Stratrix

Quick reference

Memory aid

Cows pay, stars and question marks spend, dogs go unless they hold something up.

TL;DR

Define each business's market. Plot it by market growth and relative share, sized by revenue. Put its free cash flow beside it. Then draw the arrows: where each cash cow's surplus goes, the share target and stop date for each question mark, and which dogs to exit once you have checked what they share with the rest.

What Is BCG Growth-Share Matrix?

The BCG growth-share matrix plots each of a company's businesses on two axes: how fast its market is growing, and how big its share is next to the largest rival's. That gives four boxes. Stars lead growing markets, cash cows lead slow ones, question marks trail in growing markets and dogs trail in slow ones. What it decides is where a portfolio's cash goes: from the cows that throw it off to the stars and question marks that need it, with the dogs sold or closed unless they serve another business.

The value of a product is completely dependent upon obtaining a leading share of its market before the growth slows.

— Bruce Henderson, "The Product Portfolio," Boston Consulting Group (1970)

Most people treat the matrix as a report card: stars are winners, dogs are losers, and the job is to sort. Bruce Henderson built it to answer a different question: where should the money go next? His 1970 essay is about cash, not rank. "The portfolio composition is a function of the balance between cash flows," he wrote, and "Every company needs products in which to invest cash." A cash cow is the business whose surplus, in his words, "need not, and should not, be reinvested in those products." A star is where that surplus goes to hold a lead while the market grows; a question mark is a bet that the surplus can buy one. So the matrix is not finished when every business has a label. It is finished when every arrow is drawn: which business pays, which is paid, how much and for how long. Alphabet runs it in plain sight. In 2025 Google Services earned $139.4 billion of operating income while Other Bets, the group that includes Waymo, lost $7.5 billion, according to Alphabet's annual report. The labels are easy. The arrows are the strategy.

BCG growth-share matrix

Market growth rate on the vertical axis, relative market share on the horizontal axis with high share on the left. Circles are sized by revenue; cash flows from the cash cows to the stars and question marks.

Stars

High growth, high share: fund to hold the lead

Question Marks

High growth, low share: fund to leadership or exit

Cash Cows

Low growth, high share: send the surplus elsewhere

Dogs

Low growth, low share: exit unless another business needs it

Relative market share (High → Low)

Origin & context

Bruce Henderson, who founded the Boston Consulting Group in 1963 and ran it until 1980, set out the idea in "The Product Portfolio," a short essay BCG published in 1970. It opens: "To be successful, a company should have a portfolio of products with different growth rates and different market shares." Its first rule was "Margins and cash generated are a function of market share," and its conclusion was blunt: "every product should eventually be a cash generator; otherwise it is worthless." Henderson's fourth box was not dogs but pets, products he called "essentially worthless, except in liquidation." In 2014 BCG's Martin Reeves, Sandy Moose and Thijs Venema revisited the matrix. By the early 1980s, they wrote, it "was used by about half of all Fortune 500 companies, according to estimates," citing Philippe Haspeslagh in Harvard Business Review (1982). They also found it aging: "market share is no longer a direct predictor of sustained performance," and in industries where change had sped up, "the average time spent in a quadrant halved: from four years in 1992 to less than two years in 2012."

BCG Growth-Share Matrix in action

Alphabet

One cash cow, many bets

On August 10, 2015, Larry Page told the world that Google would become one company among several, under a new parent called Alphabet. "Alphabet is mostly a collection of companies," he wrote, and he made a promise any BCG partner would recognize: "We will rigorously handle capital allocation and work to make sure each business is executing well."

The shape is Henderson's, drawn at enormous scale. Search and its sister businesses are the cash cow. In 2025, by Alphabet's annual report, Google Services earned $139.4 billion of operating income, while Other Bets, the group that holds Waymo, brought in $1.5 billion of revenue and lost $7.5 billion. The same report says Waymo announced a $16.0 billion investment round in February 2026, "the significant majority of which was funded by Alphabet."

The harder test was Google Cloud, a question mark by any reading. In November 2018 CNBC reported that "Google has failed to eat into Amazon's massive lead in the market" and that "Microsoft has clearly established itself in second place." Alphabet's 2021 annual report showed Cloud losing $4.6 billion in 2019, $5.6 billion in 2020 and $3.1 billion in 2021. Henderson's rule for a question mark is to buy leadership or get out. Alphabet kept paying through years of losses, and in 2025 Google Cloud earned $13.9 billion of operating income on $58.7 billion of revenue.

