# Eppo — Conversion Optimization

Eppo by Datadog, Inc. holds a panel rating of 2/5 from 6 AI judges (category median 2/5), as of 2026-11-17. Provenance: Rest of world; sovereignty: not determined.

- Vendor: Datadog, Inc. (https://www.geteppo.com)
- Category: [Conversion Optimization](https://whats-best.ai/software/conversion-optimization.md)
- Provenance (editorial classification): Rest of world
- Canonical page: https://whats-best.ai/software/conversion-optimization/eppo
- Alternatives: https://whats-best.ai/software/conversion-optimization/eppo/alternatives ([markdown](https://whats-best.ai/software/conversion-optimization/eppo/alternatives.md))
- Data as of: 2026-11-21 · evaluated 2026-11-17 · oldest source captured 2026-11-16

**AI-generated.** Every score, rationale and verdict here is written by a panel of AI judges from public evidence — an analysis, not a hands-on test. Every fact quotes the vendor page it was read from. No score can be bought. How scores are made: https://whats-best.ai/methodology

## At a glance

- Vendor: Datadog, Inc.
- Headquarters: Not stated on the vendor’s pages as of 16 Nov 2026
- Data hosting: Not stated on the vendor’s pages as of 16 Nov 2026
- Free plan: Not stated on the vendor’s pages as of 16 Nov 2026

## The panel's verdict

Eppo — acquired by Datadog and now branded Datadog Experiments — is a warehouse-native experimentation and feature-flagging platform. The bench scored it strongest on data and integrations, where scores cluster at 7-8, and on experimentation scope, also 7-8: experiments run on the customer's own warehouse — Snowflake, Databricks, BigQuery and Redshift are named — with no data duplication, and SDKs are documented for twelve languages. Statistical rigour clusters at 6-7 on named methods — sequential, fixed-sample and Bayesian frameworks with CUPED variance reduction — though we found no public information on sample-ratio-mismatch detection or correction for multiple metrics. The weakest areas are consent and tracking, scored 0-2, with no public information on how the script behaves relative to consent; sovereignty, scored 0-3, where the published terms point to Eppo Data, Inc. of San Francisco under Datadog's master agreement; and performance impact, clustering at 2. The genuine split is fit: one judge scored experimentation scope 8 for the server-side SDK programme while the e-commerce judge returned 0, and several judges found no public information on a visual editor or multi-page tests. We found no published prices.

**Speaks for it:**
- Experiments run natively on your own warehouse — Snowflake, Databricks, BigQuery and Redshift are named — with no data duplication
- Statistical methods are named outright: sequential, fixed-sample and Bayesian frameworks, CUPED variance reduction, and non-inferiority testing with guardrail cutoffs
- SDKs are documented for twelve languages, from JavaScript to Android, spanning client and server
- Flag configurations are published to a CDN with local caching and no real-time reliance on the vendor's API
- The vendor states that PII never leaves your system, with results described as leaving a paper trail easy to audit and reconcile

**Held against it:**
- We found no public information on when the assignment script runs relative to consent, what it stores on a visitor's device, or whether a cookieless mode exists
- We found no published figures on script weight, flicker handling or impact on Core Web Vitals
- Contracting points to Eppo Data, Inc. of San Francisco under Datadog's Master Subscription Agreement, and only subprocessor categories — without locations — are published
- We found no public information on a visual editor or on split-URL and multi-page tests, so building a test takes a developer
- We found no public information on a documented API for managing experiments and reading results, on webhooks, or on export formats and limits

**Best for:**
- You run server-side or app experiments and want results computed in your own warehouse — Snowflake, Databricks, BigQuery or Redshift — where they stay in your data and can be audited
- Your engineers ship in the twelve documented SDK languages and want feature gates, controlled rollouts, mutual-exclusion layers and global holdouts on one metric model
- You need named statistics — sequential, fixed-sample and Bayesian with CUPED variance reduction — that you can defend to finance

**Avoid if:**
- Your marketing team needs point-and-click page testing without developers; the captured pages describe an SDK-and-flag programme, and we found no public information on a visual editor
- Your procurement requires an EU contracting entity, published EU residency or a named subprocessor list; the published terms point to Eppo Data, Inc. of San Francisco under Datadog's master agreement, with subprocessor categories only
- You must satisfy a data protection officer on how the test script treats consent before deploying it on your site; consent and tracking scored 0-2, with no public answer on script behaviour, storage or a cookieless mode

## Scores by judge and criterion

Scores out of 5, each judge scoring independently against the published rubric (v2). "n/r" = not rated: the vendor publishes no price, or the judge found no public information on that criterion.

