Probant vs. Intelligems
Both tools compared point by point, based on Intelligems’s public documentation, accessed November 11, 2026. Every statement about Intelligems links to its source.
What Intelligems offers.
Intelligems presents itself as the profit optimization and A/B testing platform built for Shopify.1 Based in New York, it raised $9 million in a Series A.23 Its engine is Bayesian.4
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 Intelligems’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 | Intelligems | Probant |
|---|---|---|
| Statistical method | A Bayesian engine with Monte Carlo simulations; a frequentist mode is not mentioned. The vendor states that the model does not account for day-of-week seasonality.4 | 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 | No sequential test or always-valid interval is documented. The documentation recommends waiting for minimums (300 orders per group, 7 days); an automatic verdict through the API (opt-in) adds a stability criterion over the last few days.56 | 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) | No native SRM check in the interface is documented. The vendor provides a guide to build a Slack alert yourself (n8n + API) that includes a chi-square test, and a second guide to pause failing tests automatically.7 | 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 | No native guardrail metric is documented. An official guide explains how to pause a test automatically when it crosses health thresholds (visitors, SRM, conversion drop) with an external workflow (n8n) calling the API.8 | 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 | Deploying a winning variant ends the test and creates a personalization that carries over its changes.9 A stepped rollout with automatic rollback is not mentioned in their public documentation (as of November 11, 2026). | 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 | No native dynamic allocation is documented. An official guide shows how to build a multi-armed bandit (n8n + API) that shifts traffic in steps up to 80/20.10 | 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, average order value, and other revenue metrics, each with significance and 95% intervals (credible intervals from the Bayesian model).1112 Gross profit per visitor and per order, with intervals: COGS per variant (CSV or Shopify sync), average shipping cost per order, and a transaction fee percentage. COGS apply retroactively; missing COGS count as 0 (warning below 90% of revenue covered).13141516 | 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 | A winner’s curse correction is not mentioned in their public documentation (as of November 11, 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 mode by default: probability to beat control, probability to be best, 95% intervals on the difference and the value. The choice of prior and expected loss are not documented.1718 | 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 | A multiple comparisons correction is not mentioned in their public documentation (as of November 11, 2026). | Segment comparisons are corrected with the Holm method. With several variants, the error risk is split across comparisons (Bonferroni). Where to check |
| Multivariate testing | Up to five groups per test (A/B/C/D/E). No native factorial multivariate test is documented: the FAQ suggests creating one group per combination; no interaction analysis is documented.1920 | 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) | Variance reduction (CUPED) is not mentioned in their public documentation (as of November 11, 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 | Visual Builder that embeds the store in the app (click an element, live preview); a Styles & JavaScript editor is available in every experiment.2122 | 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 | Ready-made components for offers (quantity buttons, progress bar, offer message, upsells) and a library of checkout blocks (badges, social proof, upsell; Shopify Plus).2324 | 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 | ESM script loaded through the App Embed with blocking="render" by default: the browser does not paint the page until it loads (slightly higher FCP/LCP). Anti-flicker hides only the modified elements, or the whole page for redirects. Size not published.2526 | 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 | Access follows Shopify’s app permission.27 Custom roles or granular permissions are not mentioned in their public documentation (as of November 11, 2026). | 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 | The privacy policy states a transfer outside Europe, to Canada and the United States; the script is served from AWS S3.2829 An EU hosting region is not mentioned in their public documentation (as of November 11, 2026). | 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 | Assignment happens on the first visit and is stored in a first-party cookie and in local storage.30 Consent management is not mentioned in their public documentation (as of November 11, 2026). | 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 | Public pricing per store based on monthly order volume and the modules chosen, with no per-visitor billing or overage. From $69/month (Smart Content); Unlimited from $349/month on the App Store; 3-month minimum for Smart Pricing and Unlimited; 20% off annually.313233 | 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 Intelligems also does.
Probant shares these points with Intelligems.
Revenue per visitor, with its range
Revenue per visitor, average order value, and other revenue metrics, each with significance and 95% intervals (credible intervals from the Bayesian model).1112
A visual editor
Visual Builder that embeds the store in the app (click an element, live preview); a Styles & JavaScript editor is available in every experiment.2122
Ready-made widgets
Ready-made components for offers (quantity buttons, progress bar, offer message, upsells) and a library of checkout blocks (badges, social proof, upsell; Shopify Plus).2324
A Bayesian reading
Bayesian mode by default: probability to beat control, probability to be best, 95% intervals on the difference and the value. The choice of prior and expected loss are not documented.1718
When to choose Intelligems over Probant.
Based on Intelligems’s public documentation.
You sell on Shopify and want to measure profit
Gross profit per visitor and per order, with intervals: COGS per variant (CSV or Shopify sync), average shipping cost per order, and a transaction fee percentage. COGS apply retroactively; missing COGS count as 0 (warning below 90% of revenue covered).13141516
You want AI to build the variants
AI Visual Builder (changes described in natural language), Atlas Experience Agent (sets up an experiment from a description), Atlas QA (automated QA with screenshots), an AI agent in Slack, and skills through the MCP server.343536
You want pricing based on order volume
Public pricing per store based on monthly order volume and the modules chosen, with no per-visitor billing or overage. From $69/month (Smart Content); Unlimited from $349/month on the App Store; 3-month minimum for Smart Pricing and Unlimited; 20% off annually.313233
What sets Probant apart.
