Skip to content
Hosted in Europe. Your test data is stored in the European Union.Hosted and stored in EuropeLearn more
probant.
Log inCreate your workspace
From idea to rollout, in one tool.Product overview

Prepare

Decide

Roll out

Scale

Visual editorBuild the variant without a developer, with the assistant.Test plan and durationDuration calculated before the first visitor.IdeasEach idea estimated from your past tests.Metrics and marginNo-code goals, margin calculated from your orders.
ReportsA one-sentence verdict, with revenue and margin in euros.Session recordings and click mapsSee what visitors do with each variant.GuardrailsA test that breaks something stops on its own.
RolloutsStep by step, with automatic rollback.PersonalizationAn experience for one audience, measured against a holdout.Feature flagsTurn a feature on by rule, and off in one click.
Learnings libraryThe effects of your finished tests, on one scale.Installation and integrationsTwo lines of code, in the browser or on the server.Team and governanceRoles, launch approvals, and an audit log.
Explore the demo Pricing
The rules behind every verdict, set before launch.The full method

The method, step by step

How Probant decidesThe peeking trapThe winner’s curseThe statistics engineGlossaryVerify a report
MethodThe winner’s curseWhy the reported gain is too good, on average.See the correction
Revenue per visitor · Free shipping progress bar
Measured lift+6.0%
After correction+4.9%
+€125,700per month after correction, instead of +€153,300

Corrected using the demo store’s 48 finished tests.

Rerun the simulation with your own number of tests and duration.A/A simulator
By site area and by team.All use cases

You are…

Know what pays off, without a statistician.

The features that matter to you:

ReportsA one-sentence verdict, with revenue and margin in euros.Metrics and marginNo-code goals, margin calculated from your orders.Visual editorBuild the variant without a developer, with the assistant.See the use case
Cart · demo storeFree shipping progress barConclusive. Proven gain+€125,700 per month · revenue per visitor +6.0%, range +2.2% to +10.0%

Computed by the engine on a demo test.

By site areaCart and shippingProduct pageCheckoutMobile navigationCategory pages
Explore the demo
Blog, docs, and tools.All resources
A/A simulatorSee for yourself what peeking costs.Why checking a test every day produces false winners.Run the simulator
BlogMethod, measurement, and lessons learned.
Nov 12What is an A/B test? Definition, steps, and an exampleNov 7Measure margin, not just revenueNov 7The winner’s curse: why the announced gain falls short
All articles
Duration calculatorHow many days it takes to detect the effect you want.DocumentationInstall the script, handle consent, send events.Video tutorialsInstall, launch a test, and read a report, on video.ComparisonsProbant vs. other tools, with sources.AB TastyKameleoonVWOOptimizelyAdobe TargetDynamic Yield+11 more
ChangelogWhat changed in Probant, by date.Case study formatContext, test, result, and decision, on the demo store.Migrate to ProbantImport your AB Tasty, VWO, or Kameleoon history, recalculated by Probant.
Articles, docs, and calculators, free to use.
One fixed monthly price, no overage charges.See pricing

How many visitors do you test each month?

A tested visitor is each unique visitor who sees a test or a rollout during the month. It’s the only thing that sets the price.

Billing
Your planEssential€590/monthor €5,900 per yearFor a store that tests continuously, down to revenue per visitor.
  • Up to 250,000 tested visitors per month
  • All features, unlimited tests within your volume
  • No overage charges
Choose Essential
Your bill never goes over your plan. Once you reach the volume, running tests keep going.Billing questions

Probant vs. GrowthBook

Both tools compared point by point, based on GrowthBook’s public documentation, accessed November 11, 2026. Every statement about GrowthBook links to its source.

78 sources from 33 official pages, accessed November 11, 2026.

What GrowthBook offers.

GrowthBook presents itself as a “warehouse-native” platform for experimentation, feature flags, and product analytics.1 Data stays in the customer’s warehouse; only aggregate statistics leave it.2 Its engine is Bayesian by default, with frequentist as an option.345

Probant is for sites that want A/B test verdicts that hold up, with a plan set before launch and guardrails monitored through rollout. It measures what each conversion is worth: the margin on an order, a qualified or signed lead, a paid subscription.

