Conversion Optimization
Eppo
Rest of world Report an errorPanel rating · 6 judges · How to read the stars
Category median
Sovereignty: 1 of 4 dimensions proven
0–5 in half steps. 5 means the rubric's top anchor is met on the evidence.
by Datadog, Inc. · www.geteppo.com
Report an error on this page Is this your product? →
Read this page as one judge. Each weighs the same scores by what they care about.
The panel's verdict
Eppo — warehouse-native experimentation and feature flagging, now Datadog Experiments — earns its highest marks in data and integrations and experimentation scope, both scoring 7-8: results are computed in the customer's own cloud on Snowflake, Databricks, BigQuery or Redshift, and feature-flag experiments, mutual exclusion layers, global holdouts and contextual bandits arrive through SDKs in twelve languages. Statistical rigour sits at 6-7 on named frameworks — sequential, fixed sample and Bayesian — with CUPED variance reduction and non-inferiority testing under guardrail cutoffs. Weakest are sovereignty, scored 1-2 on contracting through Eppo Data, Inc. of San Francisco under Datadog's master subscription agreement, with no stated residency for the hosted service and subprocessors published only as categories, and consent and tracking, scored 2-3, with no public information on what the SDK stores on a visitor's device or when assignment runs relative to consent. Performance impact runs 1 to 4: the data protection officer credits CDN delivery with local caching as loading behaviour, while the remaining scores of 1 to 2 cite no published page-cost figures. No prices are published.
Speaks for it
- Results are computed in the customer's own cloud on Snowflake, Databricks, BigQuery or Redshift, from a single source of truth with no data duplication
- Feature-flag experiments, mutual exclusion layers, global holdouts and contextual-bandit personalisation are delivered through SDKs in twelve named languages
- Sequential, fixed sample and Bayesian frameworks are named in writing, with CUPED and CUPED++ variance reduction and non-inferiority testing under guardrail cutoffs
- Metric definitions are owned by the data team under version control and a semantic layer, with results described as easy to audit and reconcile
- The terms state that orders will not automatically renew and that support is included at no additional charge
Held against it
- We found no public information on what the experimentation SDK stores on a visitor's device or when variant assignment runs relative to consent
- Contracting runs through Eppo Data, Inc. at 435 Brannan St, San Francisco under Datadog's master subscription agreement, with no stated residency for the hosted service and subprocessors listed only as categories
- We found no published figures for script size, load cost, flicker handling or measured Core Web Vitals impact
- No public prices appear on any captured page, so an annual cost cannot be estimated before a sales conversation
- We found no public information on sample ratio mismatch detection or on correction for multiple metrics and variants
Best for
- You already run your metrics on Snowflake, Databricks, BigQuery or Redshift and want experiments computed in your own warehouse, where you can audit the results yourself
- Your programme is developer-led and spans web, server and mobile, matching SDKs across twelve named languages with mutual exclusion layers and global holdouts
- You defend experiment numbers to finance and want named frameworks — sequential, fixed sample and Bayesian with CUPED variance reduction — behind them
Avoid if
- Your DPO must sign off on documented consent behaviour for a script that touches every visitor to your site
- You require an EU contracting entity, stated data residency for the hosted service, or named subprocessors rather than categories
- You need to estimate your annual cost before entering a sales conversation
The scores
Experiment types & delivery
Show reasoningHide reasoning
How this is scored
What can be tested and where: client-side changes through an editor, server-side and feature experiments through SDKs, multivariate and multi-page tests, and personalisation — judged on what the documentation shows rather than on the feature grid.
0 — Simple A/B split of one page element through a visual editor; no server-side option, no targeting beyond URL.
3 — Client-side A/B and split-URL tests with basic audience targeting, and no SDK or server-side delivery.
5 — Client-side and server-side experiments through documented SDKs for common languages, multivariate and multi-page tests, audience targeting on behaviour and attributes, and rule-based personalisation.