These points are not in Intelligems’s public documentation, accessed November 11, 2026. A missing mention does not prove a feature is missing; if these points matter to you, ask Intelligems.
Winner’s curse correction
The reported gain is corrected by Bayesian shrinkage toward the effects of the store’s past tests.
IntelligemsA winner’s curse correction is not mentioned in their public documentation (as of November 11, 2026).
Learn moreSRM that blocks the verdict
A traffic split imbalance moves the test to “blocked” status, and no winner is declared.
IntelligemsNo native SRM check in the interface is documented. The vendor provides a guide to build a Slack alert yourself (n8n + API) that includes a chi-square test, and a second guide to pause failing tests automatically.7 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.
IntelligemsAssignment happens on the first visit and is stored in a first-party cookie and in local storage.30 Consent management is not mentioned in their public documentation (as of November 11, 2026).
Variance reduction (CUPED)
CUPED applies to test reports, using each visitor’s history before their first exposure.
IntelligemsVariance reduction (CUPED) is not mentioned in their public documentation (as of November 11, 2026).
Frequently asked questions.
How much does Intelligems cost?
Public pricing per store based on monthly order volume and the modules chosen, with no per-visitor billing or overage. From $69/month (Smart Content); Unlimited from $349/month on the App Store; 3-month minimum for Smart Pricing and Unlimited; 20% off annually.313233 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 Intelligems host data?
The privacy policy states a transfer outside Europe, to Canada and the United States; the script is served from AWS S3.2829 An EU hosting region is not mentioned in their public documentation (as of November 11, 2026). Probant hosts its database and application in Amsterdam, with Scaleway.
Does Intelligems measure margin?
Gross profit per visitor and per order, with intervals: COGS per variant (CSV or Shopify sync), average shipping cost per order, and a transaction fee percentage. COGS apply retroactively; missing COGS count as 0 (warning below 90% of revenue covered).13141516 Probant measures revenue and margin per visitor, each with its range, from orders and the cost of each product.
Can you look at Intelligems results while a test is running?
No sequential test or always-valid interval is documented. The documentation recommends waiting for minimums (300 orders per group, 7 days); an automatic verdict through the API (opt-in) adds a stability criterion over the last few days.56 Probant sets its checkpoints before launch (O’Brien-Fleming) and uses always-valid ranges during rollouts.
Sources.
Official pages from Intelligems, accessed November 11, 2026. Quotes are reproduced in their original language.
- Source 1: www.intelligems.io/llms.txt
Intelligems is the profit optimization and A/B testing platform built specifically for Shopify.
- Source 3: www.intelligems.io/resources/blog/intelligems-raises-9m-to-fuel-the-profit-growth-platform-for-ecommerce-brands
We’re thrilled to announce that Intelligems has raised $9 million in our Series A round, led by Stage 2 Capital.
- Source 4: docs.intelligems.io/analytics/overview/experiment-analytics/statistical-significance
Intelligems uses a Bayesian statistical model and Monte Carlo simulations to analyze A/B tests. Note that the model does not account for intra-week (daily) seasonality, or other store-specific factors.
- Source 5: docs.intelligems.io/analytics/overview/experiment-analytics/statistical-significance
We recommend that in addition to using Intelligems' provided probabilities to determine the significance of your test, you also ensure the test reaches a pre-defined minimum number visitors, orders, and full weeks live.
- Source 6: docs.intelligems.io/analytics/overview/experiment-analytics/statistical-significance
In addition to the raw probabilities and confidence intervals above, Intelligems can return an automated test verdict — a single, actionable read on a test that combines statistical significance with stability (how settled the result has been over recent days).
- Source 7: docs.intelligems.io/developer-resources/external-api/automations-and-guides/automated-test-alerting-for-slack
This guide walks you through creating a daily automation that checks running tests for traffic or data issues (low visitors, SRM failures, large conversion drops, minimum orders per variant) and posts a Slack alert when a test fails health thresholds.
- Source 8: docs.intelligems.io/developer-resources/external-api/automations-and-guides/how-to-automatically-pause-failing-tests
This guide explains how to extend your monitoring workflow to automatically pause an experiment if it fails your health thresholds.
- Source 9: docs.intelligems.io/general-features/rolling-out-tests
Rolling out a winning test variant ends the test and creates a Personalization with the same site modifications as the variant.
- Source 10: docs.intelligems.io/developer-resources/external-api/automations-and-guides/build-a-multi-armed-bandit-for-dynamic-traffic-optimization
Learn how to implement a Multi-Armed Bandit (MAB) workflow using the Intelligems API to automate traffic allocation, minimize opportunity cost, and maximize revenue during live experiments.
