Splitlab home

Book a Demo

Splitlab home

Book a Demo

Splitlab home
ClickFunnels sends orders and upsells to Splitlab by webhook, each order is matched to its variant, and Variant B wins on revenue per visitor, $3.34 vs $2.91

November 3, 2026

By

Bishal Katuwal

·

6 min

read

How to A/B Test ClickFunnels 2.0 Pages (and Why the Built-In Split Test Isn’t Enough)

November 3, 2026

By

Bishal Katuwal

·

6 min

read

How to A/B Test ClickFunnels 2.0 Pages (and Why the Built-In Split Test Isn’t Enough)

In the ever-evolving business environment, achieving a competitive edge demands innovative solutions beyond conventional methods. Organizations require smart systems that can optimize workflows, enhance customer interactions, and reveal new avenues for growth. Enter Biotix, the cutting-edge AI-powered CRM management tool.

The native ClickFunnels 2.0 split test is fine for a quick headline check. The moment you test the offer, the price or the upsell, you need revenue across every step and real statistics.

Bishal Katuwal

Bishal Katuwal

Splitlab Founder

ClickFunnels sends orders and upsells to Splitlab by webhook, each order is matched to its variant, and Variant B wins on revenue per visitor, $3.34 vs $2.91
ClickFunnels sends orders and upsells to Splitlab by webhook, each order is matched to its variant, and Variant B wins on revenue per visitor, $3.34 vs $2.91

Never run a split test on your ClickFunnels 2.0 funnel? Start today with the built-in feature. Duplicate a step, change the headline, split the traffic, watch the numbers.

This post is for what comes next. Once you test the offer, the price, the order bump or the upsell, clicks stop mattering. Money does. That’s where the native split test runs out of road.

Key takeaways

  • ClickFunnels 2.0’s native split test compares two versions of one step on that step’s conversion rate. Good for headlines, blind to everything downstream.

  • Any test that touches the offer needs revenue per visitor measured across the whole funnel.

  • With Splitlab you paste one snippet, connect ClickFunnels 2.0 by webhook, and every order is credited to the variant the visitor saw.

  • Set a stopping rule before you launch: probability to be best plus an expected-loss ceiling.

What the native ClickFunnels 2.0 split test does well

At the time of writing, the native split test lets you:

  • Create a variant of a single funnel step

  • Choose the traffic split

  • Compare each version’s conversion rate side by side

No code, no extra tool, no extra bill. For a headline or hero image test on a busy opt-in page, that’s honestly enough.

Where it runs out

Three limits show up the moment a test touches money.

  • It sees one step. You see checkout rate, but not whether buyers took the bump, bought the upsell or refunded a week later.

  • It’s A versus B. Two versions per step. Testing three prices or five headlines means months of back-to-back tests.

  • It doesn’t tell you when to stop. You get two percentages. No probability, no interval, no estimate of what a wrong pick costs.

A cheaper offer can raise checkout rate and still lose money. Revenue per visitor vs conversion rate goes deep on why.

How to run a revenue-based A/B test on ClickFunnels 2.0

Here’s the setup with Splitlab, step by step. The same principle works with any tool that assigns variants in the browser and tracks revenue by webhook.

1. Paste the snippet once

Add the Splitlab snippet to your funnel’s head tracking code. You do it once, not per page.

  • In ClickFunnels, open your funnel settings

  • Go to Settings > Tracking Code

  • Paste the snippet into Head Tracking Code and save

Splitlab Websites page, ClickFunnels tab: five steps to install the SmartCode (log in and open funnel settings, Settings → Tracking Code, paste into Head Tracking Code, save, verify) with the Include anti-flicker script toggle on

Splitlab shows the exact install steps for ClickFunnels. Keep the anti-flicker option on.

The snippet assigns each visitor a variant, renders it before the page paints, and remembers the assignment across every step.

2. Connect ClickFunnels 2.0

Open Integrations in Splitlab and connect ClickFunnels 2.0 with your workspace API key. Every order, subscription and opt-in then arrives by webhook with the buyer attached.

Splitlab Integrations page, Funnels & Revenue section: ClickFunnels Classic (tracks revenue per website) and ClickFunnels 2.0, marked Connected (tracks revenue across all funnels)

Revenue flows in by webhook, not by pixel.

Full steps live on the ClickFunnels 2.0 integration page. Still on Classic? Use the ClickFunnels Classic integration.

3. Point the test at your funnel step

Create a test and paste the URL of the step you want to change.

Splitlab test setup on the Pages step: Page Selection set to URL matches https://yourfunnel.myclickfunnels.com/sales-page

Pages is the first step of every test: tell Splitlab which URL to run on.

Then add your variations and build each one in the visual editor. You can:

  • Rewrite copy and swap images

  • Move, hide or add sections

  • Drop in a block of HTML

  • Run more than two variants at once

Splitlab Variants & Traffic step with Traffic Split on Custom Distribution: Control 60%, Variant 1 0%, Variant 2 40%

Each variation gets its own editor and its own share of traffic.

4. Pick metrics that measure the funnel

Star revenue per visitor as the primary metric. Then add guardrails:

  • Conversion rate catches anything that breaks

  • Refund rate matters if the test changes the offer

Splitlab Metrics step: Revenue per Visitor (RPV) starred as Primary, with Average Order Value (AOV) and Conversion Rate (CVR) as standard metrics and a custom Order Page Visit metric

RPV, AOV and CVR come built in. The starred metric decides the test.