| Criterion | The Growth Lead | The E-Commerce Manager | The Product Engineer | The CRO Consultant | The Data Protection Officer | The Skeptic |
| --- | --- | --- | --- | --- | --- | --- |
| Experiment types & delivery | 3.5/5 | 0/5 | 4/5 | 3.5/5 | 3.5/5 | 3.5/5 |
| Statistical method & guardrails | 3.5/5 | 0/5 | 3.5/5 | 3.5/5 | 3/5 | 3.5/5 |
| Consent & visitor tracking | 1/5 | n/r | 1/5 | 1/5 | 1/5 | n/r |
| Snippet performance & flicker | 1/5 | 0/5 | 1/5 | 1/5 | 1/5 | 1/5 |
| Analytics, data export & integrations | 4/5 | 0/5 | 3.5/5 | 3.5/5 | 3.5/5 | 4/5 |
| European sovereignty | 1.5/5 | 0/5 | 1/5 | 1/5 | 0.5/5 | 1.5/5 |
| Pricing transparency | n/r | n/r | n/r | n/r | n/r | n/r |
| **Weighted verdict** | **3/5** | **0/5** | **2.5/5** | **2.5/5** | **1.5/5** | **3/5** |

### The judges

- **The Growth Lead** (v1.0): Runs the testing programme for a German marketing team and reports uplift to the board. Wants a visual editor the team can use without a developer, targeting that goes beyond the URL, and results she can defend when finance asks where the promised revenue went.
- **The E-Commerce Manager** (v1.0): Owns a shop where a tenth of a second of load time shows up in the conversion rate. Wants personalisation that can be tested against a control, a snippet that does not flicker the product page, and a traffic-based price that does not jump a tier every December.
- **The Product Engineer** (v1.0): Ships experiments behind feature flags and would rather not ship a third-party script at all. Wants server-side SDKs for the stack in use, edge delivery, metrics computed in the warehouse the team already trusts, and an API good enough to build internal tooling on.
- **The CRO Consultant** (v1.0): Sets up testing programmes for several clients and has to recommend a platform each can afford and defend. Reads for multi-project management, clean analytics integration, a consent setup a client's DPO will sign off, and a traffic metric she can estimate before the contract.
- **The Data Protection Officer** (v1.0): Has to answer whether the test script may run before the consent banner is answered, and usually finds the vendor has not said. Wants the storage keys listed, a documented consent-pending mode, visitor data processed in the EU, and a published subprocessor list that includes the CDN.
- **The Skeptic** (v1.0): A statistician who has watched too many "winners" regress to zero. Reads for the method by name, what stops a team from peeking, sample ratio mismatch checks and multiple-comparison corrections — and treats a "probability to beat" without documented assumptions as marketing.

## European sovereignty — proven facts

Built only from facts on the vendor's own pages; a dimension we could not prove is left open, never scored as zero. Headline: not determined.

| Dimension | Finding | Points | Source |
|---|---|---|---|
| Legal entity | Not determined | — | — |
| Hosting location | Not determined | — | — |
| Hosting options | Not determined | — | — |
| Open source | Not determined | — | — |
| Data processing agreement | Not determined | — | — |
| Transfer impact assessment | Not determined | — | — |

## Key facts

Each fact is the vendor's own statement, cited to the source page it was read from ([S*n*] → Sources).

**Compliance**
- California rights: yes [S2]
- CCPA disclosed categories: Category B - California Customer Records law personal information [S2]
- Cookies notice: yes [S2]
- Dnt response: no [S2]
- GDPR: yes [S7]
- Minors: 18 [S2]
- No sale of data: no · yes [S2]
- Opt out timeframe: 15 [S2]
- SOC 2 type2: https://trust.datadoghq.com · SOC 2, Type 2 [S3]
- Third party contracts: yes [S2]
- Tracking technologies: Beacons/Pixels/Tags [S2]

**Data**
- Account deletion: some information may be retained for fraud, investigations, legal requirements · yes [S2]
- Auditability: yes [S7]
- Automatically collected: ip address · browser and device characteristics · log and usage data · device data · location data [S2]
- Egress: no [S7]
- No egress: No data egressing of data, no conflicting sources, no black boxes. [S1]
- Opt out targeted ads: http://www.aboutads.info/choices/ [S2]
- Personal information collected: names; email addresses; job titles; and other similar information [S2]
- Pii: no [S4]
- Retention: as long as necessary; no longer than account lifetime [S2]