8 — Feature-flag-based experiments sharing one audience and metric model with web tests, mutually exclusive experiment groups, holdouts, edge or CDN delivery, and personalisation that can itself be tested against a control.
10 — One experimentation programme across every surface: web, app, server and edge from the same platform, experiment interactions managed, a documented experiment lifecycle from hypothesis to archived result, and a library of past results a team can search.
The Growth Lead
Feature-flag experiments across twelve documented SDK languages with mutual exclusion layers, global holdouts and contextual-bandit personalisation, with configs published to a CDN — a genuine one-platform programme across web, server and app. For my marketing team the decisive gap is that we found no public information on a visual editor or split-URL testing, so every test needs a developer. 1 4
The E-Commerce Manager
The feature-flagging pages show experiments, gates, kill switches, controlled rollouts, mutual-exclusion layers and global holdouts, delivered through SDKs for a dozen languages with configs published to a CDN — most of the way to the top of this category. What keeps me from rating it higher: I found no public information on a visual editor, multivariate or split-URL tests, or any documented way to test the Contextual Bandit personalisation itself against a control, which is exactly the proof I want before letting an optimiser loose on my product pages. 1 4
The Product Engineer
Feature-flag-based experiments, controlled rollouts and kill switches share one platform with A/B tests, mutual exclusion layers and global holdouts, served through SDKs spanning JavaScript, Node, Python, Java,.NET, PHP, Go, Rust and mobile, with configs published to a CDN instead of a real-time API call — that is the shape I want to build on. Contextual bandits cover personalisation. I found no public information showing the personalisation can itself be tested against a control, nor any documented audience-targeting model. 1 4
The CRO Consultant
Feature-flag experiments delivered through SDKs across twelve named languages, with mutual exclusion layers, global holdouts and controlled rollouts, would cover most of a multi-client testing programme, and configs published to a CDN with local caching keep delivery resilient. We found no public information on a visual editor for client-side changes, on multivariate or multi-page tests, or on a documented experiment lifecycle with a searchable library of past results, so I stopped short of the top band. 4 1 8
The Data Protection Officer
Server-side experiments and feature flags share one platform with SDKs across twelve languages, mutually exclusive experiment layers, global holdouts, and contextual-bandit personalisation, with assignment configs published to a CDN. I found no public information on a documented experiment lifecycle from hypothesis through to archived results, or on a searchable library of past results, which is what the published pages would need to show for a higher mark. 4 1 6
The Skeptic
SDK-delivered experiments in twelve named languages, mutual exclusion layers, global holdouts and contextual bandits all sharing one warehouse metric model — the feature-flag-plus-holdouts shape I look for. What I did not find is a documented lifecycle from hypothesis to archived result or a searchable library of past results, so I stop one rung short of the top. 1 4
Statistical method & guardrails
Show reasoningHide reasoning
How this is scored
Which statistics decide the winner and what protects the customer from misreading them. Scored on what the vendor documents: the method by name, how peeking and multiple comparisons are handled, and whether sample ratio mismatch is detected.
0 — A "winner" or "probability to beat" figure with no documented method, no stated sample-size guidance and no warning against stopping early.
3 — The method is named (frequentist or Bayesian) but its assumptions are not documented, and nothing prevents a test from being called while it is still underpowered.
5 — Documented method with confidence or credible intervals, a sample-size or test-duration calculator, and stated guidance on when a result may be read.
8 — Sequential testing or an equivalent documented protection against peeking, correction for multiple metrics or variants, sample ratio mismatch detection, variance reduction such as CUPED, and guardrail metrics that can stop a harmful test.
10 — The statistics are auditable: methodology published in enough detail to reproduce a result, the choice of method explained per use case, raw per-visitor data available for independent re-analysis, and the interface refuses to present an underpowered result as a conclusion.
