BCG Growth-Share Matrix

Overview

Maps each business unit on a 2×2 grid of market growth rate vs. relative market share, revealing which units generate cash, which absorb it, and which to invest in, harvest, or exit. Four quadrants: Stars (invest), Cash Cows (harvest), Question Marks (binary decide), Dogs (exit or hold minimally). Rests on two empirical anchors: experience curve (high share = lowest cost) and industry life cycle (high growth demands reinvestment; maturity throws off cash).

Composes with: porters-five-forces to define industry boundary first · swot-analysis for internal-capability depth · ansoff-matrix to set growth direction for units worth investing in.

When to Use

• Firm operates ≥ 3 distinct business units competing for a shared capital pool
• Annual strategy or budget reviews need a forcing function for prioritization
• PE/VC portfolio requires a quick health-read across holdings; M&A teams assessing retain vs. divest
• AI capital reallocation: deciding which units to harvest to fund AI capex / AI-native bets, and whether an AI unit is a true Star or an expensive Question Mark amid AI-native competition

When NOT to use: single-product startup · highly interdependent units where divesting a Dog may destroy a Cash Cow · market in technology transition with unreliable growth data · firm-level competitive analysis within one market

Coaching Novices (Adaptive Front Door)

• Engine mode: user has specific BU data → run The Process directly.
• Coach mode: user is unfamiliar → guide step by step.

In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.

1. BCG shows which businesses fund others, which burn cash, and which need a decision — using two numbers: market growth rate and your share relative to your biggest competitor.
2. Check fit: does the user have multiple distinct units? If single-product, redirect to Ansoff or Five Forces.
3. Ask: "Which business units are you trying to prioritize?"

[WAIT — do not advance until user responds]

4. Walk through unit definition, data collection, quadrant plotting, trend analysis, and strategy assignment one step at a time.

[WAIT — do not advance until user responds]

5. Close: "The key thing BCG just revealed is [which unit is your implicit funder and which is consuming it without a clear path to self-sufficiency]."

[WAIT — do not advance until user responds]

The Process

Produce a Portfolio Map — quadrant assignments, trend arrows, and resource-allocation recommendations per SBU.

Step 1 — Define SBUs. Must: serve an identifiable customer group, have identifiable competitors, be manageable with resource independence. Stop rule: if you cannot name the primary competitor, the boundary is wrong.

Step 2 — Market growth rate. 2–3 years external data; calculate CAGR. Dividing line: 10% (raise to 20–30% for AI/clean-tech). Never use own revenue growth as a proxy.

Step 3 — Relative market share. Own share ÷ largest competitor's share. >1.0 = leader; <1.0 = follower.

Step 4 — Plot. X-axis: relative share (log, right = high); Y-axis: growth (linear, up = high); bubble size = revenue. Assign quadrant.

Step 5 — Trend arrows. 2-year trajectory per SBU. Trend often matters more than current position.

Step 6 — Strategy. Star: invest aggressively. Cash Cow: extract surplus; minimize capex. Question Mark: binary — upgrade to Star OR exit by a named date. Dog: harvest/exit; hold only if synergy is named and quantified.

Output Template

BCG Portfolio Map: <company> | Threshold: <X>% | Date: <date>
SBU | Growth | Rel.Share | Quadrant | Revenue | Profitable?
Trend: <SBU> moving <from> → <to> — reason: <…>
Cash generators: <list> | Cash absorbers: <list> | Balance: <surplus/deficit>
Strategy: <SBU A>: invest/harvest/exit by <date>
Key decision: <what the analysis forces>

→ Method in Action: Procter & Gamble's Brand Portfolio Restructuring (2012–2016) · GE's "Fix, Sell, or Close" Pruning (1981–1995) → 2026 lens: Microsoft's Portfolio as AI Reallocates Capital (2024–2026) — which units are Stars, which Cash Cows fund the AI capex build, which are Question Marks or Dogs

Portfolio Packs

→ Primary sources: references/sources.md

Common Rationalizations

[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.

Red Flags

• Market boundary defined after desired quadrant assignment is known · relative share uses own revenue not competitor's share · all Question Marks described as "likely Stars" with no exit criteria · no trend arrows · Dogs retained with unquantified synergy · Cash Cows don't cover Stars + Question Mark investment needs · matrix used as a slide with no reallocation following

Verification

• Each SBU passes the standalone-manager test (identifiable market, rivals, separable P&L)
• Relative share = own share ÷ largest competitor's share (not absolute share)
• Market growth from external data, not own revenue growth
• 2-year trend arrows plotted for each SBU
• Cash Cow generation quantified against Star + Question Mark investment needs
• Each Dog has an exit plan or a named, quantified synergy
• Each Question Mark has a decision deadline with invest-or-exit criteria
• Refresh cadence scheduled (state the date)

Part of deciqAI Knowledge Skills — 253 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. See it run → https://www.deciqai.com/s/bcg-matrix · Built by deciqAI · github.com/deciqAI · Contributions welcome.

Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/bcg-matrix.json