How to operate DataMetrixx

1. Getting started

After creating your account at datametrixx.com, you will receive a confirmation email. Once your account is verified, you can access the admin panel.

The first time you sign in, you land on the Dashboard, which summarizes your account status: active clients, loaded datasets, active publications and storage usage.

2. Roles and permissions

DataMetrixx has 5 roles with hierarchical permissions:

3. How data flows

This is the mental model you will use all the time. Read it once and the rest of the platform will make sense.

• Your account can have one or more clients, depending on your plan.
• Each client can have many independent datasets (for example: “Sales”, “Inventory”, “Production”).
• Each dataset accumulates versions. Every time you upload a CSV to the same dataset, a new version is created and the previous one is kept.
• Components (charts, tables, KPIs) are built on top of a dataset — not a specific version.
• The dataset’s active version is the one components display. When a new import finishes successfully, it automatically becomes the active one.

4. Managing clients

Each client represents a company, brand or project you will deliver data and visualizations to.

Create a client

1. Go to Clients from the sidebar.
2. Click New client.
3. Fill in the name, slug (friendly URL) and contact email.
4. Optional: create a client administrator with an email and password.
5. Choose the domain type: none, subdomain or custom domain.
6. Save. The client is ready to load data and publish.

5. Managing users

Users are created in DataMetrixx and AWS Cognito at the same time. Only Platform Admin and Tenant Owner can create and edit users or reset passwords.

Create a user

1. Go to Users → New user.
2. Fill in the full name, email and initial password.
3. Assign a role (Tenant Admin, Client Admin or Viewer).
4. For a client role, select the assigned client.
5. Save. The user receives their credentials.

Reset a password

Use the Reset password button. The new password must have:

• At least 8 characters
• At least one uppercase letter
• At least one number
• At least one special character

The form validates these requirements in real time with a visual checklist.

6. Catalogs

Catalogs define the allowed values for your data dimensions: continents, regions, countries, classifications, elements and sub-elements. They work as master lists that keep naming consistent when importing.

Go to Catalogs to view, edit or import new values. Catalogs are shared across all of the client’s datasets.

7. Datasets

A dataset is the container that groups every version of the same set of data. Think of it as “the topic”: specific rows come and go with each import, but the dataset is the stable anchor your components connect to.

Create a dataset

1. Go to Datasets and click New dataset.
2. Give it a clear name and description (e.g. “Sales 2026”).
3. Save. The dataset starts empty, waiting for its first import.

Row structure

Every row you load into the dataset follows a common dimensional schema. You don’t have to use every field — fill in the ones that apply to your business:

The combination continent + region + country + classification + element + sub_element + year acts as a unique key within each import. It is what the system uses to detect duplicates.

8. Imports and versioning

The main way to load data is by uploading CSV files. Every import you run on the same dataset becomes a new version — it does not overwrite the previous one.

Step-by-step process

1. Go to Imports → New import.
2. Select the target dataset (or create a new one).
3. Upload the CSV file (comma-delimited, UTF-8 encoding).
4. Map each column in your CSV to the matching system field: continent, region, country, classification, element, sub_element, year, value. If your CSV uses other names (e.g. “País”, “Year”), the mapping translates them.
5. Choose a duplicate policy:
   • Reject: if a row duplicates an existing key, it is skipped.
   • Upsert: updates the duplicated row’s value with the new one.
6. Start the import. The system processes it in the background and shows the progress.

Import statuses

Versioning and the active version

Each dataset has an active version: the import whose rows components display today. Previous versions remain stored — you can go back to any of them at any time.

• When an import finishes with completed status, it becomes the active version automatically.
• You can activate a previous version from the dataset’s import list.
• You can roll back a problematic import: its rows are deleted and, if it was the active one, the dataset returns to the previous version.

9. Files

The Files section manages binary assets (images, documents, logos) stored in S3 per client. When you upload a file you get a public URL you can use in image components, site builder pages or branding.

User avatars are stored under the public avatars/ prefix and can be accessed without a signature.

10. Components

A component is a reusable visualization (table, chart, KPI, map…) connected to a dataset. The key thing to understand is this:

Available types

How a component connects to the data

When you create a component, you need to answer these questions:

1. Which dataset does the information come from? You select one of the client’s datasets. The component is tied to it by id — not to a specific version, but to the dataset as a whole.
2. Which filters do I apply? For example: only the country “Mexico”, only years between 2020 and 2026, only the element “Sales”. Filters discard rows before grouping.
3. How do I group? Dimensions are the columns you sum or average by. For a bar chart of sales by country, the dimension is country. For a timeline, it is year.
4. Which aggregation do I use? sum, avg, count, max or min over the value column.
5. How do I sort and how much do I show? Sort by dimension or by aggregated value, ascending or descending, and limit to the top N results.