The differences, point by point.

On the left, what GrowthBook’s public documentation says, each sentence with its source. On the right, what Probant does. “Not mentioned” means nothing was found in that documentation, which does not prove a feature is missing.

Probant and GrowthBook, point by point. Note numbers link to the sources at the bottom of the page.
CriterionGrowthBookProbant
Statistical methodTwo engines: Bayesian (default) and frequentist (Student’s t-tests), chosen per organization or per project. Sequential testing is an option of the frequentist engine.345Frequentist, with a sequential plan set before launch: target effect, duration, and checkpoints. The verdict fits in one sentence, with a 95% range. Where to check
PeekingSequential testing with “asymptotic confidence sequences” (Waudby-Smith et al. 2023), which can be turned on by default or per test. Limited to the frequentist engine and to the Pro and Enterprise plans; off by default.6789Decisions are only made at planned checkpoints, with O’Brien-Fleming boundaries: very strict at first, close to the usual threshold at the end. During a rollout, ranges are always valid (mSPRT), so you can check them at any time. Where to check
Traffic split (SRM)Automatic chi-square test on every experiment, with a warning if p < 0.001 (adjustable threshold); the result stays displayed. In Safe Rollouts only, you can choose to pause or cancel the rollout on SRM.10111213The traffic split is checked with a chi-square test every time the report is read. If the split is off, the test moves to “blocked” status and no winner is declared. Where to check
GuardrailsIn a standard experiment, guardrail metrics are shown separately and feed the Decision Framework recommendations (“Roll back now”), with no documented automatic stop. Automatic rollback on guardrail degradation exists in Safe Rollouts (Pro and Enterprise plans).14151617Guardrail metrics are tracked with their range; a guardrail is only breached if the whole range crosses the limit. A breached guardrail blocks the test’s verdict, stops a running test on its own, and during a rollout moves visitors back to the original automatically. Where to check
Gradual rolloutsStepped Ramp Schedules (wait, approval) and Safe Rollouts: steps of 1 → 5 → 10 → 25 → 50% with guardrail monitoring and automatic rollback. Pro and Enterprise plans according to the docs; the pricing page puts Ramp Schedules in Enterprise.16181920Steps of 25% → 50% → 100% by default, 10% → 25% → 50% → 100% in cautious mode. Automatic rollback when a guardrail is breached; an exposed visitor stays exposed in the following steps. Where to check
Multi-armed banditMulti-armed bandits that reallocate traffic based on a single decision metric (Pro and Enterprise plans). Contextual bandits (weights per segment) are limited to the Enterprise plan.212223Optimization test: Thompson sampling, reallocated at most once a day, with a minimum share per variant. It maximizes a conversion rate, or an amount per visitor such as revenue: purchase probability and average order value drawn from their distributions, with values capped at a percentile as in an A/B test. Optionally, shares per device (mobile, tablet, desktop): one draw per device, each with its own learning and floor. A visitor always keeps their variant: a new split only applies to visitors who arrive after it. The screen says so: it maximizes conversions (or the amount per visitor) during the test, without proving that a variant is better. The difference is read period by period (and device by device), with no declared winner; the CSV export gives the same reading. Where to check
Revenue and margin per visitorMean metrics (e.g., revenue per user) and ratios (e.g., average order value) analyzed with a 95% confidence interval in frequentist mode or a posterior distribution in Bayesian mode.242526 Margin per visitor is not mentioned in their public documentation (as of November 11, 2026).Revenue and margin per visitor, each with its range; the gain is stated in euros. Margin is calculated on the server from reconciled orders, the cost of each product (Shopify or CSV), and store fees; no margin verdict below 90% of revenue covered by known costs. Where to check
Winner’s curseNo per-test winner’s curse correction by default (non-informative prior). An optional proper prior (Normal with mean 0, standard deviation 0.3) shrinks lifts toward 0 in Bayesian mode, and the Insights dashboard “de-biases” the program’s cumulative impact with the James-Stein estimator.27282930The reported gain is corrected for the winner’s curse by Bayesian shrinkage toward the effects of the store’s past tests. Where to check
Bayesian modeBayesian engine by default with “Chance to Win” and the lift distribution (violin plot). Non-informative prior by default; proper prior and per-metric priors as options. Risk (expected loss) is computed but no longer shown in the app.313233Optional, off by default: a Bayesian reading next to the verdict, with the probability that the variant does better, the credible interval, and the expected loss of each choice. The prior is centered on zero and as wide as the effects measured by the store’s completed tests; while there are too few of them to estimate it, the Bayesian reading is not shown. The decision always follows the sealed plan; when the two readings disagree, the report explains why. Where to check
Multiple comparisonsCorrection to choose from: none, Holm-Bonferroni (FWER), or Benjamini-Hochberg (FDR), across all goal metrics × variants (and segments). Frequentist engine only; secondary and guardrail metrics are excluded.343536Segment comparisons are corrected with the Holm method. With several variants, the error risk is split across comparisons (Bonferroni). Where to check
Multivariate testingThe docs use “multivariate A/B tests” to mean tests with several variants (A/B/n) through non-boolean flags. No full factorial design or interaction report is documented; the docs consider interactions between parallel tests rare.373839Full factorial design generated by Probant: two to four sections, each with two to four levels, up to sixteen combinations, each served as a variant. Main effects of each section across all traffic and interactions between two sections, corrected with the Holm method on the comparisons stated in the plan. Based on the workspace’s measured traffic, the plan says whether the test can reach a conclusion and suggests removing a level or a section; no best combination is declared without a check. Where to check
