Here's the secret about metadata: it doesn't need to be complicated. Minimal, well-planned metadata with controlled fields is far more powerful than dozens of freeform text fields. When users realize that structured metadata makes filters work like magic-instantly showing exactly the assets they need-everything clicks.
Setting up metadata in Data Dwell is one of the most important decisions you'll make as an admin. A thoughtful metadata plan makes assets easy to find, understand, and use. A hasty one creates confusion and frustration.
The good news? You don't need to solve everything at once. This guide will help you create a minimal, effective metadata structure that works with your team's existing workflows-not against them.
Metadata serves two critical purposes: helping users understand your assets and powering search and filters. Poor metadata planning leads to three common problems:
Assets are hard to find - Without clear, consistent metadata, users resort to browsing folders or guessing filenames. What should take seconds takes minutes. Filters become useless when metadata is inconsistent or missing.
Assets are hard to use - Even when users find an asset, they can't tell if it's approved, outdated, or appropriate for their needs. Missing context means wasted time and potential mistakes.
Workflows break down - Teams need to track review dates, content types, and approval status. When this information isn't captured in metadata, workflows rely on memory, spreadsheets, or email threads.
Metadata solves these problems-but only if it's planned thoughtfully around understanding your users and your assets.
Key Takeaway: Metadata planning isn't a technical exercise. It's about understanding how your team works with assets and what information they need to find and use content effectively.
Before planning your metadata structure, you need to understand how the pieces work together in Data Dwell.
These are the fields you define-the structured information specific to your organization. Data Dwell supports nine field types:
Best for Filters (Use These Most):
Good for Search and Context:
Critical insight: Dropdown, Dropdown Multiselect, Tags, and Autocomplete fields are your most powerful tools. They create consistent, filterable metadata that makes finding assets instant. Free text fields (Textfield, Textarea) are searchable but can't power filters because every user might describe things differently.
Best practice: Use controlled field types (Dropdown, Dropdown Multiselect, Autocomplete) wherever possible to maintain consistency and enable powerful filtering.
Profiles determine which metafields and components are visible on an asset. Think of them as templates that define what metadata structure different types of assets should have.
What Metadata Profiles control:
Every asset must have a metadata profile assigned to it. Different asset types can have different profiles if they need different metadata.
Example: Marketing images might need Campaign Name and Target Audience fields, while legal documents need Compliance Status and Retention Period. Each would have its own metadata profile.
Important note: User permissions also affect what users see and can do with assets. Metadata Profiles control what metadata structure exists on the asset, while permissions control whether users can view or edit that metadata.
Planning tip: Start with one or two profiles. Many organizations can begin with a single "Default" profile and add specialized profiles only when truly needed. Don't create separate profiles for every minor variation-that adds unnecessary complexity.
Data Dwell automatically captures technical information you don't need to recreate:
Planning tip: Focus your custom metadata on business context and organizational needs, not technical details that files already contain or system fields that are managed separately.
Folders and metadata serve different but complementary purposes:
Folders provide hierarchical structure: Marketing → Social Media → LinkedIn shows where content lives organizationally.
Metadata enables cross-folder discovery: "Show me all assets for Campaign X" works regardless of where files are stored. One campaign might have assets in Social Media, Email Marketing, and Print folders-metadata connects them.
Your folder structure gives assets an organizational home. Your metadata makes them discoverable across that structure.
Key Takeaway: Folders = where things live. Metadata = how you find and understand them.
Tags are a special type of metadata-flexible, user-friendly labels that don't require admin setup.
Use tags for: Ad-hoc groupings, project nicknames, seasonal themes, trending topics, emergent categories you didn't predict.
Use custom metafields for: Structured, consistent data that needs to be reportable, used in workflows, or drive important filters.
Tags are great for flexibility, but controlled metafields (dropdowns) are better for critical organizational needs because they ensure consistency.
Users will search using filenames, metadata values, tags, and folder names. Filters rely entirely on structured metadata-especially Dropdown and Dropdown Multiselect fields.
This is where metadata planning pays off: Well-structured metadata turns "I need to find something" into "I can see exactly what I need in seconds" because filters work instantly with controlled field types.
Test this: Before finalizing your metadata structure, imagine specific search scenarios. "I need the Q3 social media assets for Product A" should have a clear path using your planned metadata.
Quick Win: Write down 5 real questions your users ask when looking for assets (like "Where are the approved holiday campaign images?"). Make sure your planned metadata can answer each question through search or filters.
