CDP Buying Guide for Teams Without a Data Engineer

Hightouch published a CDP RFP template with nine evaluation sections covering architecture, event collection, identity resolution, audience segmentation, data activation, data onboarding, analytics, and AI. It is a thorough document. It is also written for a buyer with a data engineering team, a Snowflake instance, and a six-figure annual budget. If that describes your team, use their template. If you are a RevOps lead, a marketing ops manager, or a founder trying to get Stripe data into your CRM, you need a different CDP buying guide.

This guide is the evaluation checklist for teams under 200 people. It covers the five CDP selection criteria that actually matter when you don't have a data engineer: setup time, pricing transparency, warehouse requirement, sync modes, and self-serve access. For background on what CDPs are and how the category works, see What Is a Customer Data Platform?.

The five questions every CDP buying guide should answer before you start evaluating

Most CDP buying guide content starts with features: does the platform support identity graphs? Can it build lookalike audiences? Does it have AI decisioning? These are valid questions for enterprise buyers spending $100,000+/year. They are the wrong starting point for a 30-person company.

Before you evaluate any feature, answer five foundational questions:

1. What problem are you actually solving? "Our CRM doesn't have billing data" is a sync problem. "We need to stitch anonymous website visitors across devices" is an identity problem. These have very different solutions at very different price points.

2. Do you have a data warehouse? If no, half the CDP market is eliminated immediately. Composable CDPs require one. Traditional CDPs include their own storage but cost 10x more.

3. Do you have a data engineer on staff? If no, any product that requires SQL, dbt models, or SDK instrumentation is not a realistic option. The evaluation should include only products your existing team can operate.

4. What is your budget for data infrastructure? CDPs range from $0 (direct sync free tiers) to $500,000+/year (enterprise CDPs). Knowing your range eliminates 80% of the market and prevents wasted sales calls.

5. How fast do you need to be live? A 30-minute setup and a 6-month implementation project are not competing options. They solve different problems for different teams.

The answers to these five questions define your shortlist before you evaluate a single feature.

CDP buying guide criteria for teams under 200 people

Enterprise CDP evaluation criteria focus on scalability, governance, and compliance. Small-team CDP selection criteria focus on something else entirely: can you get value today without hiring someone to make it work?

Here are the five criteria that matter, ranked by importance:

Setup time is first because it determines whether you can evaluate the product at all. If a vendor needs two weeks of scoping calls before you can test it, you are evaluating their sales process, not their product. A tool designed for small teams lets you connect your first two tools and verify data flow in under an hour.

Pricing transparency is second because hidden pricing always means enterprise pricing. Every vendor that publishes pricing on their website serves teams your size. Every vendor that hides pricing behind a sales form has a minimum contract that exceeds what most small teams will spend on data infrastructure in a year.

Warehouse requirement matters because it determines your total cost and complexity. A composable CDP costs $1,000-5,000/month plus $500-5,000/month in warehouse compute plus dbt Cloud plus a person who can write SQL. A direct sync tool costs $0-100/month with no prerequisites.

CDP buying guide red flags: warehouse prerequisites, opaque pricing, mandatory sales calls

When evaluating data sync or CDP vendors, watch for patterns that signal the product is not built for your team size. These are not subjective preferences. They are structural indicators that the vendor targets buyers with dedicated data teams and six-figure budgets.

"Connect your data warehouse to get started." This is the first screen in composable CDP onboarding. If you don't have a warehouse, the product cannot function. This is not a limitation you can work around. It is the architecture.

"Contact sales" on every pricing tier. If the pricing page has no numbers, the minimum contract exceeds your annual data infrastructure budget. Vendors hide pricing when their lowest tier costs more than small teams expect to pay.

Mandatory demo before trial access. This means the product requires guided setup. Products built for self-serve users give you a free tier and let you connect tools immediately. Products that gate access behind demos need a sales engineer to configure the environment.

SDK instrumentation required. "Install our JavaScript snippet on your website" or "Add our SDK to your mobile app" means you need an engineer to implement and maintain the integration. For teams that just need Stripe data in their CRM, SDK instrumentation solves a problem you don't have.

Per-MTU or per-event pricing. Monthly tracked users (MTUs) and event-based pricing scale unpredictably. A traffic spike doubles your bill. Per-sync-action pricing with published tiers lets you predict costs before you commit.

CDP requirements checklist: setup time, pricing, sync modes, warehouse requirement

Use this CDP requirements checklist during your evaluation. Score each vendor on a 1-3 scale for each criterion and compare totals.

Setup and access:

- Can you sign up and connect two tools without a sales call? (Yes = 3, Demo required = 1)

- How long from signup to first data flowing? (Under 1 hour = 3, 1 day = 2, 1+ weeks = 1)

- Does the product work without a data warehouse? (Yes = 3, Optional = 2, Required = 1)

Pricing and commitment:

- Is pricing published on the website? (Full tiers = 3, Partial = 2, Hidden = 1)

- Is there a free tier with real functionality? (Yes = 3, Limited trial = 2, No = 1)

- Can you upgrade without a sales call? (Self-serve checkout = 3, Sales-assisted = 1)

Sync capabilities:

- How many sync modes? (4 modes: update, create, update-or-create, mirror = 3, 2 modes = 2, 1 mode = 1)

- Does it support bidirectional sync? (Yes = 3, One-way only = 1)

- Does it track field-level changes? (Yes, with old/new values = 3, Record-level only = 1)

Error handling and observability:

- What happens when a record fails to sync? (Dead letter queue = 3, Retry only = 2, Silent drop = 1)

- Can you see sync history per record? (Yes = 3, Aggregate only = 2, No = 1)

- Does it show which fields changed and when? (Yes = 3, No = 1)

A score of 30+ out of 36 means the product is built for self-serve teams. Under 20 means it is an enterprise tool that will require dedicated staff to operate.

How to run a CDP proof-of-concept in one day with direct sync

Enterprise CDP evaluations take 6-12 weeks. They involve stakeholder interviews, use case roadmaps, phased deployment plans, and implementation partners. That process exists because enterprise CDPs require it. The architecture demands warehouse configuration, SDK instrumentation, identity graph setup, and data modeling before a single record syncs.

Direct sync eliminates every one of those prerequisites. Here is a same-day proof of concept:

Morning: scope and connect (2 hours). List the tools that hold customer data. Pick the highest-value sync: for most teams, that is billing to CRM (Stripe to HubSpot) or database to email (Postgres to Mailchimp). Sign up for a sync tool with a free tier. Authenticate both tools with API keys or OAuth. Select record types. Map 5-8 fields: subscription status, plan name, renewal date, lifetime revenue, last login.

Afternoon: sync and validate (2 hours). Run the first sync. Open HubSpot (or your CRM) and verify that Stripe billing data now appears on contact records. Change a subscription in Stripe. Wait for the next sync cycle. Confirm the change propagated. Test what happens when you create a field type mismatch: does the record land in a dead letter queue, or does it vanish?

End of day: decision. You now have real data flowing between production tools. You know exactly how long setup takes, what the sync latency is, how errors are handled, and whether the field mapping covers your use cases. Compare that experience against the enterprise CDP evaluation timeline: 6 weeks of scoping before you see your own data moving.

Oneprofile handles every step in this proof of concept. Connect your tools with API keys or OAuth, map fields visually, choose a sync mode (update, create, update-or-create, or mirror), and data flows on a schedule you control. Property-level change tracking means only changed fields sync. Dead letter queue catches failed records for investigation. Free tier covers the entire evaluation. No warehouse, no SDK, no sales call. Your CDP proof of concept is a 4-hour afternoon project.