A summary of Richard Rumelt's classic strategy book
TLDR:
‘Bad Strategy’ mistakes ambition, leadership, vision, planning, target-setting, or the economic logic of competition, for strategy.
The ‘kernel’ of ‘Good Strategy’ comprises three elements: a diagnosis or problem statement, a guiding policy outlining the overarching approach, and coherent action - the steps to be delivered.
A ‘Good Strategy’ is designed to be focused, using proximate objectives to help unlock threshold effects.
A ‘Good Strategy’ focuses resources and actions on the critical factors that can magnify the impact of their efforts e.g. the weakest point in a chain-linked system.
A ‘Good Strategy’ prevails by deepening an organisation’s advantages, broadening their use, creating higher demand for advantaged products or services, or strengthening the ‘isolating mechanisms’ that block imitation by competitors.
A ‘Good Strategy’ anticipates the actions of competitors.
A ‘Good Strategy’ skillfully navigates waves of industry change - such as deregulation and escalating fixed costs - to capture new sources of advantage.
Implementing ‘Good Strategy’ successfully requires counteracting the forces of inertia (resistance to change) and entropy (the tendency toward disorder over time).
Excelling at ‘Good Strategy’ requires ‘thinking about our thinking’: learning to overcome our blind spots, question our own judgement, and resist premature closure on initial proposals.
Read on below for more detail, and case studies:
1. ‘Bad strategy’ mistakes ambition, leadership, vision, planning, target-setting, or the economic logic of competition, for strategy. It is characterised by vague language, a failure to face the critical challenges directly, and a lack of feasible immediate actions.
It often confuses strategy with the outcomes of strategy - characterising the past as merely ‘underperformance’, and setting future performance goals, without a robust underlying logic for why the desired outcome will be different from previous attempts.
Bad strategy may be the result of leaders being unwilling or unable to make choices among competing values and priorities. It may be masked by the belief that ‘all you need to succeed is a positive mental attitude’.
2. The ‘kernel’ of ‘Good Strategy’ comprises three elements: a diagnosis or problem statement, a guiding policy outlining the overarching approach, and coherent action - the steps to be delivered.
The diagnosis asks "What's going on here?", and explains the nature of the challenge. It classifies the situation as a certain type by linking facts into patterns, and drawing comparisons with analogous scenarios. Often expressed as a metaphor, the diagnosis simplifies the complexity of reality into a ‘story’ that calls attention to the most critical obstacles and opportunities.
The guiding policy specifies the approach to tackle the diagnosed challenges by creating or drawing upon sources of advantage. It serves to concentrate all effort on the pivotal aspects of the situation. This reduces complexity and ambiguity, and rules out a vast array of possible actions, without strictly defining the steps. (N.B. Within Roger Martin’s ‘Playing to Win’ framework, this ‘guiding policy’ would effectively be split into ‘Where to Play’ and ‘How to Win’)
Coherent action coordinates resource commitments to execute the guiding policy effectively. Moving from high level intentions and concepts, to the practical realities of execution, sharpens strategic ideas. This forces decisions about which issues are actually most important, and demands commitment to a chosen approach.
The ‘kernel’ is an internally consistent argument. It leads from facts on the ground to diagnosis, to an overall directive, and to action. A diagnosis is needed to evaluate alternative guiding policies. An initial view of the required action is a prerequisite for understanding whether the guiding policy can be implemented. The ‘kernel’ relegates visions, hierarchies of objectives, references to time span or scope, and ideas about adaptation and change, to second-order considerations.
3. A ‘Good Strategy’ is designed to be focused, using proximate objectives to help unlock threshold effects. It produces extra ‘power’ by combining the strategy’s different elements, and the application of that power to a single, appropriate target. A good strategy doesn't just draw on existing strength; it creates new strength through the coherence of its design, triggering a cascade of favourable outcomes.
