Zeotap

Editor’s note: This Zeotap review is based on my hands-on test of the Zeotap CDP, from October 2026. I weighed pricing and buying routes alongside setup, identity resolution, audiences and journeys, consent, data residency, and composable deployment.

Quick verdict: I recommend Zeotap for mid-market and enterprise brands with consent-heavy EU data, particularly in telecom, retail, banking, and media. It combines identity resolution and journeys with consent controls, either managed by Zeotap or inside Snowflake, but buying requires a demo and contract, with no public pricing or trial.

In this Zeotap review, I’ll take a closer look at Zeotap’s pricing, features and deployment options, so you can see exactly whether its consent controls and warehouse-native architecture justify a contract for your team.

Within Panoply’s “Reverse ETL and CDPs” directory category, I favor Zeotap when activation permissions matter as much as moving customer records.

Key Takeaways

• Six modules cover ingestion through administration, with 150+ packaged destinations and 280+ composable destinations.
• Ada builds editable audiences from plain-English descriptions.
• The Snowflake Native App, launched August 5, 2026, uses Snowflake credits and Marketplace billing.
• No public price or trial means package allowances need to be settled before signing.
• Batch attributes refresh daily, usually taking 24 to 36 hours end to end.

Pros and Cons

How Much Does Zeotap Cost?

Zeotap publishes no price list, free plan, or trial. Buying starts with Book a Demo and a contract; Calculated Attribute counts and other allowances depend on the pricing package, with adjustments handled through a customer success manager, or CSM.

Snowflake credits can fund the Native App, but compute and warehouse usage also consume those credits. I favor this buying route for existing Snowflake customers who can apply committed spend through Marketplace Capacity Drawdown.

• Packaged CDP (contract): This is the Zeotap-managed route for teams buying a managed customer data platform. Some capabilities require an Enterprise plan, including channel and destination restrictions for Catalogue attributes. Those restrictions also require the CDP Admin role.
• Composable CDP / Snowflake Native App: The Native App runs in your Snowflake account and uses Snowflake credits plus Marketplace billing. Snowpark Container Services compute and warehouse usage consume your credits. Buyers start by requesting a working session.
• Google Cloud Marketplace: This route puts Zeotap fees on the Google Cloud invoice through consolidated billing, introduced with the 2021 Marketplace launch. I prefer it for teams consolidating procurement through Google Cloud; it does not provide a public Zeotap price.

Buying route	How you pay	Published price	Trial
Packaged CDP	Contract via sales	No public price	No trial
Snowflake Native App	Snowflake credits and Marketplace billing	No public price	No trial; working session
Google Cloud Marketplace	Google Cloud invoice	No public price	No trial

Is Zeotap Good Value for Money?

• A quote is essential: I cannot call Zeotap cheaper than RudderStack, whose Free plan includes 250K events/month and whose Growth plan costs $265/month for 1 million events/month. Tealium Data Cloud Activation costs $1,000/month billed annually.
• Committed spend can help: Marketplace Capacity Drawdown lets Snowflake customers apply an existing commitment. Include container compute and warehouse consumption in the operating budget.
• The bundle needs a purpose: Zeotap combines identity resolution, Consent Orchestration, and Journeys. For warehouse activation alone, my starting point is Hightouch’s free Reverse ETL tier with up to 2 active syncs.
• Enablement affects access: A/B testing requires opt-in, and Segment Membership Calculated Attributes require enablement. Complex NCE Join Support remains beta, with early access through a CSM.

Reviewer’s Notes: I recommend requesting package limits in writing before signing, including Calculated Attribute counts, required destinations, region coverage, and opt-in A/B testing. For a Snowflake team, request a Native App quote against committed spend and a breakdown of compute consumption.

My Experience With Zeotap

I followed the packaged CDP from Create Source to activation, with Catalogue mapping and identity configuration between those endpoints. I prioritized the ID strategy because later changes apply only to newly ingested data.

