# Strategic Logic **J. Carlos Jarillo** | [[Strategy]]  --- > "The concept of 'profitability of a sector' is intrinsically erroneous, since this profitability is only the sum of the profitability of the activities." This is the line that collapses most industry analysis. We talk about "the profitability of retail" or "margins in software" as if these were meaningful. They're not. Entry barriers don't make entry to an industry difficult—they make entry to specific activities within the industry difficult. Economies of scale don't apply to the business as a whole but to particular activities within it. A car manufacturer has many production plants but typically one design centre. The minimum efficient size of designing automobiles is completely different from the minimum efficient size of assembling them. These are different businesses with different profit structures, even though they're housed in the same company. Analysing "the auto industry" obscures this. The decision of configuration—which activities a company will perform and which it will leave to others—is possibly the most important decision a manager can make. It determines what business you're actually in and therefore your a priori profit potential. Yet these decisions are often made implicitly, or treated as simple cost problems ("which is cheaper, making or buying?"), when the real question is what kind of company you want to be. --- ## Core Ideas ### [[Activity-Level Analysis]] The fundamental principle of strategic logic applies to activities within sectors, not to sectors themselves. Entry barriers—economies of scale, patents, brands—don't protect industries; they protect specific activities. Any economic activity that is conceptually separable from the rest is a distinct business with its own profitability structure. Owning a property and running a restaurant on it are two separate businesses. A restaurant might lose money whilst the landlord profits handsomely—not from gastronomic prowess but from property ownership. This reframes competitive analysis entirely. Instead of asking "is this industry attractive?" ask "which activities within this industry have sustainable singularity?" ### [[Configuration Decisions]] Which activities do you perform yourself, and which do you leave to others? This is configuration, and it shapes everything. Managers tend to sacrifice good businesses for bad ones because factories are there and need to be kept busy. There's an implicit definition of business in industry terms (computers, meat packing) that prevents analysing the profitability of each activity separately. Companies end up keeping unprofitable activities and subcontracting the profitable ones—exactly backwards. The golden rule: an activity that cannot be done better (so customers pay a premium) or cannot be done at lower costs than competitors, in a sustainable fashion, adds no value. It should be questioned. ### [[The Integration Question]] Vertical integration only makes sense if carrying out two activities within the same company increases quality or reduces costs. The mere fact that one company owns two links in a chain does not improve the profitability of either link. The trend in many industries is toward "stratification"—specialisation at the activity level rather than vertical integration. Personal computers went this way. Automobiles are going this way. Wherever some activities call for different scales or capabilities, there's pressure to unbundle. The alternative to integration isn't chaos—it's coordination. McDonald's doesn't own its restaurants but coordinates them carefully. This provides the advantages of integration (overall design, quality assurance, cross-training) without the disadvantages (lack of motivation, size imbalances, technological stagnation). --- ## Key Insights **Profitability comes from sustainable singularity.** Even in the most profitable companies, most of what they do is not singular. Profitability comes from a few things (sometimes just one) done better or cheaper than anyone else can replicate. The rest is necessary but not value-creating. **New technology doesn't guarantee advantage.** If a new technology is embedded in machines anybody can buy, it won't be the basis for sustainable cost advantage. The machine seller will make sure all potential buyers know how essential it is. Real advantage comes from technology that changes minimum efficient scale or creates barriers competitors can't easily cross. **The "race to volume" strategy often destroys value.** The logic sounds impeccable: set prices below cost, grow fast, move down the cost curve, drive out competitors, then profit. The problem is every well-capitalised competitor follows the same logic. With everyone pricing below cost and no one exiting, the business becomes ruinous for years. The winners in new industries are often the patient entrants who attack uncontested segments first. > "A sensible long-term strategy that combines a reasonable short-term profitability (necessary to survive so many years) with a slow but small improvement in the competitive position can end up being very successful." **Pretenders attack segments leaders don't care about.** Toyota started in small, inexpensive cars—a segment where customers were price-sensitive and willing to try unknown brands. The leaders' first reaction was curiosity, then dismissal: "let them have that segment, it's not profitable." But Toyota used that beachhead to build volume, brand recognition, and capabilities that eventually let them compete directly. **Mergers rarely create value unless they achieve one of three things.** First, reaching minimum efficient size (if you're subscale). Second, genuine synergies where combined activities become more valuable. Third—and this is the one nobody admits publicly—withdrawing capacity from an oversupplied market. Many mergers presented as "economies of scale" are really about reducing industry capacity so margins can recover. **Companies can buy profits but not profitability.** Acquiring a profitable business costs more than it returns—otherwise it wouldn't be profitable. Growth through acquisition without genuine synergies is value destruction dressed as strategy. "It is easy to buy volume. It is almost impossible to buy profitability." **Strategic work is discovery, not planning.** Almost all companies have golden nuggets buried somewhere—activities or customer segments that are far more profitable than average. The task is to find them through rigorous strategic logic, not to impose plans from above. Imitation of competitors, vague statements about "being leaders," and concentration on operational plans don't cut it. --- ## Connects To - [[7 Powers]] - both focus on sustainable advantage, but Jarillo emphasises activity-level analysis - [[Playing to Win]] - "where to play" maps to configuration decisions - [[Better, Simpler Strategy]] - both reject industry-level generalisations for more granular analysis - [[Competing Against Time]] - time-based competition is one way to create sustainable singularity - [[The Most Important Thing]] - the discipline of knowing what you don't know applies to activity selection --- ## Final Thought Most strategy fails because it operates at the wrong level of analysis. "We're in the X industry" already contains a category error. Industries are statistical conveniences, not strategic realities. The profit potential of any company depends on which specific activities it performs, whether those activities have sustainable advantages, and whether the configuration makes sense. The practical implication: stop asking whether your industry is attractive. Start mapping which activities within your value chain have real barriers, which don't, and whether you're on the right side of that line. Most companies discover they're holding onto activities with no sustainable advantage while outsourcing ones that could be. Strategic work isn't planning. It's discovering where your actual advantages lie—and having the discipline to configure around them rather than around inherited assumptions about what business you're in.