The Growth Lead
Sequential, fixed sample and Bayesian frameworks are named and implemented alongside CUPED variance reduction and non-inferiority testing with guardrail cutoffs — the stack I would want behind a number I defend to finance. We found no public information on sample ratio mismatch detection or correction for multiple metrics or variants, which is exactly what a sceptical board member asks about. 1 6 7
The E-Commerce Manager
Sequential, fixed-sample and Bayesian frameworks are all named, alongside CUPED++ variance reduction and non-inferiority testing with guardrail cutoffs — more named statistics than most vendors publish, and sequential testing covers the peeking problem. But I found no public information on sample ratio mismatch detection or correction for multiple metrics and variants, and the only planning tool is 'Experiment Forecasts' without sample-size guidance, so two of the guardrails I look for aren't evidenced. 1 6 7
The Product Engineer
Sequential, fixed-sample and Bayesian frameworks are all named, CUPED variance reduction is documented as shortening runtime, and non-inferiority testing with guardrail cutoffs exists — real machinery, not a probability-to-beat badge. But I found no public information on sample ratio mismatch detection, on correction for multiple metrics or variants, or on sample-size and duration guidance beyond an unexplained Experiment Forecasts mention. 1 6 7
The CRO Consultant
Sequential, fixed-sample and Bayesian frameworks are named alongside CUPED variance reduction, and the product updates add non-inferiority testing with guardrail cutoffs — the peeking protection plus guardrail pairing I want before a client reads a result. We found no public information on sample ratio mismatch detection, correction for multiple metrics or variants, confidence or credible intervals, or a sample-size and duration calculator, and the deeper methodology sits behind a lead-capture form for the white paper. 1 6 8
The Data Protection Officer
Sequential, fixed-sample and Bayesian frameworks are named in writing, CUPED variance reduction is documented, and non-inferiority testing with guardrail cutoffs is described alongside clustered experiments. I found no public information on sample ratio mismatch detection, on correction for multiple metrics or variants, or on a sample-size or duration calculator, so the guardrail set as published stops short of the full picture. 1 6 7
The Skeptic
The frameworks are named by name — sequential, fixed sample and Bayesian — with CUPED and CUPED++ variance reduction and non-inferiority testing under guardrail cutoffs, which is more method disclosure than most vendors show. But I found no public information on sample ratio mismatch detection, any correction for multiple metrics or variants, a sample-size or duration calculator, or the assumptions behind each framework, so a reader cannot verify what stops a team from peeking beyond the word "sequential". 1 6 7
Consent & visitor tracking
Show reasoningHide reasoning
How this is scored
Whether the test script respects §25 TDDDG and the ePrivacy rules as evidenced on the vendor's own pages: when the script runs relative to consent, what it stores on the device, whether a cookieless or consent-free mode exists, and how visitor identifiers are handled.
0 — Nothing on the vendor's pages says what the script stores on the device or whether it runs before consent; consent is described as the customer's problem.
3 — Cookies or local storage are listed, and a consent integration is mentioned, but the documentation does not say what the script does before consent is given.
5 — Documented behaviour before and after consent, integration with common consent management platforms, a published list of cookies and storage keys with their lifetime, and IP anonymisation described.
8 — A documented consent-pending mode that holds tracking until consent while variants can still be served, a cookieless or first-party-only option, retention of visitor data stated, and guidance on the legal basis for testing in the EU.
10 — Built for §25 TDDDG: the default configuration stores nothing on the device without consent, a server-side or cookieless mode documented end to end, per-visitor data deletable on request, and the vendor states in writing how each mode maps to consent requirements.