Module	Product area
Integrate	Ingestion and destinations
Unify	Catalogue, ID Strategy, Calculated Attributes
Segment	Rule-based audiences and Ada
Orchestrate	Journeys
Protect	Consent, Customer 360, Data Lifecycle
Admin	Administration

Creating My First Source

I opened Create Source and selected Server to Server. This category uses an HTTP API with a write_key, rather than a website tag or mobile SDK.

I opened the IMPLEMENTATION DETAILS tab for the implementation guide, then checked rows on PREVIEW DATA. The preview gave me a place to inspect incoming fields before mapping them to the Catalogue.

• Website Events: Web JS and pixel files.
• App Events: Android SDK, including Amazon Fire TV, plus iOS SDK and React Native.
• Flat Files: Zeotap Google Cloud Storage and SFTP.
• Tag Managers: Adobe Experience Platform Tag Extension and Google Tag Manager.
• Warehouse connections: BigQuery, Snowflake, and Databricks, alongside Amazon S3 ingestion.

Other integration options include Braze Currents, Dotdigital, HubSpot, and Airship and Cleverpush.

Mapping Data to the Catalogue

I clicked Map to Catalogue from PREVIEW DATA. The same action was available under CATALOGUE MAPPING, so I could enter mapping from either view.

I used + ADD MAPPING for consent purposes and marketing preferences, then checked REVIEW MAPPING for warnings. Consent Orchestration uses those mapped permissions when an audience reaches activation.

I also checked the three-letter country-code mapping used to resolve storage into EU, US, or IN. That field controls residency routing, with IP address as the fallback when the country code is absent.

Setting the ID Strategy

I opened Unify and the IDENTITIES tab to inspect the default, “Identify and link using all IDs.” Custom and High-Fidelity strategies instead classified identifiers as Primary or Secondary, with priority available up to level five.

I paid particular attention to the default’s irreversible choice and the restriction of subsequent strategy changes to newly ingested data. Existing profiles do not rebuild when that setting changes.

Building and Activating an Audience

I described male loyalty members who had added to cart in the last 30 days to Ada and clicked Generate Audience. Ada represented the criteria with a profile-attribute block and an event block.

I edited the generated blocks before clicking Activate Now, then Link to Destination. Set A/B Test was a separate activation option restricted to accounts that had opted in.

Reviewer’s Notes: I found the Catalogue and ID strategy stages more consequential than the Ada prompt. My setup priority is identifier classification before ingestion: later strategy changes do not rebuild existing profiles, and an editable audience cannot correct an earlier identity-model decision.

Identity Resolution and the Unified Profile

Zeotap’s UCID is persistent, not immutable. When a record matches multiple profiles, Zeotap keeps the first-created UCID as the master and merges the other profiles into it.

The identity graph belongs to an individual account. Zeotap does not resolve identifiers or profiles across organizations, so the unified profile is account-specific rather than a shared cross-company customer record.

As covered in My Experience With Zeotap, the default uses every received identifier to match or create a profile. I prioritize deliberate identifier classification before ingestion because strategy changes do not repair existing matches.

• Custom and High-Fidelity: Identifiers start as Secondary and can be classified as Primary.
• Priority: Identifier ordering supports up to five levels.
• Strategy changes: Changes apply to newly ingested data, not retroactively to existing profiles.

Calculated Attributes add derived traits, but batch supports 12 operators while real-time supports only three.

Mode	Operators	Lookback	Freshness
Batch	12	Up to 540 days	Daily; usually 24 to 36 hours end to end, worst case about 72 hours
Real-time	3: Count, Sum, Unique List	Up to 24 hours / 30 days / 4 weeks	Real time

Batch attributes suit longer-history analysis because they support up to 540 days of lookback. I reject them for campaigns requiring a newly computed trait within the same day: refreshes run daily, with worst-case freshness of about 72 hours.