**Hosting**
- Login URL: https://eppo.cloud/ [S3]
- Residency: customer's own cloud [S7]
- Warehouse native architecture: Warehouse-native architecture [S1]

**Integrations**
- Deployment flexibility: first-class support for all common deployment modes and development workflows [S4]
- Overview: Integrations with the tools you love [S1]
- Warehouse native: Warehouse Native [S3]
- Whitepaper cta: Lead-capture form to get the technical white paper [S8]

**Legal**
- Acquired by datadog: Site footer indicates Eppo is owned by Datadog [S8]
- Acquisition: Datadog [S1]
- Address: 435 Brannan St, San Francisco, CA 94107, USA [S2]
- Billing contact: billing@datadoghq.com · Datadog [S3]
- Copyright: 2026 · Datadog [S2]
- Copyright: 2026 · Datadog [S3]
- Entity: Eppo Data, Inc. [S2]
- Governing contract: Master Subscription Agreement at https://www.datadoghq.com/legal/msa/ [S3]
- Jurisdiction: California · United States [S2]
- Owner: Datadog [S7]
- Privacy notice updated: 2024-10-16 [S2]
- Terms changes: first day of calendar month after posting; immediate at next renewal term if Order is 12+ months [S3]
- Terms last updated: 2025-05-05 [S3]

**Pricing**
- Renewal: no [S3]

**Product**
- AI model evaluation module: Build more effective AI products. Evaluate AI models by running experiments with trusted business metrics. [S1]
- Architecture: dumb server, smart client; no real-time reliance on the Eppo API; configs published to CDN [S4]
- Business metrics: Run experiments on your company’s core business metrics, not on vanity metrics. [S1]
- Capabilities: Experimentation · Artificial Intelligence · Warehouse Native · Feature Flagging · Outperform [S3]
- Clustered experiments: clustered experiments and last value aggregation [S6]
- Contextual bandit: Build dynamic, personalized experiences to optimize outcomes [S4]
- Control of infrastructure: Control of tables, queries, and experiment update schedule [S7]
- Controlled rollouts: Reduce risk by slowly exposing traffic to new features [S4]
- Cost reduction: 90 [S7]
- Cuped: yes [S7]
- Cuped runtime: Shave weeks off of experiment runtime with advanced tools like CUPED++ [S1]
- Data automation: Automate experimentation analysis, deep dives, and diagnostics with trusted statistical methods [S1]
- Debugging: one interface for requesting assignment for debugging [S4]
- Docs URL: https://docs.geteppo.com/ [S3]
- Experiment forecasts: Plan roadmaps that drive impact and measure their performance with Experiment Forecasts [S1]
- Experimentation module: Warehouse-native architecture powers a world class statistical engine without becoming a black box. Automate experiment analysis while maintaining rigor. [S1]
- Experiments: Randomize users between multiple test variants to measure impact [S4]
- Experiments branding: Eppo is now Datadog Experiments [S1]
- Feature flagging: yes [S7]
- Feature flagging: Feature Flagging [S8]
- Feature flagging module: Fast and resilient feature flags for A/B tests, feature gates, controlled rollouts, kill switches, and dynamic no-code configuration. Simple to get started, and built to scale. [S1]
- Feature flags scale: Leverage the fastest, most resilient feature flags - powering billions of daily assignments [S1]
- Feature gates: Direct users to different experiences or quickly remove problematic code [S4]
- Global holdouts: Automatically exclude small amount of customers to measure business impact [S4]
- Incremental pipelines: incremental · daily [S7]
- Marketing email opt out: yes [S2]
- Metric governance: Centralize and govern metric definitions, experiment Protocols, and workflows across your org [S1]
- Metric management: Enterprise-scale metric management with metric aggregations and metric properties [S7]
- Metric type percentiles: New metric type: Percentiles [S5]
- Mutual exclusion: Easily isolate experiments to avoid unwanted interaction effects [S4]
- No data duplication: no [S7]
- Non inferiority testing: Non-inferiority testing with Guardrail Cutoffs [S6]
- Outperform community: Outperform · 4,000+ [S1]
- Performance scorecard: Performance Scorecard for experimentation programs [S6]
- Personalization module: Unlock new possibilities in personalization with Contextual Bandits. Automatically optimize user experiences in real-time, or get more value from your AI models. [S1]
- Positioning: trustworthy, self-service experimentation and feature flagging for every team [S1]
- Protocols property analysis: Accelerate Experimentation with Protocols and Property Analysis [S5]
- Redundancy: redundancy across Eppo application, CDN, and local cache [S4]
- Safe rollouts: Make every deployment into an automated safe rollout [S1]
- SDK architecture: Eppo SDKs reflect architectural principles used at Airbnb, Netflix, and Spotify [S8]
- SDK languages: JavaScript · Node · Python · Java · .NET · PHP · Go · Rust · Ruby · React Native · iOS · Android [S4]
- SDK tags dependent flags: yes [S5]
- Self service analysis: Dig into feature performance and slice-dice analysis without needing your data team [S1]
- Social login: Facebook · Twitter · yes [S2]
- Statistical frameworks: sequential · fixed sample · Bayesian · CUPED [S1]
- Statistics: Modern statistical methodologies to minimize time-to-insight [S8]
- Use range: Use one platform for everything from simple feature gates to AI personalization [S1]
- Version control metrics: Data teams own metric definitions and guardrails with version control and the semantic layer. [S1]
- Warehouse costs: Eppo’s pipelines are optimized to keep your data warehouse bills down. [S1]
- Warehouse integrations: Snowflake · Databricks · BigQuery · Redshift [S7]
- Warehouse native: Ensure trust in data with zero-copy, warehouse-native architecture [S1]
- Warehouse native: Warehouse-native experimentation tool [S7]
- Warehouse native principle: Experimentation workflows run natively on top of the data warehouse [S8]
- Warehouse native whitepaper: Eppo offers a technical white paper on the warehouse-native experimentation workflow [S8]