Variance reduction (CUPED)CUPED variance reduction (regression adjustment on pre-experiment data) in both engines, on the Pro and Enterprise plans; post-stratification (CUPEDps) is limited to the Enterprise plan.40414243CUPED applied to live test reports, using each visitor’s history before their first exposure; without history, the difference stays unadjusted and the report says so. Where to check
Visual editor“AI Visual Editor” as a Chrome side panel on the live site: AI mode in natural language, point-and-click manual mode, Figma import, global CSS, and custom JavaScript. Pro and Enterprise plans; rendered in production through a front-end or edge SDK.444546Two ways to edit, producing the same changes: a page preview loaded by Probant, without its scripts, or the live page on the site, scripts included. On the site, the editor is loaded by the script already installed, through a short-lived signed link for one draft and one variant, with no browser extension. It only sends the chosen changes to Probant. Where to check
WidgetsReady-made widgets are not mentioned in their public documentation (as of November 11, 2026).Banner, countdown, pop-up, badge, reassurance, action bar, delivery date, social proof, promo code, free shipping, and survey, set up without code in the editor and saved in the workspace library for other tests. A widget is just a set of ordinary changes: escaped text, links over https or to the site, and styles applied without a style attribute, so a strict CSP does not need loosening. The countdown follows the site’s time zone and disappears at the deadline; the pop-in opens on arrival, on exit intent, or after scrolling, closes with the keyboard, and does not reappear during the visit if consent allows. Their small modules only load for visitors in the variant. The delivery date is calculated in the visitor’s browser (cutoff time, business days, public holidays); social proof shows the exact number of purchases received over a rolling week, and only above a threshold. The survey asks up to 8 questions (closed answers, recommendation scale, open question) with conditional logic, on the page or through a link outside the site (email, QR code), based on 70 templates (recommendation, satisfaction, effort, checkout blockers, cancellation, concept test, multi-question surveys…); open answers are redacted before sending (email addresses, phone numbers, cards, IBANs), kept for 180 days, and grouped by theme; each question is read by variant, with its range and response rate, and comparisons across all questions are corrected together; a response through a link is attached to the variant the visitor saw, recalculated by Probant, or stays outside the test. Where to check
Script and flickerScript tag loaded async; native anti-flicker only on the Script Tag, off by default (blank screen, 3.5 s maximum delay by default). The docs recommend the Edge SDK to remove flicker entirely. Script size not stated in the docs.47484950Script under 7 KB compressed, loaded asynchronously, that does not hide the page while it loads. Server-side assignment available, for a display with no flicker at all. Where to check
Granular permissionsFree plan: Admin role only. Pro: standard roles and per-project roles. Custom roles with granular permissions, teams, and a “No Access” role are limited to the Enterprise plan.5152Four fixed roles, plus custom roles the owner builds from a catalog of named permissions: prepare, launch and stop, roll out, approve a launch, write variant JavaScript, manage flags, metrics, margin, members, and keys. No privilege escalation: billing, roles, transfer, and deletion of the workspace stay with the owner; no one can grant more than their own permissions, and an API key cannot exceed its author’s. Where to check
Data hostingGrowthBook Cloud is hosted on AWS, with hosting subprocessors (AWS, MongoDB Atlas, ClickHouse) located in the United States. The eu-west-1 region is only offered for the managed warehouse and the Event Forwarder. Analysis data stays in the customer’s warehouse, and self-hosting is possible.535455565758Database and compute in Amsterdam, with Scaleway, in the European Union. Scaleway, which hosts the database and compute, is a French company. Published subprocessor list: Scaleway (Amsterdam, Netherlands (nl-ams region), in the European Union), Brevo (European Union: servers in France and Germany, encrypted backups in Belgium (Google Cloud), according to Brevo), Mistral AI (European Union by default (European API endpoint), according to Mistral), ClickHouse (option not used), Stripe (Contract with Stripe Payments Europe, Ltd. (Ireland); data transferred to Stripe, LLC (United States), according to Stripe), Sentry (European Union: Frankfurt (Germany), including backups, according to Sentry). Depending on the feature used, Mistral may temporarily transfer data outside the European Union to the subprocessors it publishes; its data processing agreement limits these transfers to countries the EU recognizes as adequate, or covers them with standard contractual clauses. Stripe transfers payment data to Stripe, LLC, in the United States, under the Data Privacy Framework and, failing that, standard contractual clauses. Whatever region is chosen, Sentry may keep the data of Probant’s Sentry account (members, settings, audit log, project descriptions) in the United States; reported errors stay in Frankfurt. Where to check
Consent by defaultBy default, the Script Tag sets a “long-lived” first-party gbuuid cookie and sends exposures to gtag, GTM, and Segment. The data-no-auto-cookies attribute delays the cookie until consent; assignment then happens with a non-persisted identifier.5960616263Consent required by default: no cookie, no assignment, and nothing sent before the visitor agrees. Consent given mid-page: the page does not change in front of the visitor; their variant starts on the next page. First-party cookie, lasting 390 days at most, never extended. Where to check
Pricing modelPriced per user, unlimited traffic: free Starter (3 users), Pro at $40 per seat per month (up to 30 users), Enterprise on quote. The managed warehouse and the CDN are billed by usage above the quotas ($30 per million events, $10 per million CDN requests).6465666768Creating a Probant workspace is free; paid plans cost €590 to €1,990 per month depending on the number of tested visitors (2 months free with annual billing), with no overage charges. Agency plan: 1,000,000 tested visitors per month, shared across client workspaces, from 3 workspaces, for €1,490 per month or €14,900 per year. Where to check

