Probant vs. VWO
Both tools compared point by point, based on VWO’s and Wingify’s public documentation, accessed November 7, 2026. Every statement about VWO links to its source.
VWO is now presented under the Wingify brand.
AB Tasty and VWO merged under the Wingify brand (announced January 20, 2026).12 vwo.com and help.vwo.com now redirect to wingify.com and help.wingify.com.23 Wingify states that it keeps running both platforms, VWO and AB Tasty.4 The facts below come from the VWO platform’s help center, now help.wingify.com, and from Wingify’s pages.
What VWO offers.
VWO is published by Wingify Software, whose main address is in Delhi, India.5 Wingify presents itself as an “agentic experience optimization platform.”6 It claims more than 4,000 customers.7 Plans range from “Growth,” for small teams getting started, to “Enterprise,” for large teams, agencies, and large accounts.89
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 VWO’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.
| Criterion | VWO | Probant |
|---|---|---|
| Statistical method | Bayesian model by default; frequentist can be chosen before launch.10 In Bayesian mode, decisions rely on a region of practical equivalence (ROPE).11 | Frequentist, 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 |
| Peeking | Three approaches: sequential, fixed horizon, or “dynamic,” with no sample bounds.12 In sequential mode, the threshold adjusts to the number of looks, through an alpha-spending variant based on the planned sample size.1314 | Decisions 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) | SRM check among the report’s “Experiment Vitals.”15 When it fails, the engine warns that the results are not reliable.16 Included from the Pro plan.17 Automatic blocking of the test is not mentioned in their public documentation (as of November 7, 2026). | The 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 |
| Guardrails | When breached: a notification, or automatic disabling of the variant involved.18 Guardrails set with a minimum detectable reduction, power, and false positive rate; from the Pro plan.1719 | Guardrail 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 rollouts | Gradual rollout rules by segment and percentage, with steps scheduled over time.2021 Metric-based automation, down to 0%.22 Web rollout campaigns from the Growth plan.23 | Steps 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 bandit | “Multi-Arm Bandit” campaigns listed in Wingify’s plans.24 | Optimization 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 visitor | Revenue per visitor, with a 99% credible interval in the report.2526 RevenueIQ, from AB Tasty, appears on the pricing page.27 Margin per visitor: not mentioned in their public documentation (as of November 7, 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 curse | Not mentioned in their public documentation (as of November 7, 2026). | The 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 mode | Bayesian model by default; frequentist can be chosen before launch.10 Decision through a region of practical equivalence (ROPE).11 Former SmartStats engine: a winner when the potential loss falls below a threshold and the chance to beat all other variants reaches 95%.28 | Optional, 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 comparisons | Optional Bonferroni correction for tests with several variants.29 | Segment comparisons are corrected with the Holm method. With several variants, the error risk is split across comparisons (Bonferroni). Where to check |
| Multivariate testing | Several elements of a page change together to find the best combination.30 The documentation reserves it for high-traffic sites.31 | Full 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) | Not mentioned in their public documentation (as of November 7, 2026). | CUPED 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 | Without the SmartCode on the page, the editor loads it through a proxy server, which the site can block; the Chrome extension is then the fallback.3233 | Two 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 |
| Widgets | Widgets added without code in the visual editor, including a countdown.3435 Adjustable display frequency; widgets saved and reused across the site.3637 | 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 flicker | The asynchronous SmartCode hides the page body by default while loading, with timeouts of 2,000 and 2,500 ms.3839 Wingify recommends the synchronous SmartCode.40 Version 3.0 reduces the payload by 40%; an absolute size is not mentioned in their public documentation (as of November 7, 2026).41 | Script 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 permissions | Five roles with predefined permissions: Owner, Admin, Publish, Design, and Browse.42 | Four 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 hosting | EU data center in Belgium (GCP europe-west1), optional; other data centers in the United States, India, and Australia.43 With the EU data center, visitor data stays in the EU; account data and campaign metadata stay in the United States.4445 | Database 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 default | By default, consent management only applies to Behavior Analytics; an option extends it to web experimentation.46 Some customers run the SmartCode without consent, classifying its cookies as necessary.47 The “Completely Block” option creates no cookie while consent is pending.48 | Consent 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 model | On quote, based on traffic, products, plan, and scope; based on monthly active users (MAU).4950 Three plans, Growth, Pro, and Enterprise, with no displayed prices.9 30-day free trial, no credit card required.51 Overage policy: not mentioned in their public documentation (as of November 7, 2026). | 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. 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 VWO also does.
