{% hint style="warning" %}
**Don't connect the connector to free accounts or weak models.** It will technically connect, but a weak model can't work with an MCP server properly and will likely make up data. BizMachine doesn't support that kind of setup.
{% endhint %}

### Basic prompts: start here

**1. Preparing for a meeting.**

> "Prepare me for a meeting with \[company]. Give me a short briefing: what they do, how big they are, how they are doing financially and whether anything has been happening at the company lately."

You get a briefing from verified data in a few minutes instead of searching manually.

**2. Finding companies directly in the conversation.**

> "Find \[industry] companies in \[region] with revenue above \[amount]. How many are there?"

You narrow a segment by describing it, not by clicking through filters. The assistant returns the count, so you see the size at once.

### Prompts for everyday work

**3. Filter by growth and activity.**

> "From the companies in the \[industry] industry in \[region], pick the ones that are growing and actively hiring. Sort them by growth."

**4. Companies by technology.**

> "From the companies in the \[industry] industry, show which ones BizMachine lists as using the cloud (AWS, Azure, Google Cloud) and how often each technology appears."

**5. Targeting by e-shops.**

> "Show e-shops on \[platform] with traffic above \[number] per month that offer \[payment method]."

**6. Signals and changes at specific companies.**

> "What has changed at these companies in the last three months: \[company 1, company 2, company 3]? I am interested in hiring, management changes and similar events."

**7. Working with your labels.**

> "Take the companies from my label \[label name] and find out how many of them are growing and how many currently have a signal."

**8. Saving a filter result to a label.**

> "Find \[industry] companies in \[region] (\[size] employees) with the highest growth potential and save them to the label \[label name]."

The assistant creates the label in Prospector right away. You don't retype or export anything.

### Advanced prompts

**9. Comparing companies side by side.**

> "Compare the companies \[company A], \[company B] and \[company C] side by side. Which is the largest, which is growing fastest, and which fits our use case best?"

**10. Prioritizing a whole segment by your use case (ICP).**

> "From my segment \[name or label], pick the 20 companies that fit our use case best according to our AI Prioritization. For each one, give its value and the main reason for that value."

**11. Pulling companies from a website into a label.**

> "Go through this page \[URL], find the first \[number] companies in BizMachine and save them to a new label \[label name]."

This works in assistants that can read web pages (claude.ai, ChatGPT). The connector finds the companies and creates the label.

**12. A large cross-country analysis in one question.**

> "Compare the number of companies in the \[industry] industry with revenue above \[amount] in the Czech Republic, Poland and Germany, and put the results in a table. For the Czech companies, check how many use the cloud (AWS, Azure, Google Cloud). Then, from my label \[label name], pick the ones that fit this best."

### Connected companies and roles

Since September 2026, the connector has also shown connected companies and the roles we can find for a company in public sources. A detailed description, sample answers and limitations are in the articles [Connected companies](/en/mcp-connector/connected-companies.md) and [Who works at the company](/en/mcp-connector/who-works-at-the-company.md). Here are the prompts to copy.

**13. The customer's whole group in one answer.**

> "For the company \[Company ID], find the connected companies. Who connects them, and with what share?"

You get the parent company, subsidiaries and companies that share a person in management all at once. In detail: [Connected companies](/en/mcp-connector/connected-companies.md).

**14. Numbers for the group, not for one company.**

> "For the company \[Company ID], find the connected companies and add up how many vehicles they operate."

You can swap the end of the question: how many premises they have, how many e-shops they run. The connector goes through the profiles of the companies in the group and adds up the figures. For a large group, this takes a moment. We know the figure only for some of the companies, so the total is an estimate.

**15. What is happening at companies with a strong tie.**

> "For the company \[Company ID], find the connected companies and, for those with a strong tie, tell me what has changed at those companies in the last 12 months."

Ties and signals in one answer: hiring, management changes, a new connected company.

**16. Companies where we already know someone in the right role.**

> "Find companies in the \[industry] industry with \[number of employees] employees in \[region] where you know at least one person in IT leadership, and save them to the label \[label name]."

The result is companies, not people. Contact details for specific people are in Prospector. In detail: [Who works at the company](/en/mcp-connector/who-works-at-the-company.md).

**17. Which of your customers are hiring for finance positions right now.**

> "Go through the companies in the label \[label name] and tell me which of them have posted job ads for finance positions in the last 30 days. For each one, list the specific positions."

You can ask about IT, sales or purchasing positions in the same way.

**18. Companies that are building a sales team.**

> "Find companies that have posted job ads for sales positions in the last 30 days and have at least \[number] open positions in total. Keep only the ones where you know someone in sales leadership, and save them to the label \[label name]."

Hiring and the roles found in one filter. These companies are strengthening their sales team, and we already know someone who leads it.

### Prompts for team leads

If you set the direction for sales, the connector answers strategic questions in a few seconds.

**Market size and ICP validation.**

> "How many companies in the Czech Republic operate in the \[industry] industry, have 50 to 250 employees and revenue above \[amount]? Show me a sample of ten companies so I can see who you mean."

**Comparing markets across countries.**

> "Compare how many companies in the \[industry] industry with revenue above \[amount] there are in the Czech Republic, Slovakia, Poland and Germany. Put the results in a table."

**What is happening at key accounts.**

> "What has changed at our most important accounts in the last three months: \[list of companies]? I am interested in hiring, management changes and financial health."

You get the most out of the connector through AI Prioritization, that is, comparing companies against your specific use case (prompt 10 above). If you don't have a use case set up in Prospector yet, setting one up is the best first step.

### Frequently asked questions

**Do I have to connect the connector first?** Yes. The prompts work only after you connect it. We have guides for [Claude](/en/mcp-connector/connecting-claude.md), [ChatGPT](/en/mcp-connector/connecting-chatgpt.md) and [Microsoft Copilot Studio](/en/mcp-connector/connecting-copilot-studio.md).

**Do the prompts work in both ChatGPT and Claude?** Yes. We have tested claude.ai, ChatGPT and Claude Code. The connector also works with other clients that support MCP over OAuth.

**Do I have to write the prompt exactly like this?** No. The prompts are templates. Rephrase them in your own words. What matters is that you add the industry, region, size or revenue.

**Why does the assistant sometimes answer without data or make data up?** The most common cause is a weak model or a free account. Don't connect the connector to free accounts or weak models. It will technically connect, but a weak model can't work with an MCP server properly and will likely make up data. BizMachine doesn't support that kind of setup.

**Can I chain prompts?** Yes. You use the result of one prompt as the input for the next. For example, you first find a segment and then have it sorted by AI Prioritization.

***

### What next

* [BizMachine MCP connector: what it does and what it is for](/en/mcp-connector/mcp-connector-overview.md)
* [Connected companies: ownership structure and personal ties](/en/mcp-connector/connected-companies.md)
* [Who works at the company: roles, hiring and search by role](/en/mcp-connector/who-works-at-the-company.md)
* [How to connect BizMachine in Claude](/en/mcp-connector/connecting-claude.md)
* [How to connect BizMachine to ChatGPT](/en/mcp-connector/connecting-chatgpt.md)
* [How to connect BizMachine in Microsoft Copilot Studio](/en/mcp-connector/connecting-copilot-studio.md)