• Conditions: Each Calculated Attribute supports up to 10.
• Group By: Batch attributes support up to 3 fields; real-time attributes do not.
• Attribute counts: The pricing package determines the allowance.
• Segment Membership: This Calculated Attribute requires enablement.

For warehouse-first teams, Zeotap identity resolution runs its connected-components algorithm as SQL inside the warehouse. Merge rules and identifier families complement probabilistic matching and golden records, with results written back to the warehouse.

I favor Zeotap for a combined identity-and-activation requirement; Hightouch’s free Reverse ETL tier is my narrower starting point when the requirement is up to 2 active syncs.

Audiences, Ada and Journeys

Ada generated editable profile and event blocks, so I could inspect the conditions before activation.

• Events: Behavioral conditions.
• Profile attributes: Customer properties.
• Calculated attributes: Derived traits.
• Consent: Permission requirements.
• Marketing preferences: Channel choices.

Ada runs on Google Vertex AI. ZeoAI also includes automated data mapping, propensity models using Vertex AI and BigQuery ML, and an RFM model.

The Native App runs propensity, churn, lifetime value, and lookalike models on Snowflake Cortex AI.

Zeotap MCP remains beta, enabled per organization with a server URL supplied by the Zeotap team. Claude, ChatGPT, or another MCP-compatible assistant can inspect schema and audience or journey health, then draft or edit audiences.

Journey Traces, shipped in Q1 2026, replays a user’s path by UUID, including input and destination output payloads plus a Customer 360 link.

Error Logs contain nine error codes, split into persistent and transient errors, with response thresholds:

• Under 1%: Normal.
• 1 to 5%: Investigate within 24 hours.
• 5 to 20%: Critical; investigate within 2 hours.
• Over 20%: Pause and escalate.

Journeys also has templates and folders. Q1 2026 added My Roles and Folder Access Management, with one role per user per organization.

Folder access supports All Folders, Include, and Exclude.

Gateway frequency capping covers seven messaging destinations:

• Emarsys
• Optimizely Campaigns
• Iterable
• Elaine
• WhatsApp
• Inxmail API: Journeys
• SFMC Journey Builder

Counters reset at 00:00 UTC. Data exports, audience syncs, and non-messaging journeys remain unaffected, so I reject Gateway as a universal activation cap.

Journey Traces is my preferred tool for investigating an individual customer’s path because the payloads and Customer 360 link connect journey behavior to that user’s profile.

Consent, Security and Data Residency

Consent Orchestration can reduce 10,000 qualified profiles to 3,000 with Email preference before activation to Braze. I consider that permission enforcement a stronger buying argument than destination count alone.

• Audience level: Consent rules apply to every segment.
• Channel level: Rules can be scoped to a channel, destination, and mapping.

Zeotap meets GDPR, ISO 27001 and CSA STAR standards.

Packaged storage uses the three-letter country code set during mapping, with IP address as a fallback. The storage labels cover these areas:

• EU: Includes Africa and Russia.
• US: Covers the Americas.
• IN: Covers Asia, the Middle East, and Oceania.

I favor the Snowflake Native App when residency must follow an existing Snowflake account region. For Packaged, include the country-to-region mapping in your residency requirements rather than relying on the EU label alone.

Profile deletion is not immediate: a successful API response confirms receipt, while full deletion takes a couple of hours. The account Admin receives an email, and a deleted-profile log remains until the end of the next year.

The deleted-profile log is not connected to any activation system.

For Composable, my security checklist includes:

• Transport: TLS 1.2+.
• Credentials: A secret manager, separate from the metadata database.
• API keys: SHA-256 hashes.
• Traceability: A full trail of every user action, API call, sync run, and AI operation.
• Data handling: Zeotap stores metadata, not customer records, acts as a data processor, and offers a DPA.

Packaged vs Composable: InfraFlex and the Snowflake Native App

InfraFlex offers Packaged or Composable deployment: Packaged is Zeotap-managed, while Composable is warehouse-native and zero-copy. I prioritize the customer-record location over audience interfaces when choosing a deployment.