The cash cow paid for both kinds of bet, and one of them became a second profit engine years after a strict reading of the matrix would have forced a choice. Money also flows back: Alphabet's 2025 report says "Our early investments in AI started out as moonshots but are now incorporated into our core products," a link between businesses that no box shows.

Read the full story: how search pays for Google's moonshots →

Procter & Gamble

Keep 65 brands, let 100 go

On August 1, 2014, A.G. Lafley, Procter & Gamble's chief executive, told investors the company would get much smaller. P&G would divest, discontinue or consolidate about 100 brands and keep about 65. "This will be a much smaller and less complicated company of brands that will be easier to operate," he said, as Bloomberg News reported. The 65 it kept, the company's next annual report said, "will retain about 85% of sales and 95% of before-tax profit."

By the arithmetic of P&G's own figures, the brands that went brought in about 15 percent of sales but only about 5 percent of before-tax profit. Henderson's verdict on businesses that tie up effort and return little cash was short: "Pets are not necessary." The big pieces went in big deals. In November 2014 P&G agreed to hand Duracell to Berkshire Hathaway in exchange for Berkshire's P&G shares, a deal it valued at $2.9 billion, and on July 9, 2015 it accepted Coty's offer of $12.5 billion for 43 beauty brands. By its fiscal 2015 annual report, 93 of the roughly 100 brands had been sold, discontinued or consolidated, counting deals already announced.

Duracell is the case that shows where the two axes stop. P&G's August 2014 release listed it among the company's "leadership brands," and when the exchange closed on February 29, 2016, chief executive David Taylor called it "a strong, global business with a great future ahead of it." A leader was sold anyway. The annual report gave the test P&G used: "categories and brands that are structurally attractive and that play to P&G strengths." Share and growth were two inputs. Fit decided.

Duracell left on February 29, 2016. The brands P&G kept were the ones it expected to hold about 85 percent of sales and 95 percent of before-tax profit, with about 100 fewer brands to run.

Read the full story: why P&G cut 100 brands →

Core Components

1

Stars

High share in a fast-growing market. Profitable on paper, but holding the lead while the market grows takes most of the cash they make. If they keep the lead when growth slows, they become cash cows.

Example

Henderson: "The high share, high growth product is the 'star.'" In the invented Northfield Tools sample below, pressure washers sit here.

2

Cash cows

High share in a slow-growing market. They throw off more cash than they need to hold their place, and the surplus is meant to leave.

Example

Google Services earned $139.4 billion of operating income in 2025, by Alphabet's annual report, while Other Bets lost $7.5 billion.

3

Question marks

Low share in a fast-growing market. Henderson: "They almost always require far more cash than they can generate." Funded to leadership they become stars; left as they are they fade.

Example

Google Cloud trailed Amazon and Microsoft in 2018, CNBC reported, and lost $5.6 billion in 2020 by Alphabet's 2021 annual report.

4

Dogs (Henderson's pets)

Low share in a slow-growing market. They may show an accounting profit, but holding share eats it, so little cash comes out. Henderson: "Pets are not necessary."

Example

Henderson called a pet "essentially worthless, except in liquidation." Check what else it holds up before you sell it.

Henderson's fourth box was "pets," not dogs. Pets, he wrote, "are evidence of failure either to obtain a leadership position during the growth phase, or to get out and cut the losses."

When to Use BCG Growth-Share Matrix

Scenario 1

A BCG matrix example, worked through (invented company and figures)

Problem it solves: Shows the full sequence on a sample portfolio, from the boxes to the cash moves they point to.

Real-world application

Northfield Tools is an invented company, and every figure here is invented for illustration. It has four businesses. Hand tools: market growing 2 percent a year, Northfield's share 25 percent against 10 percent for the largest rival, a relative share of 2.5, and $40 million of free cash a year: a cash cow. Pressure washers: market growing 12 percent, share 30 percent against 15 percent, relative share 2.0, roughly breaking even on cash because it is spending to keep up with demand: a star. Smart garden sensors: market growing 25 percent, share 4 percent against a leader's 20 percent, relative share 0.2, using $15 million a year: a question mark. Garden hoses: market growing 1 percent, share 5 percent against a leader's 30 percent, relative share about 0.17: a dog. Now draw the arrows. The $40 million from hand tools pays first for pressure-washer capacity, because a star that loses its lead while the market grows never becomes a cow. The sensors get money only against a share target with a date, because a question mark that stays at 0.2 will need $15 million every year. The hoses look like an easy sale until you check what they share: if the same hardware-store buyers order hoses with the hand tools, selling them may cost shelf space for the cow. The labels took ten minutes. The decision is the arrows and the checks.