New admins often make one of two mistakes:
Trying to solve everything at once - Creating dozens of metafields to cover every possible scenario before launch. This overwhelms users and often misses the mark because you're guessing at needs rather than solving real problems.
Starting too narrow - Building metadata for one team without considering how it connects to the bigger picture. This creates silos and means rebuilding later when other teams need access.
The sweet spot is starting with an isolated, high-value use case while thinking holistically about how it fits into your organization.
Before creating a single metafield, understand your users and your assets. Ask yourself:
How do people search for assets?
What do people need to know to use an asset?
What workflows exist around assets?
What terms and categories does your team already use?
The answers reveal what metadata you actually need. Start by observing real behavior, not ideal scenarios. Honor existing terminology-forcing new language creates resistance.
Quick Win: Schedule 15-minute conversations with 3-5 users from different teams this week. Ask them to show you how they currently find assets and what information they wish they had. Watch what they do, not just what they say.
Every organization is different, but most benefit from starting with these core elements:
Help users understand what the asset is and what it's for.
Consider (use Dropdown or Dropdown Multiselect):
Clarify when and how assets can be used.
Consider (use Dropdown for consistency):
Make assets discoverable and understandable.
Consider:
Start with 5-8 metafields maximum. You can always add more later. Focus on fields that will be used consistently and drive important filters or workflows.
Quick Win: Draft your first 5-8 metafields right now. For each one, write down: (1) Field name, (2) Field type, (3) Why users need it, (4) What values it should have (for dropdowns). If you can't clearly explain why a field is needed, cut it.
Here's a practical approach to get started:
Pick a specific, high-value use case. Examples:
Create only the metadata needed to solve this specific problem. Launch it. Get feedback.
Quick Win: Identify your single biggest asset findability problem today. Design 3-5 metafields that directly solve it. Test with 10-20 sample assets to validate the fields work as expected.
As you plan your initial metadata, ask:
You're not building everything now-but you're ensuring what you build today won't create problems tomorrow.
After launch, watch how people actually use the system:
Add, refine, or remove metadata based on observed behavior, not assumptions. Empty metafields signal they're not valuable-consider removing them. Frequently requested filters signal you need new dropdown fields.
Quick Win: Schedule a metadata review 30-60 days after launch. Look at 50 random assets and note which metafields are filled in vs. empty. Interview 3-5 power users about what's working and what's missing.
Free text fields (Textfield, Textarea) are searchable but can't power filters. If ten people describe the same campaign differently, your filters become useless.
The problem: Users think they want flexibility, but inconsistent entries make assets unfindable through filters.
What to do instead: Use Dropdown or Dropdown Multiselect for any information that should be filterable. Reserve free text for truly unique descriptions or notes.
Required fields slow down uploads and frustrate contributors who don't know what to enter.
The problem: Contributors face a wall of required fields they can't fill in, so they either enter garbage data or avoid uploading entirely.
What to do instead: Only make fields required if the asset is genuinely unusable without that information. Most fields should be optional. Consider phased requirements-minimal required fields at upload, additional fields required before moving to "Approved" status.
If information is already captured elsewhere (folder structure, filename conventions, EXIF data), don't duplicate it in custom metadata.
The problem: Redundant fields create maintenance burden and confusion when information conflicts.
What to do instead: Use folder paths for hierarchical organization. Use EXIF for technical details. Use custom metadata for business context that isn't captured anywhere else.
Even perfect metadata structure fails if users don't understand it.
The problem: Users don't know which fields to use, what values to select, or why metadata matters. Inconsistent entries follow.
What to do instead: Document your metadata structure-what each field means, when to use it, what values mean what. Train contributors. Make field names and values self-explanatory.
Metadata only works if it stays current.
The problem: Assets get uploaded with metadata but never updated. "Draft" assets from 2022 are still marked as drafts. Nobody knows who's responsible for keeping information current.
What to do instead: Plan for how values will be updated (manually, automatically, through workflows). Assign owners to assets so there's clear accountability for maintaining metadata. Use workflows or scheduled reviews to catch outdated metadata.
Creating dozens of metafields with deeply nested categories before understanding real usage patterns.
The problem: Complex structures overwhelm users and often don't match how people actually work.
What to do instead: Start minimal. Add complexity only when real usage demands it. Five well-used metafields beat twenty poorly-understood ones.
Remember: metadata planning isn't about creating the perfect system on day one. It's about creating a solid, minimal foundation that can evolve with your organization's needs. Well-planned metadata with controlled fields makes filters work like magic-and that's when users truly appreciate what you've built.