The Growth Lead
The captured privacy notice governs Eppo's own website cookies and says nothing about what the test SDK stores on a visitor's device or when it runs relative to consent, and the "Easy compliance with GDPR" claim gives no documented consent behaviour. We found no public information on a consent-pending mode, a cookieless option or consent management platform integration, which I cannot sign off for a German programme. 2 4 7
The E-Commerce Manager
The privacy notice lists cookies, beacons and pixels and points to a cookie notice, and the warehouse pages claim 'Easy compliance with GDPR' with PII never leaving my system — but that is as far as it goes. I found no public information on what the SDK stores on a visitor's device, when variant assignment runs relative to consent, a cookieless mode, or integration with consent management platforms, which matters when the script touches every visitor to my shop. 2 7
The Product Engineer
The only cookie and tracking material I found is the vendor's own privacy notice for its own website — beacons, pixels and tags, a referenced cookie notice, and a statement that DNT signals are not answered — plus a bare "Easy compliance with GDPR" claim. I found no public information on what a deployed assignment integration stores on a visitor's device, whether anything runs before consent, or how visitor identifiers are handled. 2 7
The CRO Consultant
The privacy notice lists cookies, web beacons and pixels and points to a cookie notice, and the warehouse page claims easy GDPR compliance, but we found no public information on what the experimentation script stores on a visitor's device or what it does before consent is given. The statements that personal data never leaves the customer's system address identifier handling rather than consent, and no consent management integration or cookieless mode is documented on the captured pages. 2 7 4
The Data Protection Officer
The privacy notice says cookies, web beacons and pixels may be used and points to a cookie notice, and a warehouse page claims easy GDPR compliance. I found no public information on whether the assignment script may run before the consent banner is answered, no published list of the storage keys and their lifetimes, no consent-pending mode, and no retention statement or legal-basis guidance for visitor data in the EU. 2 7
The Skeptic
The privacy notice covers beacons, pixels and a cookie notice for the vendor's own website and a 15-day opt-out window, but I found no public information on what the experimentation SDK stores on a visitor's device, when assignment runs relative to consent, or any consent-pending or cookieless mode. "Easy compliance with GDPR" is a slogan, not a specification. 2 7
Snippet performance & flicker
panel disagrees
Show reasoningHide reasoning
How this is scored
The cost the client-side snippet imposes on the page it tests: blocking load, flicker of original content, script weight and the effect on Core Web Vitals — scored on what the vendor measures and publishes, not on "lightning fast".
0 — A synchronous snippet with no stated size, no flicker handling and no mention of performance.
3 — An anti-flicker snippet that hides the page until the test loads, with a timeout, and no published figures for script size or load cost.
5 — Script size and loading behaviour documented, asynchronous loading option, flicker handling explained with its trade-off, and CDN delivery of the snippet.
8 — Published performance figures including impact on Core Web Vitals, a self-hosting or first-party-domain option for the script, per-project bundles containing only active experiments, and a server-side or edge alternative for flicker-sensitive tests.
10 — Performance is a stated commitment: measured overhead published and maintained, flicker eliminated by edge or server-side rendering as a documented path, and tooling that shows the customer what their own configuration costs the page.
The Growth Lead
The architecture described — configs published to a CDN with local-cache redundancy — reads as resilience against outages rather than any statement about page cost. We found no public information on script size, load cost, flicker handling, asynchronous loading or any measured effect on Core Web Vitals. 4 1
The E-Commerce Manager
On a tenth-of-a-second conversion question the captured pages are nearly silent: no script sizes, no load-cost figures, nothing on Core Web Vitals, and no anti-flicker handling documented anywhere I can see. The only crumbs are configs published to a CDN with local-cache redundancy and a 'dumb server, smart client' design — resilience, not a measured page cost — and the one page that reads as performance is a Performance Scorecard for the experimentation programme, not for the snippet. 4 6
The Product Engineer
The delivery architecture is documented — configs published to a CDN with redundancy across application, CDN and local cache, and no real-time reliance on the vendor's API — but that is resilience, not measurement. No published figures on script weight, load cost, flicker handling or Core Web Vitals appear on any captured page, and "fastest, most resilient" is a claim, not a number. 1 4
The CRO Consultant
We found no public information on script size, loading behaviour, flicker handling, self-hosting or any measured effect on Core Web Vitals. The nearest facts are that feature-flag configs are published to a CDN with multi-level redundancy and local caching, which describe reliability rather than the cost a snippet imposes on the page. 4
The Data Protection Officer
Assignment configs are published to a CDN with local caching and no real-time reliance on the vendor's API, which does describe loading behaviour. I found no published figures for script size or load cost, no mention of Core Web Vitals impact, and no public information on flicker handling or a self-hosting option for the script. 4 1
The Skeptic
The captured pages describe config delivery through a CDN with local-cache redundancy, and an architecture essay on SDK principles, but no measured page cost anywhere. I found no published figures for script weight, load time or Core Web Vitals impact and no documented flicker path — for a product delivered by SDKs, that quiet about the client-side cost is information in itself. 4 8
Analytics, data export & integrations
Show reasoningHide reasoning
How this is scored
Getting results and raw data out: integration with analytics and tag management, export of visitor-level results, warehouse-native analysis, and an API — because an experiment result that cannot be checked in the customer's own data is a claim, not a finding.