What GrowthBook also does.

Probant shares these points with GrowthBook.

  • A sequential test

    Sequential testing with “asymptotic confidence sequences” (Waudby-Smith et al. 2023), which can be turned on by default or per test. Limited to the frequentist engine and to the Pro and Enterprise plans; off by default.6789

  • A multiple comparisons correction

    Correction to choose from: none, Holm-Bonferroni (FWER), or Benjamini-Hochberg (FDR), across all goal metrics × variants (and segments). Frequentist engine only; secondary and guardrail metrics are excluded.343536

  • Variance reduction (CUPED)

    CUPED variance reduction (regression adjustment on pre-experiment data) in both engines, on the Pro and Enterprise plans; post-stratification (CUPEDps) is limited to the Enterprise plan.40414243

  • Revenue per visitor, with its range

    Mean metrics (e.g., revenue per user) and ratios (e.g., average order value) analyzed with a 95% confidence interval in frequentist mode or a posterior distribution in Bayesian mode.242526

  • Gradual rollouts

    Stepped Ramp Schedules (wait, approval) and Safe Rollouts: steps of 1 → 5 → 10 → 25 → 50% with guardrail monitoring and automatic rollback. Pro and Enterprise plans according to the docs; the pricing page puts Ramp Schedules in Enterprise.16181920

When to choose GrowthBook over Probant.