Scenario 2

Deciding which businesses to sell, close or keep

Problem it solves: Separates the businesses that pay for the portfolio from the ones that only use it, before an exit is chosen.

Real-world application

Procter & Gamble's 2014 plan kept about 65 brands that, by its annual report, would retain about 85% of sales and 95% of before-tax profit, and let about 100 go.

Scenario 3

Funding new ventures from a mature core

Problem it solves: Makes the transfer from a cash cow to its bets visible, so it can be sized and reviewed rather than assumed.

Real-world application

Alphabet reports Google Services, Google Cloud and Other Bets separately; in 2025 Google Services earned $139.4 billion of operating income and Other Bets lost $7.5 billion.

A label is not a decision

A grid with every business in a box and no cash moving between them has not used the matrix. For each cash cow, name the business its surplus will fund and for how long; for each question mark, name the share it must reach and the date you stop paying if it does not.

How to Apply BCG Growth-Share Matrix: Step by Step

Before you start

  • →Revenue and free cash flow for each business or product line
  • →Each business's market share and the share of its largest rival, in a market you have defined
  • →The growth rate of each market, not of your own sales
Tools:A grid with the two axes and circles sized by revenueA table of each business's free cash flow, market growth and relative shareSegment reporting from your own accounts or, for a competitor, its annual report
1

Define the businesses and their markets

List the businesses, brands or product lines to plot, and write down the market each one competes in: the customers, the products and the geography.

Tips

  • ✓If two products have different customers or different rivals, plot them separately

Common mistakes

  • ✗Defining a market so narrowly that every business becomes a leader, or so widely that none is
2

Measure market growth

Find the annual growth rate of each market. Choose one line between high and low growth and use it for every business, so the boxes are comparable.

Tips

  • ✓Write down the line you chose and why; the boxes move if it moves

Common mistakes

  • ✗Using your own sales growth instead of the market's
3

Calculate relative market share

Divide your share by the share of the largest rival. Above 1.0 means you lead; 0.2 means the leader is five times your size.

Tips

  • ✓For the leader, divide by the second-largest player

Common mistakes

  • ✗Using absolute share, which hides how far behind the leader you are
4

Plot the circles

Place each business on the grid and size its circle by revenue, so the portfolio's weight is visible at a glance.

Tips

  • ✓Plot last year's position too, with an arrow, to show which way each business is moving

Common mistakes

  • ✗Treating a business on the border as firmly in one box
5

Put the cash beside the box

Write each business's actual free cash flow next to its circle. A cash cow should be producing cash and a question mark using it; where the numbers disagree with the box, trust the numbers and find out why.

Tips

  • ✓Henderson's whole case rests on cash; a box without a cash figure is a guess

Common mistakes

  • ✗Stopping at the labels
6

Draw the arrows

Decide where each cash cow's surplus goes, how much and for how long. Give each question mark a share target and a date; if it misses, the money stops.

Tips

  • ✓Fund few question marks fully rather than many partly; a question mark that never gains share keeps needing cash

Common mistakes

  • ✗Reinvesting a cow's cash in the cow out of habit
7

Check what the boxes cannot see

Before selling a dog or starving a question mark, list what it shares with the rest of the portfolio: customers, channels, costs, technology. Then ask whether the business fits what the company is good at, the test P&G used in 2014.

Tips

  • ✓Ask the sales team which orders would be lost if the business went

Common mistakes

  • ✗Selling a small business that keeps a large one on the shelf

Only a diversified company with a balanced portfolio can use its strengths to truly capitalize on its growth opportunities.

— Bruce Henderson, "The Product Portfolio," Boston Consulting Group (1970)

Value & Outcomes

Primary benefit

Shows which businesses pay for the portfolio and which spend its money, so the transfers between them can be decided rather than assumed.

Additional benefits

  • ✓Gives a board and its managers one picture of the whole portfolio
  • ✓Makes the cost of a long-running bet visible year by year
  • ✓Forces a date and a target on every question mark

What you'll learn

  • →How to place a business by market growth and relative market share
  • →How to read the boxes as sources and uses of cash
  • →When to override the boxes for synergies and fit

Typical outcomes

A list of where each cash cow's surplus goes, and for how longShare targets and stop dates for each question markExit candidates, checked for what they hold up elsewhere

Best Practices

📋 Preparation

  • •Define each market before measuring share in it
  • •Gather free cash flow by business, not only revenue and profit

🚀 Execution

  • •Use relative share, not absolute share
  • •Show last year's position beside this year's
  • •End with cash moves, not labels

🔄 Follow-Up

  • •Re-plot at least once a year; BCG's 2014 review found the average time in a quadrant halved between 1992 and 2012 in industries where change sped up
  • •Check each question mark against its share target on the date you set

💎 Pro Tips

  • •Plot a competitor's portfolio from its segment reporting to see which business it is milking and which it is feeding
  • •Ask which business would fund the next bet if the current cash cow slowed
Check the boxes against the cash

Henderson's first rule was "Margins and cash generated are a function of market share." BCG's own 2014 review found that "market share is no longer a direct predictor of sustained performance." Before you trust a box, put each business's actual cash flow beside it.

