Core Concepts

TimeTiles organizes data in a hierarchy designed to scale from a single dataset to a multi-tenant platform serving multiple organizations.

Data Hierarchy

Site (optional — multi-tenant)
  └── Catalog (project)
        ├── Dataset (data source)
        │     └── Event (data point)
        └── View (saved display configuration)

Events

An event is a single data point — something that happened at a specific place and time. Each row in your CSV becomes an event.

Every event has three required fields:

Beyond these, events carry any additional fields from your data: description, category, source, severity, URLs, images — whatever your schema includes. These become filterable and searchable in the interface.

Datasets

A dataset is a collection of events that share the same schema (field structure). Datasets correspond to a single data source or import.

• Each CSV/Excel upload creates or updates a dataset
• Events within a dataset share the same field definitions
• Scheduled URL imports keep a dataset updated automatically
• Users can filter the explore interface by dataset

Example: A “2024 Election Coverage” catalog might contain datasets for “Rallies”, “Polling Stations”, and “Incident Reports” — each with different schemas.

Catalogs

A catalog is a project-level container that groups related datasets. It’s the unit of access control and the entry point for exploration.

• One catalog per project, investigation, or topic
• Controls visibility (public or private) and permissions
• Aggregates events from all its datasets into a single explorable interface
• Has its own branding and theme settings

Views

A view is a saved display configuration within a catalog. Views define how data appears when someone opens the explore interface — which datasets are visible, what map bounds to show, and which filters are active.

• Create focused views for different audiences or stories
• Each view has its own shareable URL
• Useful for curating specific perspectives on the same underlying data

Sites

A site is an optional multi-tenant layer. When enabled, a single TimeTiles installation can serve multiple independent instances, each with its own domain, branding, and catalogs.

Most deployments don’t need sites — a single-site setup works out of the box.

Data Flow

Upload CSV/Excel/ODS or configure scheduled URL
    ↓
Schema detection — TimeTiles identifies field types and suggests mappings
    ↓
You review and approve the schema
    ↓
Geocoding — text addresses are converted to coordinates
    ↓
Events are created in the dataset within a catalog
    ↓
Events appear on the map with timeline, filters, and charts

For details on each processing stage, see Data Import.

Organizing Your Data

One catalog per project. Keep related datasets together under a single catalog — this is what your audience will explore.

One dataset per source. Each import or data source gets its own dataset. This keeps schemas clean and makes it easy to update individual sources.

Consistent fields within datasets. Use the same date format, consistent category names, and clear titles. TimeTiles generates filters from your data — consistent values produce better filters.

Include metadata for filtering. Categories, tags, and any other classifying fields become interactive filters in the explore interface. The more structured metadata you include, the richer the exploration experience.

Next Steps

• Getting Started — Import your first dataset
• Data Import — Formats, geocoding, scheduling, and schema handling
• Comparison — How TimeTiles compares to similar tools