Based on GrowthBook’s public documentation.

  • You want an open-source tool you can host yourself

    Code mostly under the MIT license (“open core”), with a few directories under an Enterprise license; self-hosting documented with the same code as the cloud.697071

  • Your data already lives in a warehouse

    Analysis runs in the customer’s warehouse (BigQuery, Snowflake, Redshift, Databricks, ClickHouse, Postgres, Athena…); only aggregate statistics come back. A managed ClickHouse warehouse is available if you do not have one.27273

  • You want cluster tests, for example by region

    “Cluster” experiments: randomization of a group (organization, region…) and analysis on the included unit, with a warning below 20 clusters per variant; Pro and Enterprise plans. No switchback test (alternating over time) is documented.7475767778

What sets Probant apart.

These points are not in GrowthBook’s public documentation, accessed November 11, 2026. A missing mention does not prove a feature is missing; if these points matter to you, ask GrowthBook.

  • Margin per visitor

    Probant measures revenue and margin per visitor, each with its range.

    GrowthBookMargin per visitor is not mentioned in their public documentation (as of November 11, 2026).

    Learn more

  • SRM that blocks the verdict

    A traffic split imbalance moves the test to “blocked” status, and no winner is declared.

    GrowthBookAutomatic chi-square test on every experiment, with a warning if p < 0.001 (adjustable threshold); the result stays displayed. In Safe Rollouts only, you can choose to pause or cancel the rollout on SRM.10111213 Automatic blocking of the test is not mentioned there (as of November 11, 2026).

  • Consent required by default

    No cookie, no assignment, and nothing sent before the visitor agrees.

    GrowthBookBy default, the Script Tag sets a “long-lived” first-party gbuuid cookie and sends exposures to gtag, GTM, and Segment. The data-no-auto-cookies attribute delays the cookie until consent; assignment then happens with a non-persisted identifier.5960616263 Consent required by default is not mentioned there (as of November 11, 2026).

  • A test stopped by its guardrails

    A breached guardrail stops a running test on its own and blocks its verdict.

    GrowthBookIn a standard experiment, guardrail metrics are shown separately and feed the Decision Framework recommendations (“Roll back now”), with no documented automatic stop. Automatic rollback on guardrail degradation exists in Safe Rollouts (Pro and Enterprise plans).14151617 An automatic stop of the test on a guardrail is not mentioned there (as of November 11, 2026).

    Learn more

Frequently asked questions.

How much does GrowthBook cost?

Priced per user, unlimited traffic: free Starter (3 users), Pro at $40 per seat per month (up to 30 users), Enterprise on quote. The managed warehouse and the CDN are billed by usage above the quotas ($30 per million events, $10 per million CDN requests).6465666768 Creating a Probant workspace is free; paid plans cost €590 to €1,990 per month depending on the number of tested visitors (2 months free with annual billing), with no overage charges.

Where does GrowthBook host data?

GrowthBook Cloud is hosted on AWS, with hosting subprocessors (AWS, MongoDB Atlas, ClickHouse) located in the United States. The eu-west-1 region is only offered for the managed warehouse and the Event Forwarder. Analysis data stays in the customer’s warehouse, and self-hosting is possible.535455565758 Probant hosts its database and application in Amsterdam, with Scaleway.

Does GrowthBook measure margin?

Margin per visitor is not mentioned in their public documentation (as of November 11, 2026). Probant measures revenue and margin per visitor, each with its range, from orders and the cost of each product.

Can you look at GrowthBook results while a test is running?

Sequential testing with “asymptotic confidence sequences” (Waudby-Smith et al. 2023), which can be turned on by default or per test. Limited to the frequentist engine and to the Pro and Enterprise plans; off by default.6789 Probant sets its checkpoints before launch (O’Brien-Fleming) and uses always-valid ranges during rollouts.

Sources.

Official pages from GrowthBook, accessed November 11, 2026. Quotes are reproduced in their original language.