“Having conflicting goals, dedicating resources to unconnected targets, and accommodating incompatible interests are the luxuries of the rich and powerful, but they make for bad strategy....strategy is at least as much about what an organisation does not do as it is about what it does” - Richard Rumelt
Design is the ‘engineering of fit among parts’. This imposed coherence alone can be a source of advantage, and is more necessary with greater competitive challenges. But the extent to which the design centrally imposes co-ordination is a trade-off: coordination is costly because it fights against specialisation gains, localised nuance, and reduces speed. But the design needs centralisation where the costs or benefits of actions are not borne by the decentralised actors who carry them out (i.e. a person’s actions may only impact other units, or in the future not the present). It takes policies devised to benefit the whole, at the heart of the designed strategy, to mediate this conflict.
The brilliance of good organisation is in imposing only the essential amount of coordination. For instance, vertical integration, with its need to master new business operations, would be wasteful when a company can buy perfectly good products and services from outside suppliers. However, it becomes almost essential when mutual adjustment of multiple elements is needed, and especially when there is important learning to be captured about interactions across business elements.
Threshold effects make this design and focus essential. Where these effects exist, a critical mass of effort is required to break a specific constraint, without which no meaningful impact can be delivered. These ‘returns to concentration’ arise when disproportionately larger payoffs are generated by such breakthroughs. As such, it’s prudent to limit objectives to only those where the critical level can be achieved, and to marshall all available resources at such a constraint if it is non-negotiable.
From a business perspective, it is often preferable to dominate a small segment over having an equal number of customers who represent only a sliver of a larger market. From a psychological perspective, there can be returns to concentration when people ignore signals below a certain threshold (a "salience effect") or when they believe in momentum-that success leads to success. In either case, the strategist can increase the perceived effectiveness of action by focusing effort on targets that will catch attention and sway opinion. It may, for example, have more impact on public opinion to completely turn around two schools than to make a 2 percent improvement in two hundred schools.
Proximate objectives are a tool to help organisations focus their energy and resources. They are clear and feasible, with symbolic value and strategic importance. JFK’s ‘man on the moon’ goal was a ‘proximate objective’ - it seemed audacious, galvanising a focused response, but was also carefully chosen as a target to create a contest in which the USA could easily outstrip the USSR. The full complexity and ambiguity of a situation can be daunting. But the leader’s role is to absorb a large part of that complexity and ambiguity, passing on to the organisation a simpler, solvable problem, rather than announcing ambitious goals without resolving a good chunk of the ambiguity about the specific obstacles to be overcome.
4. A ‘Good Strategy’ focuses resources and actions on the critical factors that can magnify the impact of their efforts e.g. the weakest point in a chain-linked system.
Good strategy acts like a lever. It leverages decisive asymmetries or imbalances where a relatively small adjustment can unleash much larger pent-up forces for maximum returns.
A new ‘lever’ or source of advantage can be created through subtle shifts in viewpoint. This could be by applying a robust and unique competence developed in one context to another, by identifying pent-up demand that has yet to be fulfilled, by reframing competitive dynamics, or by tapping into economies of scale (unit cost reductions through growth) and/or network effects (unit value increases through growth).
Chain-linked systems are one source of ‘critical factors’. These exist where the performance of a system (an organisation, an economy) is limited by its weakest component - a ‘bottleneck’. Where there is such a weak link, a chain is not made stronger by strengthening the others. If each link is managed separately, the system’s effectiveness will be limited (a phenomenon known as ‘quality matching’).
Conversely, a well-managed constellation of chain-linked capabilities can create a ‘flywheel’ of high performance. Such a system also makes competitive imitation difficult. Replicating only a single component of a chain-link system adds expense to the competitor's business without eroding the leader’s advantaged position - an existing rival would have to virtually start fresh and, in effect, compete with itself.
