# How our product team used the Hello Retail MCP to turn the monthly report into one prompt

> One prompt builds a performance report split by Search, Recommendations, Pages and Product Agents. One more line puts it on a monthly schedule.

- **Who**: the Hello Retail product team
- **Used**: the Hello Retail MCP analytics tools
- **To**: turn the monthly performance report into one prompt that runs itself
- **Result**: A report split by solution, ending in a to-do list, on the first of every month
- **Source**: Hello Retail team
- **Products**: Search, Product Recommendations, Pages, Product Agents
- **Canonical page**: https://helloretail.com/en/agentic-playbooks/monthly-performance-report/
- **Evidence**: [Fall Release 2026 webinar, from 4:26](https://helloretail.com/en/blog/mcp-ecommerce-automation/#t=266)

## Prompts

### Build the report (as typed)

```text
Please create a performance report for the months August and September, separating out numbers for Search, Recommendations, Pages, and Product Agents.
```

### Put it on a schedule (as typed)

```text
Please run this report every first of the month for the two previous months.
```

## Hello Retail MCP tools for this play

- `website_listForCompany`
- `search_getAnalyticsOverview`
- `search_getTopSearches`
- `recoms_getAnalyticsTotals`
- `pages_getAnalyticsOverview`
- `productAgents_getAnalytics`

Works with: Claude Cowork scheduled tasks

## The job

A monthly performance report usually means visiting several places, collecting data points and stitching them into one document. Someone has to remember to do it, every month.

## The play

1. Connect the Hello Retail MCP server to your AI assistant.
2. Ask for the report and name each solution you want counted on its own. That last part matters: It makes the report split revenue and traffic clearly, even where several solutions were involved in the same sale.
3. Turn the same prompt into a scheduled task and add one line, so it runs on the first of every month.

## What came back

A few minutes later, the report was ready. It showed September against August for every solution, broke search down into the top searches and how they developed, and showed where shoppers went after each solution. It ended with a list of what to do next. The first item: Fix the 10 dead-end searches it had found.

After a one-time approval of the permissions it needs, the scheduled task fills in the right months each time and the report arrives without anyone asking.

## Make it yours

- A report like this takes a little while to build. Start the next question in another chat while you wait.
- Ask for the numbers your management meeting actually uses, and the report follows that shape every month.

## Try it

Connect the Hello Retail MCP server to an AI assistant that supports MCP, then use the prompt above. Setup guide: https://helloretail.com/en/mcp/#connect

Part of the Agentic Playbook: https://helloretail.com/en/agentic-playbooks/
