Every major CDP vendor publishes an RFP template. Most of them cover the same ground: identity resolution, data integrations, privacy compliance, support, and pricing. These are necessary categories. They are not sufficient for retail.
Retail has specific data problems that generic RFP frameworks don't address. A shopper who buys in-store with a credit card, browses on mobile without logging in, and purchases online through a loyalty account generates three identity fragments that most CDPs struggle to connect. A loyalty program with millions of members accumulates duplicate profiles that inflate program costs and fragment reward histories. A retail media network needs advertiser-grade audiences backed by identity resolution confidence levels that generic audience builders can't provide.
When the RFP doesn't ask about these problems, the vendor doesn't have to prove they can solve them. And you don't find out until after implementation.
We built the Retail CDP RFP Guide to close that gap. It's a practical evaluation framework organized around the four outcomes that matter most to retail brands, with table stakes vs. next level benchmarks and ready-to-use RFP questions designed to surface the capabilities that separate a retail-ready CDP from a generic one.
Here's what generic RFP templates miss and what the retail guide covers.
Generic RFPs test identity resolution. Retail RFPs test omnichannel identity.
Most CDP RFP templates include questions about matching rules, deduplication, and data source coverage. These are important, but they test identity resolution in the abstract. For retail, the relevant question is whether the platform can resolve identity across the specific systems and scenarios that generate fragmented shopper data.
Can it handle POS transactions where the only identifier is a credit card token? Can it retroactively connect a guest checkout to a loyalty profile when the customer enrolls weeks later? Can it distinguish between a person and a household, or between an individual and a loyalty account, so that merges don't corrupt reward balances?
Retail also has an identity problem that other industries don't: different teams need different identity strategies running simultaneously. Marketing wants broad probabilistic matching to maximize addressable audience reach. Loyalty operations wants conservative deterministic matching to prevent account-level errors. Retail media wants high-confidence identity to satisfy advertiser quality expectations. Clienteling needs traceable matching appropriate for face-to-face interactions.
A generic RFP asks whether the vendor does identity resolution. A retail RFP asks whether the vendor can run multiple identity resolution strategies concurrently on the same data, each tuned to a different business context, without duplication. That's a question most vendors can't answer well, and the ones who can will differentiate themselves immediately.
Generic RFPs mention AI. Retail RFPs test whether AI is production-ready.
AI is on every CDP vendor's feature list. But for retail marketing teams operating on compressed seasonal timelines, the question isn't whether the platform has AI. The question is whether the AI is governed, auditable, and actually useful in day-to-day campaign execution.
A retail buyer evaluating AI should ask: Does the AI operate on identity-resolved profiles, or on raw fragmented data? Can a marketer describe a campaign goal in plain language and have the system propose the segments and journey logic? Does the system produce a reviewable plan before executing? Can the outputs be fully reversed? Does the AI proactively surface recommended actions based on current customer trends, or does it only respond when a user asks?
These governance questions matter because AI applied to fragmented data produces inaccurate results at speed. A churn model trained on incomplete purchase histories undervalues your best customers. A segmentation engine that can't distinguish a new buyer from a returning guest-checkout shopper puts the wrong people into the wrong journeys. The speed of AI makes these errors harder to catch and more expensive to fix.
The retail guide includes a full set of AI evaluation criteria organized as table stakes vs. next level capabilities, with specific questions that test whether the vendor's AI is enterprise-ready or still in demo mode.
Generic RFPs skip retail media entirely.
This is the biggest gap. No major CDP vendor's generic RFP template includes a section on retail media networks and audience monetization. Yet for many enterprise retailers, the retail media network is one of the fastest-growing revenue streams in the business, and the CDP is the infrastructure layer that determines audience quality.
When evaluating a CDP for retail media, the RFP should test whether the platform can produce audiences with explicit identity resolution confidence levels (high-confidence for premium placements, broader reach for prospecting). It should ask about activation speed: can a new audience segment reach the ad environment on the same day it's created, or does it take the multi-day cycles typical of legacy onboarding? It should probe closed-loop measurement: can the platform link ad exposure to purchase events at the SKU level? And it should test whether first-party audience segments can be syndicated to public ad marketplaces without requiring the retailer to build and operate a full media network.
These questions separate CDPs that treat retail media as a checkbox from those that treat it as a first-class capability. The retail guide dedicates a full section to this area, with table stakes vs. next level benchmarks and specific questions to include in your RFP.
Generic RFPs measure time to value generically. Retail RFPs measure it in use cases.
Most CDP RFP templates ask about implementation timelines and support. The retail guide goes further by asking vendors to demonstrate specific retail use cases out of the box: loyalty deduplication, omnichannel personalization, abandoned cart recovery with cross-channel identity verification, churn prevention with value-based prioritization, and sell-through optimization that matches overstock categories to high-affinity buyers.
These use cases follow a crawl/walk/run maturity model. Crawl use cases (customer suppression, welcome series, basic analytics) should be achievable within weeks. Walk use cases (lookalike expansion, conversion API integration, cross-category discovery journeys, churn propensity scoring) represent the next tier. Run use cases (real-time personalization, event-triggered activation, audience monetization, bring-your-own predictive models) are what separate a CDP that's delivering returns from one that's still being configured.
Asking vendors to map their capabilities against this maturity model gives you a clear picture of where you'll start, where you'll grow, and whether the platform can support both.
Built for the cross-functional team
One more thing generic RFP frameworks miss: different stakeholders need different sections. A CMO evaluating campaign ROI cares about different capabilities than a CTO evaluating architecture, and both care about different things than a VP of Retail Media evaluating audience quality.
The retail guide includes a persona mapping section that directs each buyer to the sections most relevant to their role. It's designed to be shared across the evaluation team, not handed off to procurement as a standalone document.
Download the guide
If you're evaluating CDPs for your retail brand, the Retail CDP RFP Guide gives you the framework, benchmarks, and questions you need to run an evaluation that actually tests what matters for retail.
Download the Retail CDP RFP Guide →
For a personalized walkthrough of how these capabilities work in practice, contact us for a demo.