5. A ‘Good Strategy’ prevails by deepening an organisation’s advantages, broadening their use, creating higher demand for advantaged products or services, or strengthening the ‘isolating mechanisms’ that block imitation by competitors. Value creation is delivered when competitive advantage increases or when the demand for the resources underlying that advantage increases. As well as deepening an organisation’s cost and / or revenue advantages to yield more profit, ‘Good strategies’ can:
Broaden the contexts in which those advantages are used: Costs and revenue advantages are not absolute - they vary depending on the characteristics of a buyer e.g. their price sensitivity, location, preferences for certain types of product. Growth can be achieved by identifying the correct aspect of the existing value proposition and capabilities, and applying these existing strengths appropriately to new audience segments. Broadening the application of IP benefits from the fact it is enhanced, rather than spent as a result. Conversely, brands and relationships may be damaged by careless extension.
Create higher demand for advantaged products or services: Various forms of marketing, deployed strategically, can engineer higher demand for scarce resources, conferring rewards on the organisation that holds them.
Strengthening ‘isolating mechanisms’: These allow advantage to be sustained, preventing competitors from replicating it. Such mechanisms include patents, reputations, commercial and social relationships, network effects, economies of scale, tacit knowledge and skill gained through experience, and rapid iteration cycles to create an ever-changing target for imitators.
6. A ‘Good Strategy’ anticipates the actions of competitors. This involves applying strength against weakness, leveraging relative strengths to impose disproportionate costs on competitors and complicating their problem of competing with you. In this fashion, the West’s Cold War policy of ‘Containment’ used a long term financial and technological advantage to force the USSR to spend scarce resources on types of defences that didn't pose risks to the West. This anticipation also involves foreseeing the reactions of opponents, based on their culture, incentives, priorities, strengths and structure, and manipulating it accordingly, moving at a speed where their lack of time prevents them from learning new lessons and altering their predictable methods.
7. A ‘Good Strategy’ skillfully navigates waves of industry change - such as deregulation and escalating fixed costs - to capture new sources of advantage.
It is possible to gain a competitive advantage (‘high ground’) by creating it yourself through pure innovation (e.g. technical inventions, novel business models). However, during industry transitions, the ‘high ground’ changes, and by anticipating these, new sources of advantage can be captured.
These waves of disruption can fell incumbents, whose powerful resource positions which once produced profit without great effort, have also produced complacency. This manifests itself in independent fiefdoms, a lack of disciplined focus, and bolt-on acquisitions which try to provide a veneer of continued growth in the face of inevitable decline.
Causes of industry changes include escalating fixed costs (especially relating to product development, often causing consolidation), and deregulation. Regulations tend to subsidise some buyers at the expense of others, and when the systems organisations have used to justify their costs and prices become obsolete, their complexity and opacity remain in the ways of working, making them vulnerable to attack.
Such waves of industry change are typically accompanied by biases in forecasting. These include advice from industry pundits to adopt ‘copycat’ strategies based on the current highest performers, assumptions of a "battle of the titans," where the current market leaders will tussle for supremacy, undercutting the middle-sized and smaller firms. Predictions that growth trends will continue and then normalise, rather than continue and then decline, are also common. While such a prediction may be valid for a frequently purchased product, for durable products, there is an initial rapid expansion of sales when the product is first offered, but after a period of time every- one who is interested has acquired one, and sales drop to track population growth and replacement demand—the faster the uptake of a durable product, the sooner the market will be saturated.
During such waves of change, an ‘attractor state’ can be used as a navigation aid. It describes how the industry "should" work in the light of technological forces and the structure of demand. It is a point of view on how the whole sector - not a single company - will evolve toward meeting the demands of buyers as efficiently as possible, and the associated accelerants and impediments to this happening.
Where industries are not apparently changing, ‘Good Strategies’ set organisations up to behave like a chess master. The grandmaster finds positions for their pieces that (a) increase the options open to them and decrease the freedom of operation of the opponent's pieces; and (b) impose certain relatively stable patterns on the board that induce enduring strength. This translates into creating advantageous positions - focusing on the sources of and barriers to success in your sector - and building agility to seize on the next opportunity.
