Every department is right. So why is inventory still piling up? Lessons from Estée Lauder for cosmetics businesses

Estée Lauder closed fiscal 2023 with sales down about 10%, from US$17.74 billion to US$15.91 billion, and net earnings down by more than half, from US$2.39 billion to US$1.01 billion. ^[1] The cause the company pointed to was not a weak brand or poor products, but excess stock in Asia travel retail, where retailers were tightening inventory.

In fiscal 2024, total sales declined again. The company said Asia travel retail was affected by inventory resets by the company and its retailers, alongside lower consumer conversion. ^[2] In the consolidated balance sheet, inventory and promotional merchandise fell from US$2,979 million at 30 June 2023 to US$2,175 million at 30 June 2024: a decline of US$804 million, or about 27%. ^[2] That difference is calculated from the accounts. It is neither cash recovered nor inventory specific to the Asian channel.

Put simply, one of the world's largest cosmetics companies was hurt for two consecutive years by the same question: how much product should it send into the channel? This is a premium travel-retail case, so it is a useful lesson, not a proxy for the entire Thai cosmetics market.

I start with this story because today I want to talk about Fast-Moving Consumer Goods, or FMCG.

What is FMCG, and why does it interest me?

FMCG means selling consumer products that turn over quickly, have a low unit price and are bought repeatedly. Sales can be offline, online or both. Think instant noodles, snacks, drinks and household products.

Today, though, I want to focus on cosmetics. I see a decision that brand owners and executives can easily get wrong. It is one I have got wrong myself.

I run Belly Thailand, a business selling tents to resorts. I have faced a similar inventory problem to Estée Lauder. The difference is that my tents can remain in saleable condition for more than a year. My unsold stock gives me more time to recover.

If you sell cosmetics, you have to make those decisions faster. How much should you produce? Should you accelerate promotions before a product expires or falls out of fashion, especially when it comes in different shades? One wrong move can quickly erase the profit opportunity. Produce too little and you lose sales. Discount too early and you lose margin. Be too optimistic and excess stock can become dead stock.

Many cosmetics are repeat purchases. That does not mean every brand, shade and channel moves at the same pace. Some products sell well during promotions; others depend on the season or a trend. Shelf life also needs to be assessed product by product, because it varies with the product, its use and its storage. ^[4]

A closer look through Business Analytics

Let me go a little deeper into Business Analytics. If it is not clear on the first read, that is fine. I have been there too. There are only two lines of equations to follow.

My observation about cosmetics is that these businesses often both manufacture and sell, and often manage several brands. That connects with a Business Play called Calculated Ambition. The longer the chain, from raw materials to production, warehouse, channel and cash, and the more brands involved, the more places there are for an upstream decision to create a downstream problem that nobody sees.

Calculated Ambition with original PL artwork and a V2 worked score example: 70 / 60 / 65. These illustrative scores are independent of the inventory models below.

This Play takes an E2E, or End-to-End Inventory Management, approach. It treats everything from the production order to the cash coming back as one connected chain.

Most businesses look at inventory in separate pieces. Purchasing looks at how much to order. The warehouse looks at how much it holds. Sales looks at how much it can sell. Finance looks at how much cash is tied up. Each is right from its own perspective, but nobody sees how an upstream decision creates a downstream problem.

Eliyahu M. Goldratt wrote in The Goal in 1984 that a factory in which every department operates at maximum efficiency is not necessarily the most profitable factory. ^[5] Forty years later, the lesson still holds. It has simply moved from the factory floor to the inventory spreadsheet.

When four departments see four different things

Department	Department	What it wants
Purchasing / production	Q = units ordered or produced	Order a buffer, prevent stockouts, reduce cost per order
Warehouse	I = units left in stock	Less stock left over
Sales	S = units sold	Sell as much as possible
Finance	C = cash remaining	Keep more cash available

These equations leave other variables out to make the idea easier to follow.

Inventory left over (I) ≈ old unsold stock + new production (Q) − units sold (S)

Cash (C) ≈ opening cash + (units sold × selling price) − (new production × cost)

Read it from left to right. Order more and more stock remains. Discount to move it and cash shrinks. Four departments pull in different directions, but all four numbers are connected. They cannot all improve at once, and finance is often the last to find out.

In practice, inventory should reflect goods actually received: closing inventory = opening inventory + actual receipts − units sold − write-offs. Cash should reflect actual receipts and payments, not just profit on paper. ^[6] Beyond month-end cash, I want to know how low cash fell during the month. Having enough on the last day does not pay a bill that was due earlier.

Calculated Ambition treats all four numbers as one problem: choose Q to maximise profit while keeping cash above a defined floor. There is one decision variable, Q. The others follow from it. The left side is opportunity; the right side is risk. That is what Calculated Ambition means: be as ambitious as your calculations show you can afford.

Four departments, one decision: choose Q with profit and cash in view.

Put numbers in, and the picture becomes clearer

Consider one hypothetical product over 30 days. All figures are invented to explain the reasoning. They are not data from Estée Lauder, Belly Thailand or any client.

Assumptions: opening inventory of 200 units, already paid for at THB 100 per unit; selling price THB 250; cost of new stock THB 100 per unit; opening cash THB 200,000; other cash obligations before sales begin THB 20,000; minimum cash reserve THB 80,000; base demand 1,200 units; compare Q = 600 / 1,000 / 1,400 units.

