Connect Data Sources
Connecting data sources is the first operational step in building a trusted DriveMetaData workspace. The goal is not to connect every system immediately. The goal is to connect the sources needed for clear reporting and safe activation.
Plan the connection
Before requesting a connector or upload, document:
- Source owner.
- Business purpose.
- Data fields needed.
- Refresh cadence.
- Identifier fields.
- Consent or suppression fields.
- Expected event or record volume.
- Downstream reports and segments that will use the data.
Recommended first sources
Most teams start with:
- Web or app behavioral events.
- Ad platform campaign and cost data.
- CRM or customer account data.
- Conversion, order, subscription, or booking data.
- Offline conversions or warehouse enrichment if needed.
Field checklist
| Field type | Examples | Why it matters |
|---|---|---|
| Identifier | User ID, CRM ID, anonymous ID, device ID | Profile stitching and reporting |
| Timestamp | Event time, import time, conversion time | Journey order and attribution |
| Event name | Signup, purchase, renewal, app open | Funnel and segment logic |
| Campaign fields | Source, medium, campaign, ad group, creative | Attribution and spend analysis |
| Value fields | Revenue, currency, quantity, plan | ROAS, LTV, and revenue reports |
| Consent fields | Email opt-in, SMS opt-in, region | Activation eligibility |
Connection workflow
-
Create the source request. Include owner, purpose, fields, and expected sync frequency.
-
Configure access. Use the approved connection method for the source. This may be a native connector, file import, SDK, server-side event stream, or warehouse sync.
-
Map fields. Match source fields to DriveMetaData customer, event, campaign, and value fields.
-
Run a sample sync. Validate a small batch before relying on reports or audiences.
-
Review data quality. Check timestamps, event counts, null fields, duplicates, and identifier coverage.
-
Publish for use. Mark the source as approved only after the owning team accepts the validation results.
Validation questions
- Do total event counts look reasonable by day?
- Are required fields populated?
- Are campaign values consistent with the naming convention?
- Are customer identifiers stable enough for profile resolution?
- Are test records excluded or labeled?
- Does the sync cadence match reporting needs?
Troubleshooting
| Problem | Likely cause | Next step |
|---|---|---|
| Missing customer profiles | Identifier field not mapped or source has anonymous-only data | Review identity fields and login events |
| Campaign data is fragmented | Inconsistent UTM or campaign names | Normalize naming and update campaign taxonomy |
| Revenue is overstated | Duplicate events or refunds not mapped | Validate deduplication and adjustment logic |
| Segment does not update | Source refresh cadence is too slow | Confirm sync schedule and segment lookback |
| Activation match rate is low | Destination requires a different identifier | Review channel requirements before export |
Handoff template
Use this template when asking another team to connect a source:
Source:
Owner:
Business question:
Fields required:
Identifiers available:
Refresh cadence:
Consent or suppression fields:
Reports or segments impacted:
Launch date:
Review owner: