Everything you need,
in one place.
From your first install to declaring your first winner — step-by-step guides for running A/B tests that move the needle, on any website.
Getting Started
5 guides-
What Is Split Test Pro?
An overview of Split Test Pro — a multiplatform A/B testing tool that runs experiments on your live site without touching your codebase.
-
Installing on an HTML Site
Add Split Test Pro to any website with a single script tag. Sign up, copy the snippet, paste it into your <head>, and verify it's running — under five minutes.
-
Installing on Shopify
Install Split Test Pro from the Shopify App Store. The Web Pixel and Theme App Extension auto-register so the full purchase funnel — add-to-cart, checkout, revenue — is tracked from the moment you go live.
-
A Tour of the Dashboard
Get oriented in Split Test Pro before you create your first experiment. Learn what each section does, where to find your workspace settings, and what the home dashboard tells you at a glance.
-
Your First Experiment
An end-to-end walkthrough of creating, configuring, and launching your first A/B test in Split Test Pro — from naming the experiment to interpreting the launch readiness check.
Running Experiments
4 guides-
Experiment Lifecycle
Understand the five states an experiment passes through — created, started, paused, completed, archived — and what each state means for your traffic and your data.
-
Traffic Allocation
How variant weights work in Split Test Pro — setting splits, when to use uneven allocations, and the practical guidance most teams need to ignore.
-
Pre-Launch QA
What to check before you start an experiment — the manual reviews that actually matter, what the launch checklist surfaces, and when to use the AI Review.
-
Running Multiple Experiments
Run several experiments at once — how visitor assignment works across concurrent tests, how the conflict warning helps, and the discipline that keeps your data clean.
Variants
5 guides-
Variant Types Overview
A quick tour of what each variant in Split Test Pro can change — CSS for visual tweaks, JavaScript for dynamic logic, and full-page redirects — and when to reach for each.
-
CSS Variants
How CSS injection works in Split Test Pro, how to find stable selectors, and the patterns you'll actually reach for — buttons, typography, hide/show, image swap, and the specificity escape hatches when your theme fights back.
-
JavaScript Variants
When to use JavaScript variants instead of CSS, how they're injected, the safety considerations that matter, and worked examples for the tests CSS can't express.
-
Redirect Tests
Compare two completely different pages by sending each variant to a different URL. When redirects are the right pattern, how Split Test Pro implements them, and the SEO trap to avoid.
-
Screenshots and Preview
Use Split Test Pro's auto-captured variant screenshots and Preview URL banner to confirm your variant looks right and your targeting is wired up before launch.
Targeting
3 guides-
URL Targeting
Match the right pages with Split Test Pro's URL targeting — the four URL parts you can target, the operators available, and worked examples for single pages, page patterns, and exclusions.
-
Device Targeting
Restrict an experiment to mobile, tablet, or desktop visitors. How Split Test Pro detects device type, when device targeting beats a media query, and how to read device-segmented results.
-
Event Activation
Defer an experiment so it doesn't activate until a specific event fires — useful for tests that only matter after a user interaction. Set it up in the Targeting tab and trigger via SplitTestPro.activate() or a DOM event.
Tracking & Goals
4 guides-
Conversion Goals
Define what counts as a "conversion" for your experiment. How built-in goals work on Shopify, how to wire up custom goals on HTML, and how to pick the right primary metric for your test.
-
Custom Events
Fire conversion events from your site with the Split Test Pro JavaScript API. Pass numeric values for revenue, mark recurring purchases, and verify events arrive correctly.
-
Continuous Metrics
When to track a numeric value alongside a conversion event — revenue, AOV, time spent — and how Split Test Pro switches from binomial to Normal-distribution modeling under the hood.
-
Shopify Funnel Tracking
How Split Test Pro's Shopify Web Pixel captures the full purchase funnel automatically — every event from product view to completed checkout — and how to read funnel breakdowns in your results.
Understanding Results
6 guides-
Reading the Results Dashboard
A walkthrough of every panel on the experiment Results page — verdict card, summary stats, variant comparison, device segments, revenue, custom metrics, and the statistical analysis accordion.
-
Bayesian Stats Explained
How Split Test Pro's Bayesian engine computes "probability to be best" and credible intervals — the priors, the Monte Carlo sampling, and why this beats traditional p-values for product decisions.
-
Segmenting Results
Filter your experiment results by device type to see how variants performed across desktop, tablet, and mobile — and what to do when the segments tell different stories than the aggregate.
-
AI Review
Get an AI-generated read on an experiment — what it checks before launch, what it summarizes after, when it's worth running, and how the cache works.
-
Anomaly Alerts
The four rule-based checks Split Test Pro runs against every active experiment — sample ratio mismatch, traffic drops, zero conversions, runaway long tests — plus the optional Claude-driven anomaly review.
-
Declaring a Winner
When to call an experiment done — the 95% probability threshold, the minimum-runtime and minimum-sample rules, and how to apply the winning variant permanently to your site after.
Integrations
3 guides-
Google Analytics 4 Integration
Connect Split Test Pro to GA4 via OAuth. Variant data flows into your GA4 property as custom events, and each variant gets its own GA4 audience for cross-segmentation.
-
Shopify Web Pixel
How the Web Pixel installed by Split Test Pro tracks the full purchase funnel automatically, what events it captures, and what to check when conversions aren't being attributed correctly.
-
Shopify Theme App Extension
How Split Test Pro's Theme App Extension injects variant CSS and JS into your storefront — what it adds to the page, how to verify it's active, and how it survives theme changes.
Account & Workspace
3 guides-
Workspaces
A workspace is the container for one site or store's experiments, team, and billing. Learn how to create and switch between them, what's editable, and what happens when you delete one.
-
Team and Invites
Invite teammates to a workspace, assign owner or member roles, accept invites, and remove members. The full team-management flow.
-
Billing and Plans
How subscriptions work for HTML and Shopify workspaces — Stripe checkout, the Shopify App Billing flow, plan changes, and cancellation.
Best Practices
3 guides-
A/B Testing Methodology
The principles that separate teams who get compounding gains from A/B testing from those who spin their wheels — hypothesis-driven testing, single-variable rigor, sequential discipline, and documenting losses.
-
Common Mistakes
The most frequent ways A/B tests produce misleading results — peeking, stopping early, multi-variable variants, mid-experiment edits, ignoring segments — and how to avoid each.
-
Building a Testing Program
Move from ad-hoc tests to a sustainable experimentation practice — backlog discipline, ICE prioritization, weekly cadences, quarterly reviews, and the compounding effect of a few well-applied winners.
Reference
6 guides-
Selector Cookbook
Patterns for finding stable CSS selectors, the DevTools workflow that makes it fast, and the specific anti-patterns that produce fragile experiments.
-
Troubleshooting
Diagnose and fix the most common Split Test Pro issues — variants not showing, conversions not tracked, results not updating, and platform-specific gotchas for both HTML and Shopify.
-
Glossary
A quick-reference list of A/B testing terms used throughout Split Test Pro and its documentation — variant, exposure, lift, credible interval, probability to be best, and more.
-
Privacy and Consent
How Split Test Pro handles privacy on HTML and Shopify — Do Not Track and Global Privacy Control respect, opt-in/opt-out APIs, the cookie-consent banner pattern, and what data is collected.
-
Single-Page Apps
Split Test Pro evaluates targeting on initial page load only — SPA route changes via pushState don't retrigger experiments. Learn the limitation and the workaround.
-
Frequently Asked Questions
Quick answers to the questions merchants and developers ask most often about Split Test Pro — flicker, performance, multiple experiments, privacy, and what's not supported today.