8. Implementing ‘Good Strategy’ successfully requires counteracting the forces of inertia (resistance to change) and entropy (the tendency toward disorder over time). An organisation’s level of inherent unwillingness to change - and therefore adopt a new strategy - will vary. Determining factors include the organisation’s age, culture, sector and location. Market forces may also drive this inertia, such as a fear of cannibalisation: a business may choose to not respond to change or attack because doing so would undermine short-term streams of profit. However, this means a rival could poach customers away without triggering a competitive response. There are five steps to breaking down organisational inertia:
Structural simplification and revised information flows to illuminate obsolete units, inefficient processes, and hidden bargains among units that mask waste, all of which were previously hidden by complex overlays of administration and self-interest.
Fragmentation of operating units which do not need to work in close coordination. This breaks political coalitions, cuts the comfort of cross-subsidies, and exposes a larger number of smaller units to leadership's scrutiny.
Triage, identifying which units have the performance and culture to be retained as part of the new structure. (N.B. Rumelt recommends using the unfortunately named ‘hump chart’ as one aspect of this triage. This chart plots the real operating profit of individual product lines, outlets, areas, segments, or any other portion of the total on a comparable basis. The largest net contributor becomes the first column on the chart. The second largest is added to the total of the first column, to create the second, and so on. If there are no cross-subsidies, the bars will rise smoothly to a maximum. But if some operations, products or locations are subsidised by others, a "hump" will emerge where the cumulative totals begin to lower).
Repair of the units which are to be retained. This can include setting challenging goals to increase urgency, and replacing the ‘alpha’ leader who sets norms and values to change the culture.
Reintegration of coordinating mechanisms once the operating units are largely working well, reversing some of the fragmentation that was initially used to break inertia.
Were inertia the whole story, an organisation with a ‘Good strategy’ would remain effective indefinitely, provided its context remained unchanged. However, the Second Law of Thermodynamics states that ‘entropy always increases in an isolated physical system’. Entropy looks like weeds in a garden, but in a business manifests itself as unfocused product lines, lacklustre pricing, slipping quality standards or complacent reward structures. This drift and deterioration must be constantly managed, with simplification and realignment efforts needed even when there is no change in strategy.
9. Excelling at strategy requires ‘thinking about our thinking’: learning to overcome our blind spots, question our own judgement, and resist premature closure on initial proposals. When faced with a question or problem to which there is no obvious answer, humans tend to welcome the first seemingly reasonable answer that comes to mind. While our instincts often produce remarkably sound judgments, they often also tell us - incorrectly - that our instincts are always correct. As such, building a set of disciplines and techniques to expand our perspectives, and resist these cognitive biases, improves our ability to design ‘Good strategy’. These include:
Adopting a ‘Create-Destroy’ mindset toward your proposed strategy, stress-testing multiple robust alternatives. Often when asked to generate more alternatives, people add one or two shallow alternatives by tweaking their initial insight, or include generic options such as "walk away," or "more study". Instead, once an initial draft has been created, actively seek to ‘destroy’ this proposal by exposing its weaknesses and internal contradictions. This ‘wargaming’ method encourages looking beyond the superficial, reconsidering core assumptions about the facts, and empathising with the perspectives of competitors, partners, end consumers, regulators and so on. If your proposal cannot withstand focused scrutiny, your strategy cannot be expected to stand in the face of real competition.
Honing in on anomalies as opportunities to learn and improve. These are facts that don't fit otherwise strong logic, or appear like edge cases to the current proposed ‘kernel’ of the strategy.
Invoking a "virtual panel" of appreciated experts to critique ideas, rather than relying on conceptual frameworks. Humans have innate capabilities for recognising, understanding and recalling personalities - this can be harnessed to test our own ideas. A mental picture can be built up of people whose judgements we respect by observing them at work if we know them personally, or by engaging with their content or with content created about them if we do not. Ideally each ‘virtual expert’ would have different areas of focus e.g. product, macro trends, technology value chain. Internal mental dialogue with this ‘virtual panel’ will both refine your own ideas and stimulate new ones, improving your thinking before sharing it with clients or colleagues.