  1. Source 1: www.growthbook.io/about
    GrowthBook is the warehouse-native platform for experimentation, feature flags, and product analytics trusted by AI-native product teams at 4,000+ companies worldwide.
  2. Source 2: docs.growthbook.io/warehouses
    One of the major benefits of GrowthBook is that your data remains securely stored in your data warehouse, and only the aggregated statistics are transmitted to GrowthBook servers or your self-hosted environment.
  3. Source 3: docs.growthbook.io/statistics/overview
    GrowthBook provides both Bayesian and frequentist approaches to experiment analysis. We default to Bayesian statistics because they provide a more intuitive framework for decision making for most customers
  4. Source 4: docs.growthbook.io/statistics/overview
    You can choose between the two statistics engines at the Organization or Project level.
  5. Source 5: docs.growthbook.io/statistics/overview
    The current frequentist engine computes two-sample t-tests for relative percent change; you can reduce variance (via CUPED) and you can enable sequential testing to mitigate concerns with peeking.
  6. Source 6: docs.growthbook.io/statistics/sequential
    Sequential Testing is available on Pro and Enterprise plans.
  7. Source 7: docs.growthbook.io/statistics/sequential
    Sequential Testing is only implemented for the Frequentist statistics engine.
  8. Source 8: docs.growthbook.io/statistics/sequential
    Sequential testing allows you to look at your experiment results as many times as you like while still keeping the number of false positives below the expected rate. In other words, it is the frequentist solution to the peeking problem in AB testing.
  9. Source 9: docs.growthbook.io/statistics/sequential
    To that end, we implement Asymptotic Confidence Sequences introduced by Waudby-Smith et al. (2023)
  10. Source 10: docs.growthbook.io/app/experiment-results
    Every experiment automatically checks for a Sample Ratio Mismatch (SRM) and will warn you if found.
  11. Source 11: docs.growthbook.io/app/experiment-results
    We only show this warning if the p-value is less than 0.001 (customizable in your Organization Settings)
  12. Source 12: docs.growthbook.io/app/experiment-results
    Under the hood, we are conducting a standard chi-squared test for Sample Ratio Mismatch
  13. Source 13: docs.growthbook.io/features/safe-rollouts
    Under Advanced Settings, you can choose how GrowthBook responds to other issues, such as a sample ratio mismatch, no traffic, or multiple exposures: hold the ramp, roll back, or warn only.
  14. Source 14: docs.growthbook.io/app/experiment-results
    Guardrail metrics are treated much like regular metrics, but they are placed in a separate part of the results view and are not part of any p-value corrections in the frequentist engine
  15. Source 15: docs.growthbook.io/app/experiment-decisions
    Clear Signals, which requires no guardrail failures and all goal metric successes
  16. Source 16: docs.growthbook.io/features/safe-rollouts
    GrowthBook automatically rolls back and disables the rule if any of them regress significantly, so there is no separate auto-rollback toggle.
  17. Source 17: docs.growthbook.io/features/safe-rollouts
    Safe Rollout is available on Pro and Enterprise plans.
  18. Source 18: docs.growthbook.io/features/ramp-schedules
    A Ramp Schedule rolls a feature out in stages instead of switching it on for everyone at once.
  19. Source 19: docs.growthbook.io/features/safe-rollouts
    1% → 5% → 10% → 25% → 50%
  20. Source 20: docs.growthbook.io/features/ramp-schedules
    Ramp Schedules is available on Pro and Enterprise plans.
  21. Source 21: docs.growthbook.io/bandits/overview
    When your bandit is running, we recalculate the percent of traffic to allocate to each variation based on which variation (or arm) is performing better on a single Decision Metric.
  22. Source 22: docs.growthbook.io/bandits/overview
    Multi Armed Bandits is available on Pro and Enterprise plans.
  23. Source 23: docs.growthbook.io/bandits/contextual
    Contextual Bandits is available on Enterprise plans.
  24. Source 24: docs.growthbook.io/app/metrics
    A Revenue per User metric would use SUM and select the column where the order total is stored.
  25. Source 25: docs.growthbook.io/app/metrics