Composable supports six warehouses:

• Snowflake
• BigQuery
• Databricks
• ClickHouse
• Redshift
• Spark Lakehouse

Identity resolution runs as warehouse SQL, including probabilistic matching and golden records. Loaders bring in SaaS data from Salesforce, HubSpot, and Stripe.

Zeotap Agent profiles raw tables, writes Data Prep recipes, and builds audiences on the Audience Canvas. The initial pipeline sequence is:

1. Connect the warehouse.
2. Create a SQL model.
3. Set up a destination.
4. Run the first sync.

The Snowflake Native App launched August 5, 2026, running the whole composable CDP inside the customer’s account on Snowpark Container Services. Its interface uses a small Zeotap control plane at agentic.zeotap.com.

The default application name is ZEOTAP_COMPOSABLE_CDP. Installation takes about five to ten minutes of operator time.

The Native App launched with 250+ connectors; the broader composable product has 280+ destinations. Residency follows the Snowflake account, making this my preferred Zeotap route for Snowflake-centered banking, insurance, telecom, and public-sector teams.

Zeotap retained its CDP when the Zeotap-Data sale to Roqad was announced September 2, 2025. It also retained exclusive identity graph access for five years.

Reviewer’s Notes: I recommend Packaged for teams wanting Zeotap-managed operation, and Composable for teams keeping the warehouse as their customer system of record. The Native App is my preferred deployment for an existing Snowflake account because the composable CDP runs inside that account.

• Best for: Enterprises choosing between managed and warehouse-resident deployment.
• Skip if: A free activation tier covers the requirement, such as Hightouch’s allowance of up to 2 active syncs.

How Does Zeotap Compare to Competitors?

I shortlist Zeotap for consent-aware customer activation, but favor Zeotap alternatives with published entry prices when a free tier or a known starting budget takes priority over its integrated consent controls.

• Twilio Segment: I shortlist it for event collection through Customer Data Pipeline or broader CDP needs through Customer Data Platform. Pricing uses monthly tracked users, or MTUs; the CDP plan has custom volume pricing. Request the CDP quote when pipeline access alone does not cover the requirement.
• RudderStack: My budget-first shortlist for warehouse-first event pipelines starts here. Free costs $0 for 250K events/month; Growth costs $265/month for 1 million events/month, with 15% off annually. Enterprise has custom pricing, so I prefer Free or Growth when their event allowances fit.
• Tealium: I shortlist it for enterprise CDP and tag management needs. Data Cloud Activation costs $1,000/month billed annually, with 300+ connectors and unlimited users. Data Collection has 1,300+ pre-defined integrations; Data Collection and CDP tiers require quotes.
• Hightouch: I shortlist its composable CDP and reverse ETL for teams with existing warehouse models. I prefer its free Reverse ETL tier when up to 2 active syncs cover the requirement. It includes unlimited destinations and seats, with usage-based pricing and no MTU limits.
• Amperity: Standard and Enterprise editions use Amps for usage pricing and require sales contact. I shortlist it for enterprise CDP needs with a retail focus. Neither edition has a public price.
• Treasure Data: Now branded Treasure AI, it offers Complete CDP and Composable CDP products. I shortlist it for teams evaluating both deployment modes, but it has no public price, so it does not solve Zeotap’s pricing-transparency drawback.
• mParticle: Now Rokt mParticle, it combines real-time and warehouse architecture. I shortlist it when that hybrid approach is a buying priority. It has no public price, making it another sales-led comparison.