Probant shares these points with VWO.
A sequential peeking correction
Alpha-spending adjusts the threshold to the planned sample size, an approach close to Probant’s checkpoints.14
Guardrails that act on their own
A variant that breaches a critical guardrail can be disabled automatically.18
Rollouts with automatic rollback
Scheduled steps, and a return to 0% if a metric degrades.2122
An SRM check
SRM is part of the report’s integrity checks.15
Visitor data in the EU
No cookie before consent, as an option
The “Completely Block” option creates no cookie while consent is pending.48
When to choose VWO over Probant.
Based on VWO’s public documentation.
You want behavior analytics and personalization alongside tests
The range also covers Behavior Analytics, web personalization, and rollouts.4650
You prefer a Bayesian reading
The Bayesian model is the default, with decisions based on a region of practical equivalence.1011
You want to try it right away
You need data centers outside Europe
United States, India, and Australia, in addition to the EU data center.43
What sets Probant apart.
These points are not in VWO’s public documentation, accessed November 7, 2026. A missing mention does not prove a feature is missing; if these points matter to you, ask VWO.
Margin per visitor
Probant measures revenue and margin per visitor, each with its range.
VWOMargin per visitor is not mentioned in their public documentation (as of November 7, 2026).
Learn moreWinner’s curse correction
The reported gain is corrected by Bayesian shrinkage toward the effects of the store’s past tests.
VWOA winner’s curse correction is not mentioned in their public documentation (as of November 7, 2026).
Learn moreSRM that blocks the verdict
A traffic split imbalance moves the test to “blocked” status, and no winner is declared.
VWOWingify describes a notification that results are unreliable.16 Automatic blocking is not mentioned there (as of November 7, 2026).
A published script size
Probant’s script weighs less than 7 KB compressed and does not hide the page while it loads.
VWOWingify publishes a relative reduction of the SmartCode; an absolute size is not mentioned in their public documentation (as of November 7, 2026).41
Learn more
Frequently asked questions.
Does VWO still exist after the merger with AB Tasty?
Yes. Wingify states that it keeps running both platforms.4 Existing contracts and pricing are announced as unchanged.52
Does VWO correct for peeking?
In sequential mode, yes: you can check the report at any time without inflating the false positive rate.13 The “dynamic” mode checks significance continuously, with no sample bounds.12 In Probant, an A/B test is only decided at checkpoints set before launch; only rollouts and personalizations, which have no planned end, are read continuously, with always-valid ranges.
Sources.
Official pages from VWO and Wingify, accessed November 7, 2026. Quotes are reproduced in their original language.
- Source 1: wingify.com/blog/vwo-and-ab-tasty-join-forces
I am incredibly excited to announce that VWO and AB Tasty are joining forces, subject to customary closing conditions.
- Source 4: wingify.com/llm-info
The merger expanded Wingify’s product capabilities, global reach, and expertise while continuing to support the trusted VWO and AB Tasty platforms.
- Source 5: wingify.com/imprint
Wingify Software Pvt. Ltd. doing business as Wingify Address: India Office 703, 7th floor, KLJ North Tower, Netaji Subhash Place, Pitampura, Delhi-110034, India
- Source 7: wingify.com/llm-info
Wingify serves more than 4,000 customers worldwide from 11 offices across North America, Europe, Asia Pacific, and Latin America.
- Source 9: wingify.com/pricing
Enterprise For large teams, agencies & enterprises with advanced requirements
- Source 10: help.wingify.com/hc/en-us/articles/58838252139545-Achieve-Accurate-Campaign-Results-with-SmartStats-Configuration
The Bayesian Model is the default. It is selected automatically when you create a new campaign. You can switch to the Frequentist Model before starting the campaign.
- Source 11: help.wingify.com/hc/en-us/articles/58838252139545-Achieve-Accurate-Campaign-Results-with-SmartStats-Configuration
Evaluates when the improvement distribution crosses the ROPE boundary.
- Source 12: help.wingify.com/hc/en-us/articles/58838252139545-Achieve-Accurate-Campaign-Results-with-SmartStats-Configuration
Retrospective testing with no fixed sample size bounds. Continuously checks for statistical significance until the campaign is paused.