- Source 11: docs.intelligems.io/analytics/overview/experiment-analytics/metric-definitions/revenue
Revenue per Visitor: Net revenue divided by number of unique visitors.
- Source 12: docs.intelligems.io/analytics/overview/experiment-analytics/statistical-significance
Value confidence interval: this is a 95% confidence interval around the value measurement for a specific metric.
- Source 13: docs.intelligems.io/analytics/overview/experiment-analytics/metric-definitions/profit
Intelligems can report down to gross profit. We recommend using gross profit per visitor to analyze most tests, especially those that may affect price, product mix, order volume, or shipping rates.
- Source 14: docs.intelligems.io/analytics/overview/experiment-analytics/metric-definitions/profit
You can provide a CSV with COGS for each variant, or we can sync the information from Shopify if it’s available there.
- Source 15: docs.intelligems.io/analytics/overview/experiment-analytics/metric-definitions/profit
If you leave out COGS information for some products, their COGS will be treated as 0, which inflates profit numbers (these products are treated as having a 100% gross margin).
- Source 16: docs.intelligems.io/analytics/overview/how-to-add-profit-to-intelligems-analytics
Updating cost assumptions, including COGS per product, will apply retroactively to existing experiments, so you can do this even after an experiment has started collecting data.
- Source 17: docs.intelligems.io/analytics/overview/experiment-analytics/statistical-significance
Probability to beat control: this is the probability that a test group is better than the control group, on the basis of the selected metric
- Source 18: docs.intelligems.io/analytics/overview/experiment-analytics/statistical-significance
Probability to be best: this is the probability that a test group is best, on the basis of the selected metric.
- Source 19: docs.intelligems.io/getting-started/general-faqs
Set this up as one large experience. Make a group for each combination of variables so that you can properly measure and compare (a multivariate experiences).
- Source 20: www.intelligems.io/pricing
multi-variant testing (up to five groups per test, so A/B/C/D/E, not just A/B)
- Source 21: docs.intelligems.io/general-features/editors/visual-builder
It embeds your store directly within Intelligems, so you can click any element to edit it and see changes live, without leaving the platform.
- Source 22: docs.intelligems.io/content-experiences/content-experiences-overview/how-to-set-up-a-styles-and-javascript-experience
A Styles & JavaScript experience injects your own CSS and JavaScript for a chosen group of visitors, so you can find out whether a change lifts conversion before you commit it to your theme.
- Source 23: docs.intelligems.io/merchandising-experiences/offers-integrating-widgets
When setting up a test or personalization, you can add onsite components to merchandise your offers directly on your site to ensure your customers are aware of the offers they are eligible for, improving conversion and driving additional AOV.
- Source 24: docs.intelligems.io/checkout-experiences/getting-started-with-checkout/checkout-experiences-library
Learn how to build 6 Checkout Experiences in your store
- Source 25: docs.intelligems.io/performance-optimization/site-performance
It stops the browser from painting the page until the Intelligems script has loaded, which prevents flicker on your tests at the cost of slightly higher FCP and LCP times.
- Source 26: docs.intelligems.io/performance-optimization/site-performance/anti-flicker-modes
Price, content, and image tests: Intelligems temporarily hides only the elements your experience changes, then shows them once the right version is applied.
- Source 27: docs.intelligems.io/getting-started/general-faqs
Access to your Intelligems account (sometimes called your Intelligems portal or dashboard) is managed through Shopify, so you won't find a place to add users inside the Intelligems app.
- Source 28: www.intelligems.io/privacy-policy
Additionally, please note that your information will be transferred outside of Europe, including to Canada and the United States.
- Source 29: docs.intelligems.io/developer-resources/incident-report-12-28-2025-cdn-outage
AWS rotated SSL certificates on the S3 service backing our CDN.
- Source 30: docs.intelligems.io/getting-started/general-faqs
Intelligems uses sticky bucketing — each visitor is randomly assigned to a test group on their first visit, and that assignment is persisted in a first-party cookie and `localStorage`.
- Source 31: www.intelligems.io/pricing
You pay a flat monthly fee for your volume tier, with no per-visitor charges and no overages. Build Your Own plans start at $69/month at the lowest order volume tier.
- Source 32: www.intelligems.io/pricing
So Smart Pricing (Limited and Unlimited) and the Unlimited plan carry a 3-month minimum. All other modules are month-to-month.
- Source 33: apps.shopify.com/intelligems
Unlimited $349 / month or $3,468/year and save 17% Pricing scales based on sitewide order volume
- Source 34: docs.intelligems.io/general-features/editors/ai-visual-builder
The AI Builder lets you describe changes to your storefront and have them applied automatically.
- Source 35: docs.intelligems.io/general-features/atlas-ai/atlas-experience-agent
The Atlas Experience Agent configures an experience from a written description.
- Source 36: docs.intelligems.io/general-features/atlas-ai/atlas-qa
Atlas QA is Intelligems' quality assurance (QA) agent.
Migrate from Intelligems
Your Intelligems history can be imported as a generic CSV (visitors and conversions per variant) 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