The BCG matrix isn't dead — you're just using it wrong
Five decades in, the growth–share matrix still earns its place on the boardroom wall. What doesn't hold up are the five ways most teams misuse it.
Fifty years after Bruce Henderson sketched the growth–share matrix on a napkin, it still gets pinned to boardroom walls. It survives because nothing communicates portfolio logic to a board faster: four quadrants, a few bubbles, and everyone in the room is arguing about the right things within thirty seconds. The problem isn't the tool — it's the five ways most teams misuse it.
Mistake 1: Linear share axes
The horizontal axis on a BCG matrix is relative market share — your share divided by the largest competitor's. That means a value of 0.5× and a value of 2× are equally far from parity, just in opposite directions. On a linear axis they aren't; 2× visually dwarfs 0.5×, and every "question mark" collapses into the left margin.
Put relative share on a log scale from 0.1× to 10×, with the 1.0× line down the middle. Suddenly a business at 0.3× and a business at 3× read as mirror images — because strategically, they are.
Mistake 2: Bubbles sized by radius
The default in most charting tools maps your value (usually revenue) to radius. That's the wrong dimension. The eye reads area, and area grows with the square of radius — so a business twice the revenue of another ends up looking four times bigger.
- Radius-scaled: $200M looks 4× the size of $100M. Executives read "dominant."
- Area-scaled: $200M looks 2× the size of $100M. Executives read "twice as big."
The fix is one line of math: r ∝ √value. Every serious portfolio tool (STRATObubbles included) does this by default — but check yours, because the ones that don't are the reason your cash cows look monstrous.
Mistake 3: Treating the thresholds as sacred
The classic 10% growth line and 1.0× share line are defaults, not commandments. They encode assumptions about your industry: what counts as fast growth in enterprise SaaS is catastrophic decline in cloud infrastructure. Yet teams routinely paste the defaults on a chart and then argue about which quadrant a business "really" belongs in.
The matrix is a sentence, not a formula. The thresholds are where you say what "good" means in your industry.— Prof. Anna Roth, University of St. Gallen
Calibrate the thresholds once a year with a three-step process:
- 1. Growth threshold: take the weighted-average growth of every market you play in. That's your dividing line between "growth" and "mature."
- 2. Share threshold: decide whether you're benchmarking against the leader (1.0×) or the top three (0.5×). Boards should agree on which definition of "strong" they're using.
- 3. Document it: put the threshold and the reason on the chart itself. "Growth > 8% (market average, FY26)." Now future you can't drift.
Mistake 4: Reading quadrants as verdicts
The quadrants are conversation starters, not sentences. A "dog" isn't automatically for the chopping block; a "star" isn't automatically safe. The mistake is walking into a review with the strategy already written in the four labels.
| Quadrant | Lazy reading | Better questions |
|---|---|---|
| Stars | Invest everything. | Which stars are structurally defensible, and which are riding a market wave that will normalize? |
| Cash cows | Milk quietly. | What's the sustaining investment needed to keep this cow producing for another five years? |
| Question marks | Kill or double down. | Which of these has a plausible path to 1.0× share, and how much capital would it actually take? |
| Dogs | Divest. | Is this dog strategic ballast — a customer relationship, an entry point, a tax shield — worth more than its P&L suggests? |
Mistake 5: Presenting a snapshot
A single snapshot is the least interesting version of this chart. What tells the story is the trajectory — where each business was three years ago, where it is now, and where you think it's headed. Ghost the past positions, connect them with a dashed line, and the conversation shifts from "where are we?" to "where are we going, and at what velocity?"
The trajectory version is what separates a matrix that decorates a deck from one that drives a decision.
- Put relative share on a log scale, 0.1×–10×, with the 1.0× midline visible.
- Size bubbles by area (r ∝ √value), never by radius.
- Calibrate growth and share thresholds annually — and document the reasoning on the chart.
- Treat quadrants as conversation openers, not verdicts.
- Show trajectories, not snapshots — the story is in the direction and speed of travel.
Sara spent a decade as a strategy consultant at a top-tier firm before joining STRATObubbles. She writes about the tools senior teams actually use to make portfolio decisions.