Limitations & Pitfalls

⚡

Market share stands in for cost advantage, and the stand-in has weakened. Henderson's first rule was "Margins and cash generated are a function of market share"; BCG's 2014 review found that "market share is no longer a direct predictor of sustained performance."

Mitigation: Compare each business's margins and cash flow with its rivals' directly instead of inferring them from share

⚡

Some growth markets never pay. A question mark can need cash year after year and never reach the lead; Henderson wrote that question marks "almost always require far more cash than they can generate."

Mitigation: Set a share target, a cash limit and a stop date before funding, and keep to them

⚡

The boxes ignore what businesses share. A dog can carry costs, customers or shelf space for a cow, and a bet can feed the core: Alphabet says its early AI investments "started out as moonshots but are now incorporated into our core products."

Mitigation: Before an exit, list what the business shares with the rest of the portfolio and what would be lost with it

⚡

Two axes leave out fit. P&G sold Duracell, a brand it had listed among its leadership brands, because it chose categories that "play to P&G strengths."

Mitigation: Add a fit test, and use the GE-McKinsey matrix when market attractiveness depends on more than growth

⚡

The boxes move faster than they used to. In industries where change sped up, BCG found, the average time spent in a quadrant fell from four years in 1992 to less than two years in 2012.

Mitigation: Re-plot every year, show the direction of travel, and treat a label as current, not permanent

⚡

The market definition decides the box. Draw a market narrowly and almost anything leads it; draw it widely and almost nothing does.

Mitigation: Define each market by the customers and rivals the business actually faces, and write the definition down

Where this comes fromBruce Henderson / Boston Consulting Group introduced this (1970).

Sources & further reading

Companies that faced this decision
2015Anheuser-Busch InBev4 dated outcomes2022Activision Blizzard1 dated outcome2009Adobe1 dated outcome
All 98 verified Adjacency Expansion records →

More Management Frameworks

Other frameworks in the Stratrix library.

Strategic Analysis

Porter's Five Forces

Five pressures decide how much money an industry can make — and usually only one of them actually does. Porter's framework finds that force before you waste time scoring the other four.

Strategic Analysis

Activity System Mapping

An activity system map draws a strategy as the activities that carry it and the lines where one supports another. Michael Porter's point: a rival can copy one activity but rarely the whole system. The same lines show what else moves when you change one yourself.

Strategic Analysis

Coopetition and the Value Net

Coopetition is cooperating with a competitor where both gain. The value net, from Adam Brandenburger and Barry Nalebuff, maps the four players around a company: customers, suppliers, competitors and complementors. Its lesson is that the same company can be your rival and your partner at once, so map roles, not names.

Strategic Analysis

SWOT Analysis

A strategic planning framework that evaluates an organization's internal Strengths and Weaknesses alongside external Opportunities and Threats to inform decision-making and strategy formulation.

Strategic Analysis

Profit Pools

A profit pool is the total profit an industry earns, mapped across every activity in its value chain. The money rarely sits where the revenue is. It gathers where customers stop shopping around, and the map exists to find that place first.

Strategic Analysis

7 Powers

7 Powers names the seven ways a business can earn more than its rivals for years, not quarters. Its test is strict: an advantage counts only when it pays and when something stops rivals from copying it. A great product that a rival can match is not a power.

Tools to try

More from Stratrix

The Strategist’s Craft

The methods behind the job: the planning cycle, initiative charters, decision rights, the memo that gets a yes.

Methods and templatesStart with the craft →

Ask Stratrix

Three questions a month are free.

Stratrix Team

Every Casebook, three Strategy Packs a seat each year and Ask Stratrix, for teams that face these decisions all year.

$1,999 a year · $499 a seatWhat is included →

New analyses, by email

The defining moves in business, and what they teach, as they publish.

We'll email you a link to confirm. Unsubscribe anytime. Privacy policy.

See BCG Growth-Share Matrix in the wild.

Follow BCG Growth-Share Matrix across the companies and lenses where it actually shaped the strategy — the decision, the rationale, and how it played out.

Explore the lenses →