Do I have to redeploy my store to test a redesign?

No. The challenger theme is served through CustomFit to its cohort without replacing your published theme. If it loses, nothing was ever swapped; if it wins, you promote it with confidence.

What metrics decide the winner?

Theme tests are judged on revenue per visitor and AOV, not just click-through — because a prettier theme that lowers AOV is a loss. Guardrails on bounce and page speed keep the comparison honest.

Theme A/B Testing — Test Whole Shopify Themes for D2C — CustomFit.ai

Theme A/B testing — prove the redesign before you ship.

A new theme is the biggest bet your store makes all year — and most brands ship it on a gut feeling. CustomFit runs your challenger theme live against the current one, on real traffic, and tells you what it does to revenue before you commit. AI calls the winning theme on revenue, not guesswork.

✓ No redeploy✓ Revenue-per-visitor verdict✓ Instant rollback

🎭

Whole theme, not one block

Serve an entire alternate theme — layout, navigation, PDP, cart — to a cohort and compare the full experience, not a single element.

💰

Judged on revenue

Winners are decided on revenue per visitor and AOV — a prettier theme that quietly drops basket size is a loss, and we'll catch it.

🔁

Zero-risk rollout

Nothing is republished until you promote the winner. If the challenger loses, your live theme was never touched.

Why theme tests matter

The "we just relaunched" gamble, de-risked.

Most replatforms and redesigns go live everywhere at once, and teams cross their fingers. With theme testing you expose the new design to a slice of traffic, watch real purchase behavior, and only roll forward when the numbers say so.

✓ Run challenger vs. current on simultaneous live traffic
✓ Control the cohort size — start at 10%, scale as it proves out
✓ Roll back instantly if engagement dips

See the stats engine →

Rollout plan

Week 1Challenger at 10% — winning+9%

Week 2Scaled to 35% — holding+11%

Week 3Promote to 100%+12.6%

Seasonal theme swap

🎉 Festive theme — Diwali palette, gifting nav, bundle-first PDP

☀️ Summer theme — lighter hero, SPF cross-sell, cooling copy

🗓 Scheduled — auto-swap on/off by date, no late-night deploy

🎭 Schedule a seasonal theme, A/B it first, and revert automatically when the sale ends.

Seasonal & campaign themes

Dress the whole store for the moment.

Run a festive theme for Diwali, a minimalist theme for a launch, or a campaign theme for a big sale — test it against your evergreen design, schedule it to flip on and off by date, and never touch code on a Friday night again.

✓Schedule themes by date & campaign
✓ A/B the seasonal theme before going wide
✓ Auto-revert when the window closes

How a winner is decided

A prettier theme isn't a winning theme

Theme tests are judged on the metrics that actually matter to the business — with guardrails that catch a redesign that looks great but sells worse.

💰

Revenue / visitor

The headline metric — the true measure of a theme.

🛒

Average order value

Catches designs that convert more but sell cheaper.

↩️

Bounce rate

Guardrail — a flashy theme that scares people off loses.

⚡

Page speed

Guardrail — Core Web Vitals must stay green.

The fundamentals

What is theme A/B testing?

Theme A/B testing serves an entire alternate theme or template to a slice of your traffic and compares it against your current design on live shoppers — judged on revenue per visitor and AOV, not just clicks. It lets you prove a redesign lifts revenue before you roll it out to everyone.

Quick facts

✓Runs a whole challenger theme against the current one

✓Decided on revenue per visitor and AOV

✓No redeploy — nothing is published until you promote

✓Schedule seasonal themes and auto-revert by date

Why it matters

Why theme A/B testing matters for D2C

🎭

De-risk redesigns

Don’t bet the store on a gut-feel relaunch — prove it on real traffic first.

💰

Judge on revenue

A prettier theme that lowers AOV is a loss; theme tests catch it.

🔁

Roll back instantly

If the challenger dips, your live theme was never touched.

Who it’s for

Built for your whole growth team

🎨

Brands redesigning

Validate a new look before going wide.

🛒

Seasonal sellers

A/B festive and campaign themes, then auto-revert.

🧑‍💻

Dev & design teams

Pilot replatforms on a cohort with data in hand.

The complete guide

Understanding theme A/B testing

Theme A/B testing serves an entire alternate theme or template to a slice of your traffic and compares it against the current design on live shoppers — judged on revenue per visitor and AOV, not just clicks. It lets you prove a redesign lifts revenue before rolling it out to everyone.

A new theme is the biggest bet a store makes all year, and most brands ship it on a gut feeling. With theme testing you expose the challenger to a controlled cohort, watch real purchase behavior, and scale only when the numbers say so — with instant rollback if engagement dips.

The same mechanism powers seasonal and campaign themes: run a festive theme for a sale, A/B it against your evergreen design, and schedule it to flip on and off by date with automatic revert. No redeploys, no late-night theme surgery.

Do I need to redeploy to test a theme?

No. The challenger theme is served through CustomFit to its cohort without replacing your published theme, so there is nothing to roll back if it loses.

What metric decides the winning theme?

Revenue per visitor and AOV, not just click-through — with bounce and page speed as guardrails — because a prettier theme that lowers basket size is a loss.

Can I test two completely different themes?

Yes. CustomFit serves an entire alternate theme or template set to a randomized cohort while everyone else sees the current one — both live at once, compared on the same traffic and seasonality.

Will a second theme slow my store down?

No. Themes resolve at the edge before first paint, so the cohort gets the challenger with no flicker and no Core Web Vitals hit.

Do I need a developer to use CustomFit.ai?

No. Marketers build experiments and personalized experiences in a no-code visual editor; developers can use the API and SDKs when they want deeper control.

Live · Right now

Mamaearth — free-shipping band +12.4% AOVGIVA — festive collection page +34% revenueBellavita — PDP CTA test +27.4% CVRKapiva — Quiz-driven recs +9.48% CTRThe Sleep Co — landing personalized 2× capturesPlum — Returning shopper swap +18.2% CVRMamaearth — free-shipping band +12.4% AOVGIVA — festive collection page +34% revenueBellavita — PDP CTA test +27.4% CVRKapiva — Quiz-driven recs +9.48% CTRThe Sleep Co — landing personalized 2× capturesPlum — Returning shopper swap +18.2% CVR

Theme A/B testing — prove the redesign before you ship.