Collection Page Iteration: $42K/Month Additional Revenue Case Study

Winning tests aren't endpoints. They're starting points.

We ran a follow-up test on the collection page for an 8-figure slime brand. The previous test added a hero banner, reviews, and labels. It drove +$114K/mo.

This test pushed further. Horizontal filter pills for faster navigation. Carousel images. "NOTIFY ME" buttons on sold-out items.

Results: +$42,308/mo additional revenue.

That's on top of the previous win. Compound gains.

The Problem With Stopping at One Win

The previous collection page test was a big win. It would be easy to call it done and move on.

But the page still had friction:

• Filters required opening a modal. Extra tap, extra friction.
• One product image meant a click to see more of the product.
• Sold-out products were dead ends. No way to capture interest.

A winning test tells you what works. It doesn't tell you what else could work better.

The Hypothesis

Enhancing filters/sorting, clearer navigation, and back-in-stock sign-up on sold-out items will reduce friction, improve product discovery, and capture demand, increasing add-to-carts and SMS sign-ups.

Three improvements to test:

1. Horizontal filter pills for one-tap category access
2. Carousel format on the product images
3. "NOTIFY ME" on sold-out items to capture demand

Test Setup

Page: Collection Page
Location: The Whole Page
Platform: Intelligems
Test Type: A/B/n with 5 variations

Control

Previous winning version:

• "BEST SELLERS" hero banner with lifestyle imagery
• "LIMITED STOCK" labels on applicable products
• Star ratings with "1M+ SLIMES SOLD WORLDWIDE" social proof
  
• Filter button (opens modal)
• Product cards with slime type, scent, price

Variation 1 (Winner)

Same hero banner. Added improvements:

• Horizontal scrolling filter pills below banner (FILTER, BINGSU, CLEAR, etc.)
  
• Same labels and ratings
• "QUICK ADD" buttons visible on cards

Faster filtering. More products visible. Same trust elements.

Variation 2

Same as Variation 1. Filter pills and carousel style product images.

Variation 3

Same layout as Variations 1/2. Added "NOTIFY ME" button on sold-out products instead of dead-end "SOLD OUT" label.

Captures demand for out-of-stock items.

Variation 4

Same layout as Variations 1/2/3, but different approach - removed hero banner.

Prioritized products over banner.

Results

Winner: Variation 1 (Filter Pills + Two-Column Grid)

Metric	Improvement
Monthly Revenue	+$42,308

The filter pills drove the win. "NOTIFY ME" (Variation 3) and the no-banner approach (Variation 4) didn't outperform.

Why It Worked

1. Filter pills reduced friction to one tap

The control required: tap "FILTER" → wait for modal → select category → close modal → see results.

The variation: tap "BINGSU" → see results.

One tap vs. multiple taps. For a brand with distinct slime categories (Bingsu, Clear, Cloud, etc.), letting customers jump directly to their preferred type removes unnecessary steps.

2. The hero banner still mattered

Variation 4 removed the hero banner to prioritize products. It didn't win.

The banner earned its place in the previous test. It sets context, creates visual hierarchy, and establishes the "BEST SELLERS" frame. Removing it hurt more than it helped.

Lesson: don't undo what's working to make room for new ideas. Add to winners, don't subtract from them.

3. "NOTIFY ME" didn't move the needle (in this test)

Variation 3 added back-in-stock notifications for sold-out items. It didn't beat Variation 1.

That doesn't mean "NOTIFY ME" is bad. It captures demand and builds an SMS list. But in a head-to-head revenue test, the navigation improvements drove more immediate impact.

"NOTIFY ME" might show value in longer-term metrics like repeat purchase rate or customer lifetime value. It's worth implementing even if it didn't win this specific test.

What This Means for Iteration

One winning test isn't the end. It's the new baseline.

After every win, ask:

• What friction still exists?
• What elements could we add?
• What's working that we can amplify?
• What's the next bottleneck?

This brand now has a collection page that's been optimized twice. Hero banner, reviews, stock labels, filter pills, two-column grid. Each layer added revenue.

The next test might add something else. The compounding continues.

FAQ

How do you decide when to iterate vs. move to a different page?

Diminishing returns. When tests stop producing significant lifts on a page, move on to the next high-traffic area.

This test added $42K/mo on top of $114K/mo. Still very much worth it. When iterations start producing single-digit percentage lifts, consider shifting focus.

Should filter pills replace traditional filters?

Not replace. Supplement.

Filter pills are great for primary categories (slime types in this case). Keep traditional filters for secondary attributes (price range, scent, etc.) that don't need top-level visibility.

This test was run using Intelligems as part of a CONVERTIBLES personalization program. Want to see what iterative optimization could do for your store? Book a call to get 3 personalized recommendations for your store.