0 — Results visible in the vendor's dashboard only; no export, no analytics integration, no API.
3 — CSV export of aggregated results and one analytics integration, with no visitor-level data and no documented API.
5 — Integrations with common analytics and tag managers, export of results, and a documented API for managing experiments and reading results.
8 — Visitor-level raw data export or streaming to a data warehouse, warehouse-native analysis on the customer's own metrics, CDP integration for audiences, and an API with stated limits.
10 — The platform treats the customer's warehouse as the source of truth: metrics defined once and computed there, full historical experiment data exportable in open formats, and a versioned API a team can build its own programme tooling on.
The Growth Lead
Results are computed natively in the customer's own warehouse — Snowflake, Databricks, BigQuery or Redshift — on metrics the data team owns with version control and a semantic layer, with no data duplication and a stated audit trail I could show finance. We found no public information on a documented API with stated limits or on CDP integration for audience building. 7 1 4
The E-Commerce Manager
This is where the product earns its keep: experiments run natively on my own warehouse (Snowflake, Databricks, BigQuery, Redshift), metric definitions are owned by my data team under version control in a semantic layer, and results leave 'a clear paper trail, making results easy to audit and reconcile' — the winner is computed in my own data, which is exactly how I would verify a lift claim. The gap is delivery out: I found no public information on a documented API with stated limits, CDP audience integration, or export formats on these pages. 1 7 8
The Product Engineer
This is the model I would actually buy: metrics computed entirely in the customer's own cloud on Snowflake, Databricks, BigQuery or Redshift, single source of truth with no data duplication, daily incremental pipelines, and metric definitions owned by the data team with version control and a semantic layer. The gap is the API — I found no public information on a documented API, its limits or versioning, which is what I would need to build internal programme tooling on top. 1 7
The CRO Consultant
This is the platform's strength for my clients: experiments are computed in the customer's own cloud on Snowflake, Databricks, BigQuery or Redshift, from a single source of truth with no data duplication, and metric definitions are governed with version control and a semantic layer — so visitor-level raw data stays in a warehouse I can audit myself. We found no public information on a documented API with stated limits, on analytics or tag-management integrations named beyond a generic line about tools you love, or on export formats. 7 1 4
The Data Protection Officer
The warehouse-native model is documented in the vendor's own words: metrics computed completely in the customer's own cloud, a single source of truth with no data duplication, integrations with Snowflake, Databricks, BigQuery and Redshift, and results described as easy to audit and reconcile. I found no public information on a documented API with stated limits or on a CDP integration for audience building. 7 1 3
The Skeptic
This is where the evidence is genuinely strong: metrics defined once and computed in the customer's own cloud on Snowflake, Databricks, BigQuery or Redshift, single source of truth, no data duplication, results auditable — a finding I can re-check in the customer's own data. What I found no public information on is a CDP integration for audiences or API rate limits and versioning. 1 4 7
European sovereignty
panel opinion
Show reasoningHide reasoning
How this is scored
Where visitor data is processed and stored and who the contracting entity is. Independently sourced by the sovereignty pipeline; weighted higher here than in categories that hold only the customer's own data, because the script runs on every visitor to the customer's site and their behaviour is what the platform records.