Tool	Published entry price	Pricing unit or terms	Best fit
Zeotap	No public price	Contract/package; Snowflake billing available	Consent-aware activation
Twilio Segment	CDP custom volume pricing	MTUs	Event collection through Customer Data Pipeline and broader CDP needs
RudderStack	Free $0; Growth $265/month	Events/month	Warehouse-first event pipelines on a budget
Tealium	Activation $1,000/month, billed annually	Data Collection and CDP by quote	Enterprise CDP and tag management
Hightouch	Free Reverse ETL tier	Usage; no MTU limits	Composable CDP and reverse ETL with existing warehouse models
Amperity	Sales contact	Amps	Enterprise CDP with a retail focus
Treasure AI	No public price	Complete CDP and Composable CDP products	Teams evaluating Complete CDP and Composable CDP modes
Rokt mParticle	No public price	No public pricing terms	Hybrid real-time and warehouse architecture

How I Reviewed Zeotap

I walked the packaged CDP from Create Source through activation, including Catalogue mapping, ID strategy configuration, and an Ada audience. I focused on customer matching and permission enforcement before activation to Link to Destination.

I evaluated Composable and the Snowflake Native App’s account boundary and billing model. My comparison covered Calculated Attribute freshness and limits, consent filtering, and country-based residency routing.

I also assessed the Q4 2025 Enterprise restrictions and Q1 2026 journey controls against their practical buying implications. For value, I compared Zeotap’s contract routes with RudderStack’s published event allowances, Tealium’s Activation price, and Hightouch’s free sync allowance.

My buying criteria prioritized consent enforcement and residency over destination count, with batch freshness a separate exclusion for same-day computed-trait requirements.

Prices current as of October 2026.

Should You Use Zeotap as Your Customer Data Platform?

I recommend Zeotap for EU-heavy, consent-bound enterprises that need identity resolution and activation permissions in the same CDP. For existing Snowflake customers, the Native App adds a warehouse-resident option without requiring the packaged architecture.

My strongest reasons to shortlist it are Consent Orchestration and the choice between Packaged and Composable. Ada makes audience conditions editable, while Journey Traces exposes an individual user’s path with input and destination payloads.

I advise skipping Zeotap if transparent pricing or a free tier is essential. I also reject batch Calculated Attributes for sub-day computed-trait requirements: their usual 24 to 36 hour freshness conflicts with that expectation, and real-time attributes have only three operators.

Book a demo with Zeotap and request these requirements in the proposed contract:

• Calculated Attribute allowances: Include counts and required modes.
• Destination coverage: Name the activation endpoints.
• Residency behavior: Specify country routing or the Snowflake region.
• Enterprise restrictions: Include the Catalogue controls you need.
• A/B testing access: Request opt-in access explicitly.

FAQs

How much does Zeotap cost?

Zeotap has no public price list; pricing requires a demo and contract. Calculated Attribute allowances and other limits depend on the package arranged through a CSM.

The Snowflake Native App uses Snowflake credits and Marketplace billing, while Google Cloud Marketplace puts Zeotap fees on the Google Cloud invoice.

Does Zeotap have a free trial?

Zeotap has no free trial or free plan. Buying starts with Book a Demo, and the Snowflake Native App asks buyers to request a working session.

Request package limits during that evaluation, especially Calculated Attribute counts and Enterprise-only Catalogue restrictions. A/B testing requires opt-in access.

Where does Zeotap store data?

Packaged Zeotap stores data in EU, US, or IN regions, resolved from a three-letter country code with IP fallback. The Snowflake Native App follows the customer’s Snowflake account region.

The EU storage label also covers Africa and Russia, while IN covers Asia, the Middle East, and Oceania.

Is Zeotap a composable CDP?

Zeotap offers a Composable CDP alongside its Packaged product through InfraFlex. Composable supports six warehouses and performs identity resolution as warehouse SQL.

I favor the Snowflake Native App when the CDP must run inside an existing Snowflake account; it uses Snowpark Container Services, Snowflake credits, and Marketplace billing.

What happened to Zeotap Data?

Zeotap sold its data business, Zeotap-Data, to Roqad, with the transaction announced September 2, 2025. Zeotap retained the CDP and exclusive identity graph access for five years.

Integrations with around 30 ad tech partners moved to Roqad. The customer data platform remained with Zeotap, including its Packaged and Composable deployment choices.