Evaluating how the important and the actionable intersect in a given situation, to frame the most urgent next steps.
Focusing on underlying motives for choices and directions, assessing the problem statement or ‘why’ of a given situation than ‘what’ course of action is being taken.
Writing down your judgments in advance, so that you can assess their accuracy, avoid the tendency to post-rationalise your position as being essentially correct, and so improve your discernment over time.
Appendix | Case Studies:
Rumelt’s book is illustrated with a range of case studies, including:
Paccar, which specifically targeted the needs of owner-drivers of large high-end trucks, a small niche which wasn’t worth others’ investment to attack. As these drivers spent significant time sleeping in the truck, and significant time with other truckers, influencing them - rather than fleet managers - with a premium offering, created a winning position. With its designers and engineers expertise focused in this one direction, alongside its network of dealers, image, and loyal customers, its advantages were difficult to replicate.
Crown Cork & Seal, which, by focusing on short runs, smaller customers, rush orders and high touch technical service, avoided being captive to large scale beverage companies, and despite having higher costs from changeovers, drove much higher prices and profitability than competitors. It competed in the same industry as much larger rivals, but played by different rules, by concentrating on a carefully selected part of the market, specialising and increasing its bargaining power with respect to its buyers. As such, it captured a larger fraction of the value it creates.
Roll Global, which identified that, unlike small farmers, it could disproportionately increase the value of its agricultural businesses because it could afford to invest in stimulating demand (for the whole sector) by creating vertically integrated snacks and drinks businesses, and investing in processing plants.
Frank Gilbreth, whose 1909 work to double bricklayers’ productivity without increases in effort showed that incentives alone are not enough to drive performance improvements: it required reexamining the details of how work is done, and reengineering it.
General Motors’ coherent value proposition designed by Alfred P. Sloan, which set up"quality competition against cars below a given price tag, and price competition against cars above that price tag." This reduced cannibalisation and complication for customers - ‘ordinary people drove Chevrolets, the foreman a Pontiac, the manager a Buick, and the CEO a Cadillac’.
IKEA’s differentiated, capabilities, which make them difficult to compete with. The chain-linked activities form an unusual grouping such that expertise in one does not easily carry over to expertise at the others: adopting only one element of their strategy adds expense to the competitor's business without providing any real threat to IKEA.
NVIDIA which has a dedicated chapter, covering its:
Establishment of three separate development teams to reduce delays, have a best-in-class offering more often, and provide more learnings to the engineers;
Taking control of software drivers from board makers, given the sophistication required, the risk of IP leakage, and the conflicting relationships between boardmakers and chip makers. Driver development only started once working chips were received, different board makers created different drivers for the same board, and learnings were often not passed back to chip makers;
Short-circuiting of the existing distribution channels to cut a deal with Dell. In general, with a generic product, fragmented retail buyers are an advantage, while a significant buyer such as Dell can help a superior product like NVIDA’s see the light of day; and
Failure to acquire ArtX, in an industry with sparse and well-understood human capital. While Nvidia did not need the extra expertise, acquiring it would have denied these scarce competencies to a competitor.
The 2008 financial crisis’ roots in cognitive biases, such as:
Engineering overreach, where designers built systems whose workings and consequences exceeded their ability to analyse;
Smooth-sailing fallacy, where people assumed that a lack of recent tremors and storms means that there is limited risk, ignoring the possibility of a ‘threshold condition’ being reached;
Social herding, where individuals assumed that at least some others knew something they didn’t,
Where the ‘inside view’ prevailed - the tendency to ignore related pertinent data and believe that "this case is different";
While risk-seeking incentives drove unwise behaviour in organisations and individuals.
You can find out more about Rumelt's other work on this McKinsey Podcast.
Comments