0 — Non-EU vendor and contracting entity, hosting unstated, subprocessors unnamed, and visitor data leaving the EU without a stated safeguard.
3 — EU data residency offered as an option or an enterprise add-on while the contracting entity is non-EU, or the subprocessor list is absent.
5 — EU processing of visitor data as standard and an EU contracting entity, but parts of the chain — CDN, support access, analytics — are non-EU without an explained safeguard.
8 — EU hosting on named infrastructure including the delivery of the snippet, EU contracting entity, subprocessor list published, and a DPA covering the visitor data the script collects.
10 — Sovereign end to end and evidenced: vendor, entity, hosting, snippet delivery and every subprocessor European, certification published, and no visitor data reaching a non-EU party at any point.
The Growth Lead
The contracting entity is Eppo Data, Inc. of San Francisco under Datadog's master subscription agreement, and we found no public information on where the application or snippet delivery is hosted, with subprocessors listed only as categories rather than names. The stated "completely in your own cloud" model keeps experiment analysis in our EU warehouse, but it supplies no European entity, no named hosting and no published data processing agreement for the visitor data path. 2 3 7
The E-Commerce Manager
Everything contractual points across the Atlantic: Eppo Data, Inc. at a San Francisco address, a Datadog master subscription agreement with Datadog billing, and subprocessors named only as broad categories rather than entities. The one mitigation is that analysis runs 'completely in your own cloud', but I found no public information on where the hosted service or snippet-config delivery is processed, no EU data-residency statement, and no DPA covering the visitor data the script collects, so I could not run this on German shoppers with confidence. 2 3 7
The Product Engineer
The contracting chain is American on its face — Eppo Data, Inc. at 435 Brannan St, San Francisco, with use governed by Datadog's master subscription agreement and billing through Datadog — and I found no public information on where the hosted service processes data or which subprocessors are involved, only unnamed categories in the privacy notice. The one mitigation in evidence is architectural: experiment metrics are computed in the customer's own cloud. 2 3 7
The CRO Consultant
Contracting runs through a California entity under its American parent's master subscription agreement, the hosted service's data residency is unstated, and subprocessors are published only as categories rather than named organisations. The one fact a client's DPO can weigh is that experiment data is computed in the customer's own cloud and personal data never leaves the customer's system; we found no public information on EU residency, an EU contracting entity, or a data processing agreement covering the visitor data the platform touches. 2 3 7 4
The Data Protection Officer
The contracting entity is Eppo Data, Inc. of San Francisco under Datadog's master subscription agreement, and I found no public information on where the application or its config CDN are hosted; the privacy notice gives only broad subprocessor categories, with no names and no CDN. The claim that personal data never leaves the customer's system and the customer's-own-cloud warehouse model are real safeguards for experiment data, but no captured page states EU processing of visitor data or names a single European subprocessor. 2 3 7 4
The Skeptic
Contracting runs through Eppo Data, Inc. at a San Francisco address under a Datadog master subscription agreement, subprocessors appear only as categories, and I found no stated residency for the application behind the eppo.cloud login. The single mitigating fact is that the analysis runs in the customer's own cloud — which is the customer's location, not the vendor's commitment to European processing. 2 3 7
Pricing transparency
not rated — the vendor publishes no price
Show reasoningHide reasoning
How this is scored
A category priced by traffic — monthly tracked users, visitors or impressions — where the tier a site lands in depends on numbers the buyer has to estimate. Whether a buyer can compute the real annual cost including traffic limits, overage, server-side or personalisation modules and seats — from public pages alone.
0 — No public prices at all; every tier is a sales conversation.
3 — A starting price or a free tier exists, but the traffic metric, the limits and what happens above them are unstated — the invoice is unknowable.
5 — Tier prices public with the traffic metric and its limits defined, but at least one commonly needed piece (server-side SDKs, personalisation, overage) is unpriced or "contact sales".
8 — Every tier priced publicly with the traffic metric defined, limits, overage rates, module prices, minimum term and VAT treatment stated.