Example 1 — Bar chart: sales by country

I want to see how much each country sold in 2026, sorted from highest to lowest:

{
  "dataset_id": "sales-dataset-uuid",
  "dimensions": ["country"],
  "aggregation": "sum",
  "filters": {
    "element": "Sales",
    "year": 2026
  },
  "sort_by": "value",
  "sort_order": "desc",
  "limit": 10
}

This translates to: “sum value grouped by country, only rows where element = "Sales" and year = 2026, sorted by total from highest to lowest, top 10.”

Example 2 — Timeline: monthly sales of a product

{
  "dataset_id": "sales-dataset-uuid",
  "dimensions": ["year"],
  "aggregation": "sum",
  "filters": {
    "element": "Sales",
    "sub_element": "Online"
  },
  "year_range": [2020, 2026],
  "sort_by": "year",
  "sort_order": "asc"
}

Example 3 — KPI: total sales for the month

{
  "dataset_id": "sales-dataset-uuid",
  "dimensions": [],
  "aggregation": "sum",
  "filters": {
    "element": "Sales",
    "year": 2026
  }
}

Without dimensions, the result is a single number — perfect for a KPI.

Creation flow in the panel

1. Go to Components → New component.
2. Give it a name and description.
3. Choose the type (table, bars, line, KPI, etc.).
4. Select the source dataset.
5. Set up filters, dimensions, aggregation and sorting.
6. Set up the style: colors, labels, legend, number format.
7. Use Preview to see the result with real data.
8. Save. The component is ready to be used in templates.

Components that don’t come from a dataset

The Text, Image, PDF and File types don’t need a dataset. They add static content to a dashboard: a title, a logo, a PDF manual, a video. They are uploaded directly to the component from the panel.

11. How do components update when I upload new data?

This is probably the most important question about the system. The short answer: they update on their own, without touching the components.

The step-by-step flow, under the hood

1. You have a dataset whose active version = Import #1.
2. Your bar chart reads rows only from Import #1.
3. You upload a new CSV to the same dataset.
4. The system processes it and creates Import #2.
5. When it finishes, Import #2 is marked as the active version — automatically.
6. The next time someone opens the dashboard, the chart reads from Import #2.

What you do need to check

• Column structure: if the component filters by element = "Sales" and in the new CSV you renamed the column to “Sale”, the filter will stop finding matches. Keep names consistent or re-map the columns when importing.
• Duplicate policy: if you choose reject, rows that already exist are not updated — your active version could end up incomplete. If you choose upsert, records are updated with the new values.
• Import errors: if an import ends in failed status, the previous version stays active. Review the error summary, fix the CSV and upload it again.
• Rollback: if the new CSV introduced a problem, go to the import and roll it back. The dataset returns to the previous version, and so do the components.

When it does NOT update on its own

If instead of uploading to the same dataset you create a new dataset, existing components keep pointing to the old one. In that case you need to edit each component and point it to the new dataset — or, better, upload the data to the original dataset.

12. Templates

A template is a composition of several components in a grid layout. It works as the dashboard you will later publish.

1. Go to Templates → New template.
2. Set the name, description and owning client.
3. Drag components onto the grid and adjust their size.
4. Save it with draft status.
5. When it is ready, change it to published.

13. Publications and embeds

Publishing a template generates a public token that lets you display the dashboard on any website without authentication.

1. Select a template with published status.
2. Create a publication. A public_token is generated automatically.
3. Set the visibility: public or controlled.
4. Define the list of allowed domains (CORS).
5. Copy the embed snippet and paste it into your website.

Embed snippet

<div id="datametrixx-container"></div>
<script src="https://api.datametrixx.com/datametrixx.js"
        data-token="YOUR_PUBLIC_TOKEN"
        data-container="datametrixx-container">
</script>

14. Site Builder

Each client can have a public website with its own branding, pages and embedded publications, without writing a single line of code.

1. Go to Site Builder.
2. Set up the header, footer and navigation menu.
3. Create pages with a slug, title and SEO description.
4. Drag publications (dashboards) into the content.
5. Publish the page. It becomes available on the client’s domain.

15. Branding

In Branding, each client sets up its visual identity:

• Logo (uploaded as a file).
• Favicon for the browser tab.
• Primary and secondary colors.
• Typography (when available).

Changes are applied instantly to the public site and the embeds.

16. Domains

Each client can publish under one of these schemes:

17. Billing

The Billing section shows your current plan, usage, next charge date and payment history.

• The trial lasts 7 days, no card required.
• When the trial ends, you can choose a monthly plan.
• Prices are in MXN.
• Some plans include a one-time setup fee.

18. Audit log

The system automatically records every important action: creations, edits, deletions, imports, permission changes and password resets.

In Audit you can filter by user, action type, entity, date and range. Each entry includes a diff of the changes made and the source IP.

19. Frequently asked questions