The “average customer” causes planning fiction. They never perfectly existed, but for decades, the averages held well enough. Size curves were relatively stable. Categories behaved predictably season to season. A planner with enough historical data and a reasonable read on the macro environment could build a defensible plan. That's no longer true.
Consumer demand is fragmenting; not gradually, but structurally, and across every major retail category at once. The causes are varied: demographic shifts, new sports, changing beauty philosophies, trade policy, commodity price volatility, and a post-pandemic consumer who is simultaneously more value-conscious and more willing to spend on the right thing. Each of these trends is real on its own. Together, they're creating a planning environment where historical data is increasingly unreliable and category-level forecasts are smoothing over populations that are actually moving in opposite directions.
This is the moment when most articles would say something about the "ever-changing retail landscape." Here's what's actually happening.
Why Demand Fragmentation Is a Different Problem Than Trend Volatility
Retail has always had trend cycles. Planning teams have always accounted for style volatility, seasonal variance, and the occasional category spike. That's normal. Fragmentation is different in kind, not just degree.
Normal trend volatility shifts what customers want within a category. Fragmentation changes who the customer is by splitting a single category into two or more distinct demand pools with different size profiles, different price tolerances, different replenishment patterns, and different sell-through curves. When a category fragments, the averages don't just get noisier; they start lying to you.
Your size curve from three seasons ago doesn't describe any real customer in the room today. Your blended category forecast is hitting a number that no individual segment actually produces. You end up with too much inventory in the middle of a distribution that's splitting at both ends, and you miss the emerging segment entirely until it's large enough to show up in aggregate data. By then, you're behind.
How Structural Shifts Are Splitting Demand Across Every Major Category
Size run planning is one of the clearest early signals. Apparel companies are shifting to stock more smaller sizes and slightly fewer larger sizes as demand patterns shift across the customer base. If you haven't revisited your size curve assumptions in the last twelve months, your buy is already misaligned because the underlying population data has moved.
The same split is happening at the category level in sporting goods. McKinsey's data shows an expanding gap between highly active consumers and a growing sedentary cohort. Two populations buying fundamentally different products at fundamentally different price points, increasingly within the same category classification. The casual fitness middle is shrinking, and a blended category forecast misses both ends simultaneously.
In health and beauty, it looks different but creates the same planning problem. Consumers are consolidating their skincare routines, choosing fewer products with higher functional value rather than layering multiple single-purpose items. SKU breadth is contracting. Average unit retail is rising. A planning model that expects breadth and depth to move together will get both wrong.
In jewelry, the bifurcation is the starkest. Natural diamonds have seen a comeback in engagement ring share, rising from 52.6% to 57.3% while lab-grown continues growing in volume, now accounting for more than 45% of U.S. engagement ring purchases. Lab-grown diamonds now cost approximately 85–90% less than comparable natural stones. These are not two segments of the same customer. They require completely different inventory strategies: lab-grown needs to be planned like a commodity: high volume, fast turn, price-competitive while natural requires curation, scarcity management, and longer sell-through tolerance. Planning them with the same model destroys margin on one end and misses demand on the other.
When Your Comps Don't Cover the Category
Some of the hardest planning problems right now are caused by entirely new demand pools forming faster than systems can absorb them.
Pickleball grew 22.8% in 2025, reaching 24.3 million U.S. participants. The dedicated apparel market for the sport is projected to reach $337 million in 2026. Search volume peaks sharply in July, creating a seasonal demand spike with its own replenishment logic. This is not a trend you can absorb with a few incremental SKUs in racquet sports. It's a category that needs its own size curve, its own location strategy, its own open-to-buy. Planners who folded it into general activewear are sitting on the wrong depth at peak season.
Indoor climbing is following a similar arc. Participation jumped 13.1% to 7.1 million in 2025, while sport and boulder climbing grew 20.1% to 3.2 million. It's a technically distinct category with a different size run profile than anything adjacent to it. If you're not planning it separately, you won't know that until the stockouts show up.
The planning challenge in both cases is the same: there's no clean historical data to anchor the forecast. You're building a buy for a category that's growing faster than your comps can reflect. That requires a different kind of planning process, one that can incorporate forward-looking signals rather than relying exclusively on what sold three seasons ago.
When External Shocks Disguise Themselves as Demand Signals
Not every shift in the demand curve reflects a real change in consumer preference. Some of it is noise introduced by external conditions. Planners who can't distinguish the two will make the wrong bet.
Almost all U.S. apparel imports will be subject to higher tariffs in 2026. The near-term effect is demand pull-forward: consumers stocking up before price increases hit, producing a spike that looks like genuine category growth. The back-half correction, lower demand as consumers deplete stockpiled inventory at higher prices, looks like a slump. Planners who read the pull-forward as a trend signal will be overinventoried when the hangover arrives.
Gold pricing is creating a similar distortion in jewelry. As of early 2026, gold prices surged above $5,100 per ounce, roughly 76–77% higher than March 2025 levels. Buyers are recalibrating purchase decisions to stay within budget, and assortment preferences are shifting in ways that blend genuine taste change with price-driven substitution. Planning teams that can't separate the two will misread the demand signal, and make the wrong bet when commodity prices eventually normalize.
The deeper issue is that external shocks: tariffs, commodity prices, macroeconomic pressure are now stacking on top of structural demand fragmentation. It used to be enough to model the trend and apply a macro adjustment. Now you're layering shocks onto a customer base that's already fragmenting at the segment level.
When Historical Data Stops Being a Reliable Input
The "average customer" assumption is baked into most planning infrastructure: in size curves built from blended historical data, in category-level forecasts that aggregate across diverging segments, in replenishment logic that treats all demand within a classification the same way.
That assumption held when customer behavior was relatively homogenous within a category. It breaks down when the category starts containing multiple distinct populations.
That assumption held when customer behavior was relatively homogenous within a category. It breaks down when the category starts containing multiple distinct populations.
The problem compounds when external conditions: tariffs, commodity swings, macroeconomic pressure layer on top of structural fragmentation. It used to be enough to model the trend and apply a macro adjustment. Now you're applying shocks to a customer base that's already splitting at the segment level. The inputs to your forecast are noisier, the historical comps are less reliable, and the margin for error on your initial buy is narrower than it's ever been.
How Planning Teams Are Adapting to a Fragmented Demand Environment
The teams navigating fragmented demand well are planning with more granularity, more frequency, and more willingness to override historical patterns when the data signals something has structurally changed.
Specifically: they're working at the segment level rather than the category level, building size curves from recent behavioral data rather than multi-season averages, and forecasting emerging categories separately rather than folding them into adjacent classifications. They're treating external shocks — tariff pull-forward, commodity swings as distinct planning inputs rather than noise to be averaged out.
And they're planning with enough flexibility to course-correct mid-season. In a fragmented demand environment, initial forecast accuracy matters less than the ability to reforecast and reallocate quickly when the plan diverges from reality. Tighter OTB cycles, shorter commitment windows, and faster replenishment loops aren't just operational improvements. They're structural responses to a market that's moving faster and more unpredictably than it did a few years ago.
The plan will be wrong. The question is how fast you can adjust and whether your planning process gives you the visibility to know when that moment has arrived.
Toolio is a merchandise planning platform built for retail teams managing complex, fast-moving assortments. If your planning process is struggling to keep pace with how fast your customer base is changing, see how Toolio works.