Assume the new stock is paid for upfront, arrives before sales begin, and customers pay immediately. The base case looks like this:

Additional order Q	Units sold in 30 days	Closing inventory	Gross profit*	Minimum cash	Meets THB 80,000 floor?
600	800	0	120,000	120,000	Yes
1,000	1,200	0	180,000	80,000	YEs, exactly at the floor
1,400	1,200	400	180,000	40,000	No

*Gross profit in this example = units sold × (price − unit cost), before operating expenses, interest and tax.

Ordering 1,400 units gives us more stock. But in the base case, we sell exactly as much as we would with an order of 1,000, while 400 units remain. Cash also falls below the floor before sales begin. Looking only at closing cash would hide that breach.

Under these assumptions, 1,000 units produces the highest gross profit among the options that pass the cash constraint. But sitting exactly at the floor means there is no buffer. Additional payments or slower customer collections require a new calculation. And if demand is lower or higher than expected, the same plan leaves different amounts of surplus or unmet demand. This is what I want analysis to reveal before we decide: what we gain, what we risk and what must hold true for the plan to work.

Model A · THB · Immediate customer payment. Ordering 1,000 units gives the highest gross profit among the three cash-feasible options. Illustrative data.

Use Business Play to work through the problem

It is not necessarily a lack of data. Sometimes the problem is that data has not been turned into an analytical process, and priorities have not been set strategically. Some cosmetics become very hard to sell once their moment has passed, turning into dead stock. These products need low leftover inventory. But if nobody identifies which products belong in that group, every group receives the same precautionary buffer.

One data source cannot answer the question. If we ask how much to produce and look only at sales, we are looking at just one cell among many. We are more likely to make a poor decision than if we look across the whole chain.

The example above shows how to think about replenishing one product. It is not an Opportunity Score or a Financial Readiness Score (FRS) for an entire company. Nor does having several brands, by itself, determine the appropriate Business Play. With multiple brands and channels, we can then expand the question: which brand should receive the same pool of cash, or how much additional working capital does a sales promotion require?

I tried it with my own money first

This Play is not just theory. I have run it at Belly Thailand, my own business, since July 2025. Inventory fell by 24%, dead stock by 45%, and inventory turnover rose from four to seven times a year. Those figures belong to my business. They are not results you should expect for yours. But they answer the one question I wanted settled before offering this to anyone: does it actually work?

The screen below shows what this Play looks like when run through Streamlit in a company's own Snowflake account. The figures shown are hypothetical: one SKU and one ordering cycle.

Model B · USD · 60-day customer credit · Q tested in 10-unit steps, selecting 880 units. Illustrative mockup, separate from Model A; no live Snowflake connection.

There is one thing I want you to notice in the image. Expected profit can still rise until Q is around 990. But at that point, paying the supplier pushes cash below the minimum before customers pay. The Play therefore stops at 880. General inventory-planning software would answer 990 because it does not include that final constraint. That is the difference Business Play makes.

Lessons for business owners

1. Every department is right, but only within its own cell. The problem sits in the other cells nobody is watching.

2. Before placing a production order, do not ask only, "How much can we sell?" Ask, "If we order this much, how much cash will remain before customers pay, and how low will it fall along the way?"

3. The longer the chain and the more brands involved, the more we need one connected problem instead of four separate ones.

Power Ladder turns your business data into a Business Play that captures opportunity and manages risk.

---

Discuss your business with Power Ladder

If you owned the brand, how many units would you order?

Make the call in the Calculated Ambition simulation and see which hits first: overstock or a cash squeeze.

Try it now →

Before we finish: if you started with just one decision in your company, which one should bring data, business opportunity, and financial readiness to the same table?

If you do not yet have an answer and would like us to help you find it, get in touch.

Contact us via form ->

Contact us via the form →

Tell us about your business and the decision you would like help with.

---

One question before we finish: which does your company see first today, shipments into the channel or what consumers actually buy?

References and scope

Sources checked around 16–17 September 2026. The Estée Lauder case is historical, not a statement of its latest quarterly performance.

1. The Estée Lauder Companies Inc., Fiscal 2023 Results, 18 August 2023 (SEC Form 8-K / Exhibit 99.1). https://www.sec.gov/Archives/edgar/data/1001250/000100125023000110/elq4fy2023exhibit991.htm

2. The Estée Lauder Companies Inc., Fiscal 2024 Results, 19 August 2024. https://www.elcompanies.com/en/news-and-media/newsroom/press-releases/2024/08-19-2024-114531935 : includes Asia travel-retail commentary and inventory and promotional merchandise. The US$804 million / approximately 27% reduction is calculated from the accounts.

3. IHL Group, The 2026 Inventory Distortion Study, public summary. https://www.ihlservices.com/product/inventory-distortion-study-2026/ : the public web summary is used; 6.2% is IHL's global estimate under its definitions.

4. U.S. FDA, Shelf Life and Expiration Dating of Cosmetics. https://www.fda.gov/cosmetics/cosmetics-labeling/shelf-life-and-expiration-dating-cosmetics : cited for variation in shelf life, not as a statement of Thai labelling law.

5. Eliyahu M. Goldratt and Jeff Cox, The Goal: A Process of Ongoing Improvement, North River Press, 1984.

6. IFRS Foundation, IAS 7 Statement of Cash Flows. https://www.ifrs.org/issued-standards/list-of-standards/ias-7-statement-of-cash-flows/ : supports the distinction between profit and cash receipts/payments. The article's formulas are decision illustrations, not a complete cash-flow statement.

The Belly Thailand figures are results reported by the author for his own business. Do not use them to forecast your business. The Q / cash table and screen are hypothetical illustrations, not measured results for Estée Lauder or a client.

Contact: powerladder.tech · dithanon@powerladder.tech

By Dithanon Khrutmuang (O), founder of Power Ladder and Belly Thailand.