**Subprocessors**
- Categories: Data Analytics Services · Cloud Computing Services · User Account Registration & Authentication Services · Communication & Collaboration Tools · Data Storage Service Providers · Performance Monitoring Tools · Product Engineering & Design T… [S2]

**Support**
- Included: regular business hours · slack · other standard channels · yes · no [S3]
- White glove: Eppo's white-glove support gives you access to world-class experimenters and the building blocks to nurture an experimentation culture in your org. [S1]
- White paper: yes [S7]

## Questions

### Is Eppo European?

No. whats-best.ai classifies Eppo as Rest of world.

### Where does Eppo store data?

Not stated on the vendor’s pages as of 16 Nov 2026.

### Does Eppo have a free plan?

Not stated on the vendor’s pages as of 16 Nov 2026.

### What are the best alternatives to Eppo?

As of 19 Nov 2026, the panel rates [Convert Experiences (3.5/5)](https://whats-best.ai/software/conversion-optimization/convert-experiences), [Varify.io (3/5)](https://whats-best.ai/software/conversion-optimization/varify) and [Webtrends Optimize (3/5)](https://whats-best.ai/software/conversion-optimization/webtrends-optimize) highest among the other 45 Conversion Optimization products evaluated. Ratings that look equal are rounded to half stars; the order uses the unrounded rating, then the name. The highest-rated European alternative is [Varify.io](https://whats-best.ai/software/conversion-optimization/varify) (EU-Made, Germany). Ratings are the AI panel’s opinion, formed from public evidence: [how scores are made](https://whats-best.ai/methodology#ordering). The full list, with European products first: [Eppo alternatives](https://whats-best.ai/software/conversion-optimization/eppo/alternatives).

## What this page leaves out

- 2 product facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 1 compliance fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material.
- 1 legal fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material.

## Sources

- [S1] Vendor homepage: https://www.geteppo.com/ (checked 2026-11-16)
- [S2] Privacy policy: https://www.geteppo.com/privacy-notice (checked 2026-11-16)
- [S3] Terms of service: https://www.geteppo.com/terms-and-conditions (checked 2026-11-16)
- [S4] Experiment types & delivery — found from sitemap: https://www.geteppo.com/feature-flagging (checked 2026-11-16)
- [S5] Experiment types & delivery — found from sitemap: https://www.geteppo.com/product-updates/feature-flag-updates-sdk-tags-and-dependent-flags (checked 2026-11-16)
- [S6] Snippet performance & flicker — found from sitemap: https://www.geteppo.com/product-updates/introducing-the-performance-scorecard-elevate-your-experimentation-program (checked 2026-11-16)
- [S7] Analytics, data export & integrations — found from sitemap: https://www.geteppo.com/warehouse (checked 2026-11-16)
- [S8] Analytics, data export & integrations — found from sitemap: https://www.geteppo.com/warehouse-native-workflow-white-paper (checked 2026-11-16)

---

Cite as: whats-best.ai, "Eppo — Conversion Optimization", AI panel analysis, as of 2026-11-21. https://whats-best.ai/software/conversion-optimization/eppo

Corrections: https://whats-best.ai/correct/411?target=page