Because revenue is tied to the visitor, a purchase on step four is credited to the variant seen on step one. That’s the whole point of the setup.

5. Choose how traffic gets split

You have three options:

  • Equal Distribution for a clean test you want to learn from

  • Custom Distribution to send less traffic to a risky variant

  • Multi Arm Bandit to shift traffic toward the higher-revenue variant as evidence comes in

Splitlab Traffic Split menu open, enlarged: Equal Distribution (an even split and a clean read), Custom Distribution (you set each share) and Multi Arm Bandit (shifts traffic to the higher-revenue variant)

Pick the split per test. Use the bandit when earning during the test matters more than a clean read.

Multi-armed bandit vs A/B testing covers when each one makes sense.

What to test on each ClickFunnels step

Opt-in page

  • Headline and lead magnet promise

  • Form length, and whether to ask for a phone number

Primary: lead rate. Also watch downstream purchases. More emails from people who never buy isn’t a win.

Sales page or VSL

  • Offer framing and price presentation

  • Guarantee

  • Buy button position, long form versus short

Primary: revenue per visitor.

Order form

  • One-step versus two-step

  • Bump copy, price, pre-checked or not

  • Trust badges and payment plans

Primary: revenue per visitor across the rest of the funnel, because this step changes who reaches the upsell.

One-time offer

  • Price anchoring, video versus text

  • One-click accept versus a second form

  • Countdown timer and the “no thanks” copy

Primary: revenue per funnel visitor. Testing order bumps, upsells and downsells is the full playbook.

Thank-you page

  • Cross-sell, community invite, book a call

Usually forgotten, usually cheap to test.

Attribution details that quietly wreck ClickFunnels tests

Four things to get right before you trust a result:

  • Sticky assignment. A visitor who sees B on the sales page and control on the order form saw neither. If checkout is on another domain, the visitor id has to travel too.

  • One conversion event, fired once. A pixel that double-fires on one variant is the most common source of a fake winner. Webhook revenue avoids it.

  • Refunds count. If refunds aren’t deducted, a variant that oversells and refunds will look like a winner.

  • One testing system per step. Running the native split test and an external test together gives you four combinations and no way to read them.

Reading the result

Look for three numbers per variant.

  • Probability of being best. “Variant B has a 94% probability of higher revenue per visitor.”

  • Expected loss. “If we pick B and we’re wrong, we lose about $0.03 per visitor.” This makes a result safe to act on below 99%.

  • Lift range. “B is probably between 2% and 19% better.” Wide range means keep running.

Write your decision rule down before launch. For example: ship at 95% probability to be best with expected loss under a figure you can live with. Then don’t move the goalposts.

Bayesian A/B testing for marketers explains where these numbers come from.

Frequently asked questions

Can I A/B test ClickFunnels 2.0 without code?

Yes. The native split test needs no code. Splitlab needs one snippet pasted into the head tracking code once. After that, every variant is built in a visual editor.

Does an external testing tool slow down my ClickFunnels pages?

Any script adds some weight. What matters is CDN delivery, no flash of the original content, and no render blocking. Splitlab renders the variant before the page paints instead of swapping content afterwards.

Can I test more than two variants on a ClickFunnels page?

At the time of writing, not with the native split test. Splitlab runs as many variants as your traffic supports, plus multivariate tests that combine changes across several elements.

How does revenue from the upsell get credited to a test on the sales page?

The snippet stores the visitor’s variant. When they buy anything later in the funnel, ClickFunnels sends the order to Splitlab by webhook. Splitlab matches the buyer to the visitor and credits the variant they saw.

Do I need ClickFunnels 2.0, or does Classic work too?

Both work. Splitlab has separate integrations for ClickFunnels 2.0 and ClickFunnels Classic.

My final thoughts

The built-in ClickFunnels split test is a good place to start, and I’d use it for a quick headline check any day. But a funnel makes most of its money after the first step, in the bump, the upsell and the downsell, and that’s where the native test can’t see.

So test the step, but judge the whole funnel. Credit every order and upsell to the version the buyer saw, and pick the winner by revenue per visitor. Get that right and you stop shipping winners that only win on the first page.

Test the step. Judge the funnel.

Test the whole funnel, not one step. Connect ClickFunnels 2.0 to Splitlab and find out which variant actually made more money. Start testing or book a demo.

Built for revenue

Test for revenue, not just clicks.

Run A/B, multivariate, and multi-armed bandit experiments — with real revenue tied to every variation.

We’ll increase your conversion rate by 30% or you don’t pay. Terms apply.

✓

Revenue per variation

✓

A/B & MVT

✓

Multi-armed bandit

Test for revenue, not just clicks.

A/B, multivariate and multi-armed bandit experiments, with real revenue tied to every variation.


Start optimizing for revenue today not just conversion rates.

Run experiments, track revenue per variation, and scale what actually makes you money.

We’ll increase your conversion rate by 30% or you don’t pay. Terms apply.


Start optimizing for revenue today not just conversion rates.

Run experiments, track revenue per variation, and scale what actually makes you money.

We’ll increase your conversion rate by 30% or you don’t pay. Terms apply.


Start optimizing for revenue today not just conversion rates.

Run experiments, track revenue per variation, and scale what actually makes you money.

We’ll increase your conversion rate by 30% or you don’t pay. Terms apply.