    Ratio Metrics let you pick a custom denominator and allow for metrics such as Average Order Value (denominator is number of orders) or Session Duration (denominator is number of sessions).
  26. Source 26: docs.growthbook.io/app/experiment-results
    The graph now represents a 95% confidence interval
  27. Source 27: docs.growthbook.io/statistics/overview
    At GrowthBook, we default to an improper, uninformative prior.
  28. Source 28: docs.growthbook.io/statistics/overview
    By default, we use a Normal distribution with mean 0 and standard deviation 0.3.
  29. Source 29: docs.growthbook.io/statistics/overview
    In effect, it will shrink positive and negative results towards 0, but eventually will be overcome as more data is collected.
  30. Source 30: docs.growthbook.io/insights
    De-biasing is achieved through the positive-part James-Stein shrinkage estimator, which mitigates the natural bias in experiment outcomes (where more decisive results tend to show larger effects).
  31. Source 31: docs.growthbook.io/statistics/overview
    Chance to Win is straight forward. It is simply the probability that a variation is better.
  32. Source 32: docs.growthbook.io/overview
    Custom Priors - For Bayesian stats, set custom prior distributions for each metric
  33. Source 33: docs.growthbook.io/statistics/details
    Note: we no longer use risk in the GrowthBook app, but still compute it in the experiment results database.
  34. Source 34: docs.growthbook.io/statistics/multiple-corrections
    Growthbook allows you to choose no multiple testing correction, a correction that controls the Family-wise Error Rate (Holm-Bonferroni), and a correction that controls the False Discovery Rate (Benjamini-Hochberg).
  35. Source 35: docs.growthbook.io/statistics/multiple-corrections
    Multiple testing corrections are only implemented for the Frequentist statistics engine.
  36. Source 36: docs.growthbook.io/statistics/multiple-corrections
    We correct using all p-values for all Goal metrics; this excludes Secondary and Guardrail metrics.
  37. Source 37: docs.growthbook.io/features/basics
    Remote config, multivariate A/B tests
  38. Source 38: docs.growthbook.io/overview
    Experiment - Run a controlled hypothesis test between 2 or more feature values
  39. Source 39: docs.growthbook.io/using/fundamentals
    In practice, meaningful interaction effects are rare. More often, running tests in parallel simply increases variance without changing the overall results.
  40. Source 40: docs.growthbook.io/statistics/cuped
    CUPEDps (Controlled-experiment Using Pre-Experiment Data + post-stratification) increases the velocity of experimentation by reducing the uncertainty in estimates of experiment uplift.
  41. Source 41: docs.growthbook.io/statistics/cuped
    CUPED is available on Pro and Enterprise plans.
  42. Source 42: docs.growthbook.io/statistics/cuped
    Post Stratification is available on Enterprise plans.
  43. Source 43: docs.growthbook.io/statistics/overview
    In fact, tools like CUPED are available for both engines.
  44. Source 44: docs.growthbook.io/app/visual
    The AI Visual Editor is a Chrome side-panel extension.
  45. Source 45: docs.growthbook.io/app/visual
    AI Visual Editor is available on Pro and Enterprise plans.
  46. Source 46: docs.growthbook.io/app/visual/manual-mode
    Inject a JavaScript block. Use sparingly — it's an escape hatch for changes that genuinely need imperative code
  47. Source 47: docs.growthbook.io/lib/script-tag
    <script async data-client-key="YOUR_CLIENT_KEY_HERE"
  48. Source 48: docs.growthbook.io/app/url-redirects
    The Script Tag is the only front-end SDK that natively provides an anti-flicker mechanism (off by default).
  49. Source 49: docs.growthbook.io/app/url-redirects
    antiFlickerTimeout: The maximum number of milliseconds that the anti-flicker blank screen will be shown. Defaults to 3500 (3.5 seconds).
  50. Source 50: docs.growthbook.io/app/visual/running-on-your-site
    It eliminates flicker entirely and produces a far better experience for users.
  51. Source 51: docs.growthbook.io/account/user-permissions
    On the Free plan, members can only be assigned the Admin role — assigning any other role (including custom roles) requires a Pro or Enterprise account.
  52. Source 52: docs.growthbook.io/account/user-permissions
