CDP RFP Template: SaaS and Warehouse-Native Evaluation (2026)
A Customer Data Platform RFP that puts the SaaS-vs-composable architecture question front and centre, instead of letting the incumbent SaaS CDPs run away with the process. Sections covering ingestion, identity, activation, governance, and pricing format that surfaces the real 5-year cost.
Section 1: Organisation and data context
- Company description and primary customer type (B2C, B2B, B2B2C, marketplace).
- Current data stack: warehouse (Snowflake / BigQuery / Databricks / Redshift), pipeline (Fivetran / Airbyte / native), BI (Looker / Tableau / etc.).
- Estimated monthly tracked users (MTU) at year-1 and projected year-3.
- Event volume: events per month at year-1; ratio of server-side to client-side.
- Compliance scope: GDPR, CCPA, HIPAA, region-specific data residency.
- What you have today and what is broken (no CDP, legacy CDP, point-to-point integrations sprawl).
Section 2: Architecture preference (explicit)
State whether you are open to both SaaS and warehouse-native composable architectures, or only one. If only one, say why - it lets vendors not in your preferred camp self-disqualify rather than waste a response.
SaaS CDP
Full-stack: ingest, store, resolve identity, activate. Examples: Segment, mParticle, Tealium, Treasure Data, Adobe RT-CDP, Salesforce Data Cloud. Faster time to value, higher subscription cost, owns a copy of your data.
Warehouse-native composable CDP
Activation layer over your existing warehouse. Examples: Hightouch, Census, RudderStack (composable mode), Polytomic. Your warehouse is the source of truth; CDP handles only the activation. Lower subscription cost, more engineering ownership.
Section 3: Ingestion requirements
- Server-side event SDKs required: Node.js, Python, Ruby, Java, Go, .NET (list the ones you need).
- Client-side SDKs required: JS for web, native iOS / Android, React Native if applicable.
- Mobile-specific: deep-linking, attribution, push notification support.
- SaaS source connectors needed: [list every named tool] - confirm presence and freshness of each.
- Warehouse / database source connectors: [list specific databases] with replication frequency.
- EU / regional data residency requirement: ingest endpoint in EU or US (or both).
- Schema enforcement: blocking vs. warning on non-conforming events at ingest.
Section 4: Identity resolution
- Deterministic merge on shared identifier (email, hashed phone, user_id).
- Probabilistic / fuzzy stitching when deterministic identifiers do not match.
- Anonymous-to-known stitching on first identify event in a session.
- Merge / unmerge tooling for fixing wrong stitches.
- Cross-device tracking: web visit + mobile app + customer support call → one profile.
- Identity graph export to warehouse for analyst-level access.
Section 5: Activation destinations
List every system that needs to consume CDP audiences or events. Specify the freshness requirement per destination - some need real-time, most do not.
- Marketing automation: [your tool] - audience sync, event-triggered sends.
- Email service provider: [your ESP] - list sync, suppression.
- Ad platforms: Google Ads, Meta Ads, LinkedIn Ads, TikTok - audience sync with hashed PII matching.
- Push notification provider: [your tool].
- SMS / messaging: [your tool].
- Customer support: [Zendesk / Intercom / Help Scout] - profile enrichment with traits.
- Product analytics: [Amplitude / Mixpanel / native].
- Personalisation / experimentation: [Optimizely / LaunchDarkly / native].
- Data warehouse: full event stream + audience membership table for analyst use.
Section 6: Governance and compliance
- Consent management: integration with [OneTrust / TrustArc / native CMP].
- Right-to-be-forgotten: end-to-end deletion across the CDP and all activation destinations.
- PII handling: field-level encryption, role-based access, audit log.
- Data residency: EU-only path for EU-resident user data.
- Compliance certifications: SOC 2 Type II, GDPR, CCPA, HIPAA (if applicable).
Section 7: Pricing format required
- Pricing model: MTU-based, event-volume-based, destination-based, or seat-based.
- Base platform fee separate from usage fee.
- Overage rate per MTU or per million events.
- Per-destination cost (if applicable).
- Year-2 and year-3 escalators.
- Implementation cost (vendor-led or partner-led).
- 5-year TCO at your projected growth curve.