Business-unit strategy asks, "how does this business win against its competitors?" That's where Porter's five forces come in, along with the value chain and the moat (M4.03–M4.05). Corporate-portfolio strategy asks something else entirely: "should the parent keep owning this business at all, and if so, on what terms?" A business can be an excellent, well-run, defensible franchise and still be the wrong thing for a particular parent to own. That happens when the parent adds no advisory, capital, brand or operational value that a different owner, or the business standing alone, wouldn't add just as well or better. Goold, Campbell & Alexander call this parenting advantage (from their Corporate-Level Strategy). It puts a harder question to the parent than whether this is a good business: "are we, specifically, the best owner of it? Do we add more value here than we subtract through complexity, distraction and capital competition with our other businesses?" A center that can't answer that for each business it owns is running an accidental holding company and calling it corporate strategy.
Advertisement
How does the own, grow, fix, exit grid work?
The classic tool for turning that judgment into a decision is a two-axis grid. One axis is market attractiveness, meaning growth and structure, the Five Forces reading you already know how to run. The other is competitive/economic position. Is this business earning a real, durable spread over its cost of capital (M2.03's machinery), and does that spread look like a moat when you put it through the moat tests? Score every business the parent owns on both axes and four zones fall out. Here's the shorthand for the call each one gets:
- Own: strong position, low-growth market. It's a cash generator you keep and harvest. It should fund itself and throw off capital for the rest of the portfolio, and it shouldn't absorb more than it needs.
- Grow: strong position, attractive (higher-growth) market. Feed it capital aggressively. The parent's marginal rupee or dollar should go here first.
- Fix: weak position (often unprofitable or thin-spread) in an attractive, still-forming market. It's worth a bounded, milestone-gated bet. Don't write it a blank check, and don't exit it automatically either, because the position hasn't been proven one way or the other yet.
- Exit: weak position, unattractive market, no credible path to either strong position or attractive growth. It's a capital sink with no offsetting story. Stop funding it and, usually, sell it to someone who can do more with the assets than you can.
How should you score a business on the grid?
Two checks keep this from turning into a slide-deck ritual. First, score the position with economic profit and leave the ROIC league table out of it. M2.03's spread arithmetic, (ROIC − hurdle) × capital employed, tells you the sign (value-creating or value-destroying) and the size (how much capital is at stake) at once. Take a business earning 8% on ₹2,000 crore of capital against a 12% hurdle. It's destroying far more value than one earning 8% on ₹50 crore, even though the two ROICs look identical on a slide. Second, a negative or thin economic profit doesn't automatically mean "exit." Beginners get this wrong every time. Picture a young, fast-growing business that's below its hurdle today but on an evidenced trajectory toward it. That's a Fix. Whether it's profitable today is the wrong question. The diagnostic question is whether the loss (or thin spread) curve credibly bends before the capital runs out, and whether there's a real, falsifiable milestone that would tell you if it isn't going to. Worked Example 1 makes the distinction concrete on two businesses with different-signed but comparably ugly economics.
The annual portfolio review, run as a real process
A portfolio review that happens once, produces a two-by-two on a slide and changes nothing is theater. Run properly, it's a recurring, dated capital-allocation decision, and three rules hold it together. First, score every business on the same three axes every cycle: market attractiveness (growth, structure), competitive/economic position (the ROIC-versus-hurdle spread and its trend) and parenting fit, meaning whether the center adds value here or subtracts it. Second, the review has to produce an actual capital-reallocation decision. Dollars or rupees move from an Own or Exit business toward a Grow business. Labels alone don't count, and if nothing moves, the review wasn't real. Third, track each classification's trajectory year over year. A business drifting from Grow toward Fix, or from Own toward Exit, is more urgent news than any single year's snapshot, because it tells you which direction the spread is heading. That's M4.04's incremental-ROIC test applied to the portfolio. The finance galaxy drills the "headline versus incremental" distinction relentlessly at the business-unit level, and it applies unchanged one level up.
Does a conglomerate discount mean you should spin off a business?
A well-run portfolio review will often turn up a conglomerate discount: the market values the whole company below the sum of its disclosed parts. Measuring that discount and working out why it persists belongs to M3.10, which covers the sum-of-the-parts mechanics and the five durable causes of India's unusually deep holding-company discounts (cash that never reaches you, tax friction at two gates, agency risk, illiquidity, and no forced terminal event). Your job here is the advisory judgment on top of that valuation. A discount is a description, and on its own it isn't a thesis. What the portfolio review needs to settle is whether a specific business inside the discount is a Grow trapped under a low multiple, or an Exit the market is correctly discounting because it should be sold, discount or no discount. The Grow case is the classic spin-off candidate, since pure-play re-rating actually helps it. Confusing "there is a discount" with "therefore spin something off" is a common and expensive category error. The portfolio grid tells you which business, if any, is the candidate. The discount alone can't.