Variant Strategy

Optimize your product variant mix — colors, sizes, materials, bundles, and configurations — by analyzing sales performance patterns, market demand signals, and inventory holding costs. This skill helps ecommerce operators eliminate underperforming variants that drain resources while identifying high-potential variant gaps that competitors are filling.

Use when

• A seller says "I have 12 color options but only 3 are selling well, should I cut the rest" and needs a data-driven framework to decide which variants to keep, retire, or add
• An ecommerce operator asks "what sizes should I stock for my new clothing line launch on Shopify" and needs a size curve recommendation based on category benchmarks and target demographics
• A brand manager wants to "figure out why my variant conversion rates are so different across colors" and needs an analysis connecting variant attributes to purchase behavior
• A marketplace seller needs help deciding "whether to add a bundle variant or a new standalone SKU" on Amazon or TikTok Shop to maximize catalog performance without cannibalizing existing sales

What this skill does

This skill takes your existing product variant data — including sales volumes, return rates, inventory turnover, and margin per variant — and produces a comprehensive variant optimization plan. It segments variants into performance tiers (hero, core, long-tail, and candidate-for-retirement), identifies attribute patterns that drive conversion (such as color preferences by season or size distribution by category), and recommends specific actions: which variants to discontinue, which to replenish more aggressively, and which new variants to test based on market gaps and competitor offerings. The analysis accounts for inventory carrying costs, minimum order quantities from suppliers, and platform-specific considerations like how variant count affects search ranking.

Inputs required

• Current variant catalog (required): A list of your product variants with attributes like color, size, material, and current retail price. Example: "Blue-S, Blue-M, Blue-L, Red-S, Red-M, Red-L for Product X at $29.99 each"
• Sales data by variant (required): Units sold per variant over a defined period, ideally 30-90 days. Example: "Blue-M sold 145 units, Red-S sold 12 units last quarter"
• Return rate by variant (optional): Percentage of returns per variant, which helps identify sizing issues or color mismatch problems that inflate costs
• Competitor variant offerings (optional): What variants your top competitors offer for similar products, which helps identify market gaps and potential opportunities
• Supplier constraints (optional): Minimum order quantities, lead times, and cost differences between variants, which shapes the feasibility of adding or removing options

Output format

The output is a structured variant optimization report with four major sections. First, a Variant Performance Matrix that ranks every existing variant across revenue contribution, margin, sell-through rate, and return rate in a sortable table format with color-coded performance tiers. Second, a Recommended Actions List specifying exactly which variants to keep as-is, which to mark for clearance, which to discontinue at next reorder, and which new variants to introduce with a test quantity recommendation. Third, a Variant Attribute Analysis that breaks down how each attribute dimension (color, size, material) correlates with conversion and satisfaction, highlighting the strongest and weakest attribute values. Fourth, an Implementation Timeline with phased steps for executing variant changes, including inventory rundown periods for retiring variants and initial test order quantities for new additions.

Scope

• Designed for: ecommerce operators, product managers, merchandising teams, and inventory planners
• Platform context: Amazon, Shopify, TikTok Shop, Shopee, or platform-agnostic
• Language: English

Limitations

• Does not pull live sales or inventory data from your store; you must provide the data for analysis and the recommendations are only as accurate as the inputs
• Cannot predict consumer preference shifts or fashion trend changes with certainty; variant recommendations reflect current and historical patterns
• Not a substitute for supplier negotiations or manufacturing feasibility assessments when adding new variants