- Source 13: help.wingify.com/hc/en-us/articles/58838252139545-Achieve-Accurate-Campaign-Results-with-SmartStats-Configuration
Sequential Testing adjusts statistical thresholds dynamically to account for how many times you have checked the results. You can review the report at any time without inflating the false positive rate.
- Source 14: help.wingify.com/hc/en-us/articles/58838274775705-Understanding-Sequential-Testing-Correction
Wingify utilizes a variation of a statistical approach called Alpha-spending. This method adjusts the decision threshold for declaring a winner based on the planned sample size.
- Source 15: help.wingify.com/hc/en-us/articles/58838251378457-Keep-Your-Campaigns-Healthy-With-Experiment-Vitals
Click the heart icon in the report header to view key integrity checks such as Data Tracking, Conversion Tracking, Minimum Runtime Alert, Guardrails, Experimentation Conduct, and Sample Ratio Mismatch (SRM).
- Source 16: wingify.com/pricing
Sanity check to identify data collection and test setup discrepancies. If an SRM check fails, our stats engine will notify you of unreliable campaign results.
- Source 17: wingify.com/pricing
Pro For businesses looking to scale experimentations Everything in Growth plan, plus Advanced Targeting, Triggers & Reporting Guardrails Metrics Sample Ratio Mismatch Detection
- Source 18: help.wingify.com/hc/en-us/articles/58838315584793-Understanding-Guardrails-in-A-B-Testing
In the event of a guardrail breach, Wingify offers two options: you can either receive a notification alerting you to the breach or, for more critical guardrails, disable the variation that caused the breach and then notify you.
- Source 19: help.wingify.com/hc/en-us/articles/58838315584793-Understanding-Guardrails-in-A-B-Testing
Minimum Detectable Reduction (MDR): It represents the maximum reduction you want to be able to confidently detect as statistically significant.
- Source 20: help.wingify.com/hc/en-us/articles/58766111504409-Set-Up-Rules-in-Feature-Management
Instead of launching a new feature to 100% of your audience at once, you can define specific audience segments and percentages of traffic to receive the feature over time.
- Source 21: help.wingify.com/hc/en-us/articles/58766111504409-Set-Up-Rules-in-Feature-Management
Time-Based Automation: Schedule percentage increases (or decreases) on specific dates.
- Source 22: help.wingify.com/hc/en-us/articles/58766111504409-Set-Up-Rules-in-Feature-Management
Safety Rollback: Start at 10%. If the Error Rate (metric) goes above 0.5% at any point, immediately roll back to 0%.
- Source 25: help.wingify.com/hc/en-us/articles/59066072471833-Interpreting-an-A-B-Test-Report
Expected conversion rate/Expected revenue per visitor: This is the median conversion rate you can expect from the variation. The "best case" and "worst case" conversion rates represent the 99% credible interval where the conversion rate is likely to be contained.
- Source 26: help.wingify.com/hc/en-us/articles/58840369824537-Wingify-Campaign-Reports-Statistical-Models-and-Inference-Levels
For continuous metrics, for example, revenue per visitor, it is labeled Total value (Unique Conversions/Visitors).
- Source 28: help.wingify.com/hc/en-us/articles/59066062029977-How-Wingify-Calculates-a-Winning-Variation
We call a variation a winner when its potential loss is below the threshold of caring and its chance to beat all is more than or equal to 95%.
- Source 29: help.wingify.com/hc/en-us/articles/58838252139545-Achieve-Accurate-Campaign-Results-with-SmartStats-Configuration
Bonferroni Correction compensates for this by tightening the significance threshold for each individual comparison, keeping the experiment-wide false positive rate at the configured level.
- Source 30: help.wingify.com/hc/en-us/articles/58790783421209-Introduction-to-Multivariate-Testing
multiple elements on a single page are changed to determine the optimal combination of those changes.
- Source 31: help.wingify.com/hc/en-us/articles/58790783421209-Introduction-to-Multivariate-Testing
is useful only when the website receives high traffic
- Source 32: help.wingify.com/hc/en-us/articles/58870730666009-Why-is-My-Website-Not-Loading-Correctly-in-the-Wingify-Visual-Editor
This usually happens when the Wingify SmartCode is missing from the webpage you are trying to edit, and your servers block the page from loading via proxy.
- Source 33: help.wingify.com/hc/en-us/articles/58870730666009-Why-is-My-Website-Not-Loading-Correctly-in-the-Wingify-Visual-Editor
If you cannot install the SmartCode right away, add the Wingify chrome extension to your browser.