    Enterprise organizations using GrowthBook also have the added flexibility of defining custom roles, which enable organizations to fine-tune a role's permissions.
  53. Source 53: www.growthbook.io/legal/subprocessors
    Amazon Web Services, Inc. Cloud hosting provider United States
  54. Source 54: www.growthbook.io/platform/security
    Hosted by GrowthBook on AWS
  55. Source 55: www.growthbook.io/legal/subprocessors
    MongoDB, Inc. Data storage and processing United States MongoDB Atlas is hosted on AWS and provides a database for GrowthBook.
  56. Source 56: www.growthbook.io/legal/subprocessors
    ClickHouse, Inc. Data storage and processing United States ClickHouse Cloud stores all event data sent to the Managed Warehouse product.
  57. Source 57: docs.growthbook.io/app/managed-warehouse
    If you selected eu-west-1 as your Data Region when creating your Managed Warehouse, set the ingestor host option shown for your SDK above to https://eu-west-1.gb-ingest.com.
  58. Source 58: www.growthbook.io/platform/deployment-options
    Self-Hosted GrowthBook gives you maximum control: deploy within your infrastructure, behind your firewall, with your own security policies applied.
  59. Source 59: docs.growthbook.io/lib/script-tag
    By default, this SDK persists a random unique identifier in a first-party cookie named gbuuid.
  60. Source 60: docs.growthbook.io/lib/script-tag
    id - creates a long-lived gbuuid cookie if it doesn't exist already
  61. Source 61: docs.growthbook.io/lib/script-tag
    If you must delay persisting the gbuuid cookie until a user consents, you can add a data-no-auto-cookies attribute to the script tag.
  62. Source 62: docs.growthbook.io/lib/script-tag
    This will still generate a UUID for the user, but will not persist it.
  63. Source 63: docs.growthbook.io/lib/script-tag
    Defaults to gtag,gtm,segment, and none turns tracking off.
  64. Source 64: www.growthbook.io/pricing
    Pro For small and mid-sized teams with advanced statistics and analytics. $40 / seat / month Up to 30 users
  65. Source 65: www.growthbook.io/pricing
    Enterprise For teams running experimentation and releasing features at scale. Custom
  66. Source 66: www.growthbook.io/pricing
    Managed Warehouse 1M events/mo 2M events/mo (then $30/M) Custom
  67. Source 67: www.growthbook.io/pricing
    CDN Requests 1M/mo 2M/mo (then $10/M) Custom
  68. Source 68: www.growthbook.io/platform/deployment-options
    Only cloud offers a Pro tier at a low $40 per month, per seat.
  69. Source 69: github.com/growthbook/growthbook/blob/main/README.md
    GrowthBook is an Open Core product. The bulk of the code is under the permissive MIT license.
  70. Source 70: github.com/growthbook/growthbook/blob/main/LICENSE
    Content outside of the above mentioned directories or restrictions above is available under the "MIT Expat" license as defined below.
  71. Source 71: www.growthbook.io/platform/deployment-options
    Same codebase, same features, same development roadmap.
  72. Source 72: docs.growthbook.io/overview
    No raw user-level events or PII are ever sent to GrowthBook.
  73. Source 73: docs.growthbook.io/app/managed-warehouse
    We create 3 tables in a ClickHouse database managed by GrowthBook:
  74. Source 74: docs.growthbook.io/experimentation-analysis/cluster-experiments
    A cluster experiment randomizes a group (the cluster) but analyzes a smaller unit inside it (the subunit).
  75. Source 75: docs.growthbook.io/experimentation-analysis/cluster-experiments
    A marketplace randomizes by region to limit interference between buyers and sellers, but measures conversion per user.
  76. Source 76: docs.growthbook.io/experimentation-analysis/cluster-experiments
    Cluster Experiments is available on Pro and Enterprise plans.
  77. Source 77: docs.growthbook.io/experimentation-analysis/cluster-experiments
    GrowthBook warns when any variation has fewer than 20 clusters.
  78. Source 78: docs.growthbook.io/experimentation-analysis/interleaving
    An A/B test (or a switchback or synthetic control test where interference is a risk) on the one or two finalists

Migrate from GrowthBook

Your GrowthBook history can be imported and is reread by Probant’s engine: every verdict is recalculated from visitors and conversions.

Migrate from GrowthBookExplore the demo
  • Open demo, no sign-up
  • Free to start, no overage charges
  • Database and compute in Amsterdam, with Scaleway