Collection Page Filter Strategy: $39K/Month Case Study

Not all filters are created equal. The framing matters.

We tested filter pill options on the collection page for an 8-figure dog treats brand. Same concept (horizontal filter pills for faster navigation), different filter categories.

We tested four approaches:

• Need-based: Powerful Chewers, Picky Eaters, Sensitive Stomachs
• Dietary concerns: Single-Ingredient, Poultry-Free, Beef-Only
• Product type: Bones, Bully Sticks, Yak Chews, Jerky
• Dog size: Puppy, Small Dog, Medium Dog, Large Dog

The winner? Dog size.

Results: +$39,779/mo revenue.

The Problem With Default Filter Categories

The control page had a filter button and a sort dropdown. Standard stuff.

But filters hidden behind a button require extra effort. Tap to open, scan options, select, close. For customers who just want to find treats for their small dog, that's too many steps.

Filter pills solve the visibility problem. But which filters you surface changes everything.

The Hypothesis

Introducing clear filters (e.g., size, chew type, allergy-friendly, plant-based) will reduce decision friction, increase product discovery relevance, and improve add-to-cart and conversion rates, especially for first-time buyers with safety or dietary concerns.

The question wasn't whether to add filter pills. It was which filter dimension would drive the most conversions.

Test Setup

Page: Collection Page
Location: Above the fold
Platform: Intelligems
Test Type: A/B/n with 5 variations

Control

"All Products" header. Filter button (opens modal). "Featured" sort dropdown. Two-column product grid with images, names, star ratings, prices, and "Quick Shop" buttons.

No filter pills visible.

Variation 1: Need-Based Filters

Filter pills:

• Powerful Chewers
• Picky Eaters
• Sensitive Stomachs

Same filter approach that won on the homepage. Would it work on the collection page?

Variation 2: Dietary Concern Filters

Filter pills:

• Single-Ingredient
• Poultry-Free
• Beef-Only

Targeting customers with allergy concerns or ingredient preferences.

Variation 3: Product Type Filters

Filter pills:

• Bones
• Bully Sticks
• Yak Chews
• Jerky

Traditional category-based filtering by product format.

Variation 4: Dog Size Filters (Winner)

Filter pills:

• Puppy
• Small Dog
• Medium Dog
• Large Dog

Filtering by who the product is for, not what the product is.

Results

Winner: Variation 4 (Dog Size Filters)

Metric	Improvement
Monthly Revenue	+$39,779

Dog size beat need-based, dietary, and product type filters. The simplest framing won.

Why It Worked

1. Everyone knows their dog's size

"Is my dog a powerful chewer?" requires judgment. "Is my dog small, medium, or large?" is objective.

Dog size is the easiest filter to self-select. No thinking required. Customers know immediately which category applies.

The lower the cognitive load, the faster the click.

2. Size is a safety concern

Giving a small dog a treat meant for a large dog is a choking hazard. Pet owners know this.

Size filtering isn't just convenience. It's reassurance. "These products are appropriate for my dog" is a trust signal.

Safety concerns drive immediate action more than preference-based filters.

3. Need-based filters work better for discovery

On the homepage, need-based categories (Powerful Chewers, Picky Eaters, etc.) won. On the collection page, they didn't.

Why the difference? Context.

On the homepage, customers are in discovery mode. "What do you have for dogs like mine?" Need-based framing helps them find their path.

On the collection page, customers are in shopping mode. "Show me what's safe for my dog." Size filtering helps them narrow quickly.

Same brand, different pages, different winning filters.

4. Dietary filters were too niche

Single-Ingredient, Poultry-Free, Beef-Only. These matter to some customers, but not most.

Filter pills work best when they apply to a large portion of visitors. Most visitors don't have dogs with specific allergies. Most visitors do have dogs of a specific size.

Surface the filters that serve the majority. Keep niche filters accessible but not prominent.

5. Product type is how brands think, not customers

Bones, Bully Sticks, Yak Chews, Jerky. That's merchandising language.

Customers don't always know the difference between product types. They know they have a small dog who needs something to chew. The product format is secondary.

Customer-centric filters (who is this for) outperform product-centric filters (what is this).

What This Means for Filter Strategy

Filters should match how customers think, not how you organize inventory.

Test different filter dimensions:

• Who it's for: Size, age, experience level, skin type
• Problem solved: Chewing needs, dietary concerns, specific issues
• Product type: Category-based traditional filtering
• Attributes: Ingredients, materials, features

The winning dimension depends on your product and audience. For pet products, size/safety won. For other categories, it might be different.

FAQ

Should we remove the other filter options entirely?

No. Keep them accessible in the full filter modal.

Filter pills surface the most useful options. Power users who want to filter by ingredient or product type can still do that. The pills just prioritize what matters to most customers.

Why did need-based win on homepage but not collection page?

Different stages of the journey. Homepage is awareness and discovery. Collection page is evaluation and decision.

Need-based helps customers find their category. Size helps customers find safe products within a category. Match the filter to the customer's current task.

How many filter pills is too many?

Four to six visible, with horizontal scroll for more. Too many pills defeats the purpose of quick access.

Prioritize the dimensions that matter most. Everything else stays in the traditional filter.

Does this apply to non-pet brands?

The principle does: filter by who it's for, not just what it is.

Skincare: filter by skin type. Clothing: filter by body type or occasion. Supplements: filter by goal. Put the customer first in your filter strategy.

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