BCG matrix: definition, worked example and limits

The BCG matrix, or growth-share matrix, is a portfolio analysis tool designed by Bruce Henderson, the founder of the Boston Consulting Group, and published in 1970. It is meant for a company with several businesses that has to allocate a limited resource, cash, between them.

It rests on two ideas. The first is that a business that dominates its market makes more money than its competitors, because it produces more and its unit costs are lower. The second is that a fast-growing market consumes cash, since you have to invest to keep up, whereas a mature market releases it.

Each business is therefore placed on two axes.

• Market growth, which indicates its investment needs. By convention, growth above 10% a year is often called high.
• Relative market share, meaning the company's share divided by that of its largest competitor. A company with 30% of a market where the next player has 15% has a relative share of 2. Above 1, it is the leader.

Crossing the two gives four categories, which everyone remembers for their colourful names.

The overall logic is a cash circuit. Cash cows, leaders in mature markets, generate more money than they consume. That money funds the stars, leaders in growing markets that must invest to stay there, and a few selected question marks, which you hope to turn into stars. Dogs, small players in markets with no growth, do not deserve investment.

Over time, market growth slows. A star that has kept its position then becomes a cash cow, and a question mark that has not broken through becomes a dog. The matrix therefore describes a cycle: you always need stars in the making to replace tomorrow's cash cows.

Here is the prompt. Solène is a cosmetics group with four businesses. Its CFO has €50M of cash a year, generated by fragrances, and asks how to allocate it. The interviewer provides the following table.

Place the businesses. Fragrances lead a mature market: this is the cash cow, which matches what the prompt says about where the cash comes from. Skincare leads a growing market: a star. Make-up is small in a fast-growing market: a question mark. Haircare is small in a flat market: a dog.

Read what the matrix suggests. Fund skincare first, make a call on make-up, stop investing in haircare.

Ask for the missing numbers. You ask what each option costs. Skincare needs €30M a year to keep up with its market. For make-up, the prompt says it would take €60M over three years to hope to triple market share. There is exactly €20M a year left after skincare, which is €60M in three years. The numbers add up, with no safety margin.

Go beyond the boxes. You ask the question the matrix does not ask: are these businesses linked? The interviewer answers that haircare is made in the same plant as skincare and absorbs part of its fixed costs. Stopping it would therefore make the group's star more expensive to produce.

Recommendation. Invest €30M a year in skincare. Commit to make-up with a checkpoint at eighteen months: if relative share has not doubled, sell the business while its market is buoyant. Keep haircare without investing in it, because it covers costs of the shared plant, and only consider a sale if the plant can be filled another way.

The matrix breaks first on the definition of the market. A premium electric bike brand is a dwarf in the bicycle market and a leader in its niche. Depending on the scope you choose, the same business is a dog or a star. The matrix does not say which scope is right, and that choice determines the whole result.

It also breaks because market share is not profitability. The idea that the leader always earns more holds in industries where volume drives costs down. In luxury, niche software or services, a small specialist can be far more profitable than the number one.

Third limit, shown by the example: it treats each business as if it stood alone. A shared plant, common customers, an umbrella brand: a business classed as a dog may support the others, and removing it costs more than it loses.

Fourth limit: two criteria do not sum up how attractive a business is. Market growth says nothing about its profitability or how intense competition is, which Porter's five forces look at more closely.

Finally, its thresholds are arbitrary. A business growing at 9% is not different in kind from one growing at 11%.

In an interview, the BCG matrix is rarely asked for by name. It is useful when a case is about allocating resources between several businesses, countries or products. You can then draw on it without naming it, keeping its two good questions: which business generates cash and which consumes it, and where the company is in a position of strength.

Avoid sticking on a label and deducing an action from it. Saying "make-up is a question mark, so we must invest or exit" adds nothing. The value comes afterwards: what the investment costs, whether the company can afford it, what would win it market share, and at what point you decide to stop.

Always ask about links between businesses before recommending a closure or a sale. And check the market definition: ask against whom market share is calculated.

A portfolio case ends with numbers, not with a ranking. For each business you should be able to say what it brings in, what it requires and what you do with it, which often means a quick profit tree. The FrameLab course files the matrix under business mental models: a lens worth knowing to save time, which does not replace a structure built from the client's question. The real interview cases will show you that it almost never appears as is.

A media group has three businesses. Its print daily holds 40% of a market shrinking by 5% a year, ahead of a competitor at 20%. Its subscription news site has 5% of a market growing 20% a year, where the leader has 25%. Its conference business has 10% of a stable market, where the leader has 30%. Classify each business, then say what the matrix would wrongly lead you to conclude.