- Source 34: help.wingify.com/hc/en-us/articles/58838242926361-Add-Widgets-in-Visual-Editor-to-Create-Variations
The Visual Editor lets you add new, interactive widgets to your website variations without coding.
- Source 35: help.wingify.com/hc/en-us/articles/58838242926361-Add-Widgets-in-Visual-Editor-to-Create-Variations
Displays a live clock counting down to a specific time or duration.
- Source 36: help.wingify.com/hc/en-us/articles/58838242926361-Add-Widgets-in-Visual-Editor-to-Create-Variations
Control how often the widget is shown (for example, Every time, Once per session, or Every X days).
- Source 37: wingify.com/pricing
Build, store, and reuse widgets to quickly deploy consistent experiences across your website.
- Source 38: help.wingify.com/hc/en-us/articles/58762214319257-Customize-your-Wingify-SmartCode
The default functionality of the asynchronous SmartCode is to hide the body of the web page initially. This is to avoid any flash of the original content on the web page while Wingify applies any change.
- Source 39: help.wingify.com/hc/en-us/articles/58762214319257-Customize-your-Wingify-SmartCode
By default, the time-out values for the two parameters are 2000 and 2500 milliseconds, respectively.
- Source 40: help.wingify.com/hc/en-us/articles/58925849963033-SmartCode-3-0-What-s-New-and-How-It-Works
It is recommended that you use the Synchronous SmartCode. Place it at the end of the <head> section of your website.
- Source 41: help.wingify.com/hc/en-us/articles/58925849963033-SmartCode-3-0-What-s-New-and-How-It-Works
Wingify serves only the JS logic your account actively uses (feature-based serving), resulting in a 40% reduction in payload size.
- Source 42: help.wingify.com/hc/en-us/articles/58759151167769-Roles-and-Permissions-in-Wingify
By assigning roles with predefined permissions, you can ensure that users have the right level of access based on their responsibilities.
- Source 43: help.wingify.com/hc/en-us/articles/58766336786201-Wingify-Data-Residency
EU data center: Situated in Belgium (europe-west1 of GCP), this data center caters to those who prefer to store and manage their visitor data within the EU.
- Source 44: wingify.com/compliance/gdpr
If you choose EU data centre, all your visitor/user data shall be stored exclusively in the EU and will not be transferred into a country that is not a member of the EU.
- Source 45: help.wingify.com/hc/en-us/articles/58766336786201-Wingify-Data-Residency
But Wingify’s actual customer account data, such as Wingify account information and Wingify campaign metadata, is stored only in the US data center.
- Source 46: help.wingify.com/hc/en-us/articles/58793817111961-Configuring-Wingify-SmartCode-for-Your-Website-s-Cookie-Consent
By default, Wingify’s behaviour and data collection based on cookie consent status is applied to Wingify Behavior Analytics. To apply it to Web Experimentation, Web Personalization, and Web Rollout, turn on the Enable Advance Cookie consent options toggle.
- Source 47: help.wingify.com/hc/en-us/articles/58793866121881-Cookie-Consent-and-Wingify-s-Adherence
Some of our customers have classified Wingify's cookies as necessary, allowing the Wingify SmartCode to be executed without the visitor's consent, while others have not and require the visitor's consent before executing the Wingify SmartCode.
- Source 48: help.wingify.com/hc/en-us/articles/58793817111961-Configuring-Wingify-SmartCode-for-Your-Website-s-Cookie-Consent
If the Completely Block Wingify option is selected and no timeout is set, Wingify does not create any cookies.
- Source 49: wingify.com/pricing
Wingify offers custom pricing tailored to your needs, based on factors like traffic volume, selected products & plans, implementation scope etc.
- Source 50: wingify.com/pricing
Yes. Pricing is based on your Monthly Active Users (MAUs), depending on the product. However, for Behavioural Analytics, pricing is based on Monthly Tracked Users (MTUs) which can be subset of your MAU.
- Source 51: wingify.com
You can start on a 30-day free trial with no credit card, or book a demo for a guided setup.
- Source 52: wingify.com/blog/vwo-and-ab-tasty-join-forces
All existing client relationships, contracts, pricing, service levels, and support remain unchanged.
Migrate from VWO
Your VWO history can be imported and is reread by Probant’s engine: every verdict is recalculated from visitors and conversions.
- Open demo, no sign-up
- Free to start, no overage charges
- Database and compute in Amsterdam, with Scaleway