10 — Complete price computability: annual invoice derivable for a given traffic volume, set of modules and team size, with overage and every add-on published.
The Growth Lead
No public prices at all: the captured pages show an order-and-billing regime with no automatic renewal but not a single tier, traffic metric, limit or overage rate. The real annual cost is unknowable without a sales conversation, which is precisely the position I refuse to put myself in with procurement. 3 1
The E-Commerce Manager
No public prices at all: the homepage and the terms give me Order-based contract language, no free tier, no traffic metric, no limits and no module pricing — every tier is a sales conversation. For a shop whose December traffic doubles, an invoice I cannot compute in advance is disqualifying; the only priced item I could find is support, stated as included at no additional charge. 1 3
The Product Engineer
No public prices at all: the terms describe an Order-based subscription that does not automatically renew, with support included at no additional charge, but no tier, traffic metric, limit or overage rate appears on any captured page. Every number is a sales conversation. 3
The CRO Consultant
We found no public prices on any captured page — no tiers, no traffic metric, no limits, no free tier — so every engagement starts as a sales conversation and I cannot estimate a client's annual cost before the contract. The terms add that orders do not auto-renew and support is included at no additional charge, which helps once signed but does not make the invoice computable. 3 1
The Data Protection Officer
I found no public prices at all: the terms speak of Orders and Order Terms, confirm support is included at no additional charge, and the captured pages publish no tier price, traffic metric or limit. Every price is therefore a sales conversation, and a buyer cannot estimate an annual cost from public information. 3 1
The Skeptic
No public prices on any captured page — no tier, no traffic metric, no limits — so the invoice is not computable from public information. The terms state that support is included at no additional charge and that orders do not auto-renew, which is candid about two side-terms while the price itself stays in the sales conversation. 3
European sovereignty — proven facts
1 of 4 dimensions provenBuilt only from facts shown on the vendor's own pages. A dimension we could not prove is left open, not scored as zero.
| Legal entity | Not determined | — | uncited Report an error |
|---|---|---|---|
| Ownership | Foreign-controlled ⚠ unverified | 0/2 pts | 1 Report an error |
| Data residency | Not determined | — | uncited Report an error |
| Subprocessors | Not determined | — | uncited Report an error |
Where this could be wrong
- Evidence ages. The oldest capture behind this page is from 22 Sep 2026. Vendors change pricing and policies without notice; every fact reflects its source as of the capture date shown in the registry.
- Weak sourcing — Ownership. The page states the acquisition by Datadog but not Datadog's domicile or listing status; Datadog is a US-headquartered company (per its own MSA/trust links), so this is non-EU control, though the excerpt alone does not say where Datadog sits.
- AI can misread a source. Extraction and judgement are automated; a citation guarantees traceability, not infallibility. If something here is wrong, say so — no account needed, every report is decided within 5 business days, and accepted corrections are published.
What we left out
A claim that does not survive our checks costs us the claim, not the page. This is what was taken off this one.
- 2 legal 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. Know more? Tell us
- 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. Know more? Tell us
- 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. Know more? Tell us
Sources (8)
The pages every claim on this page was read from — each one checked, dated, and kept verifiable.
- 1 Vendor homepage www.geteppo.com Checked 22 Sep 2026 Details →
- 2 Privacy policy www.geteppo.com Checked 22 Sep 2026 Details →
- 3 Terms of service www.geteppo.com Checked 22 Sep 2026 Details →
- 4 Experiment types & delivery — found from sitemap www.geteppo.com Checked 1 Oct 2026 Details →
- 5 Experiment types & delivery — found from sitemap www.geteppo.com Checked 1 Oct 2026 Details →
- 6 Snippet performance & flicker — found from sitemap www.geteppo.com Checked 1 Oct 2026 Details →
- 7 Analytics, data export & integrations — found from sitemap www.geteppo.com Checked 1 Oct 2026 Details →
- 8 Analytics, data export & integrations — found from sitemap www.geteppo.com Checked 1 Oct 2026 Details →