Beauty runs on newness: constant launches, endless shade ranges, dated formulations. Each one is a forecasting problem, and together they move the gross margin more than most teams realise.
Source: The Estée Lauder Companies, FY2023 and FY2025 Form 10-K filings (SEC).
It is easy to treat inventory as an operational detail. In beauty it shows up directly in the gross margin line. The largest beauty group reported that excess and obsolescence charges knocked a full 100 basis points, a whole point, off its gross margin in fiscal 2023, and as it worked that excess back down the same line recovered by roughly a point over the following two years. When a single company's margin can swing a point on inventory alone, inventory is not a back-office concern. It is a board-level one.
Beauty sells novelty. New launches, limited editions and seasonal collections drive the business, but a launch is the hardest thing there is to forecast: little or no history, heavy marketing noise, and a curve that is steep at the start and uncertain after. Over-build and the launch that underdelivers becomes a markdown and then a write-off. Under-build and you stock out on the hero product during its one moment of attention. Either way the launch forecast is where a lot of the margin is won or lost.
A single lipstick is forty shades; a serum is three sizes and two formulations. That proliferation multiplies the number of forecasts and makes each one thin and noisy. The total for a franchise can look healthy while you are sold out of the three shades everyone wants and buried in the long tail of the ones they do not. Like apparel, the planning has to live at the shade-and-size level, because that is where service and markdown are actually decided.
Many cosmetics carry a period-after-opening and a shelf life, and actives degrade. So excess is not patient here. A shade or formulation overbuilt for a launch that fizzled can date on the shelf before it ever clears at a discount, which turns slow stock into destroyed stock. Catching it early, while it can still be promoted or moved, matters far more than in a non-dated category.
The groups that keep excess from eating the margin forecast launches and seasonal demand reproducibly, modelling the curve rather than guessing an uplift, and they update fast as early sell-through comes in. They plan at the shade and size level, size buffers on real variability, and run a pre-obsolescence view so dated stock is caught while it still has value. It is the deterministic, auditable approach DemandIntel brings to cosmetics and personal-care makers and distributors, on your own data, on top of the ERP you already run.
Excess and obsolescence charges hit the gross margin line directly. At the largest beauty group they cost a full point of gross margin in fiscal 2023, and recovering that excess added roughly a point back over two years, a board-level swing driven by inventory alone.
Model the launch curve reproducibly using comparable launches and real signals, then update quickly as early sell-through arrives, rather than committing to a single padded uplift. The goal is a defensible build quantity, not a perfect long-range forecast.
Because service and markdown happen there. A franchise can hit its total while stocking out of the top shades and piling up the slow ones, so the plan has to be built at the shade-and-size granularity, not the product.
Dated formulations mean excess can expire before it clears, turning slow stock into a write-off. A pre-obsolescence view catches dated stock early so it can be promoted or moved while it still has value.
In beauty, a launch you over-built is not just slow stock. It is a dated product racing the calendar to a write-off.
Picture your next S&OP meeting: one number on screen that nobody argues with. See it free, on your own data, in 45 minutes; and if it does not hold up, we part as friends.