Book Digests/General Business

Book Digest: Good Strategy/Bad Strategy

Posted on by Mike Pinkel

In Good Strategy/Bad Strategy: The Difference and Why it Matters, Richard Rumelt argues that the essence of strategy is discerning the one or two critical issues in a business problem and concentrating effort on them. Developing a strategy requires making a clear diagnosis of the problem, defining an overall approach to solving it, and creating a set of mutually reenforcing coherent actions to carry out the overall approach.

Rumelt believes that most organizations either have a bad strategy or no strategy at all. This is because good strategy requires prioritization, which requires saying no to internal constituencies advocating for their preferred projects. It also requires hard analytical work that is much more demanding than the standard routine of setting goals, creating incentives, and fostering a shared vision.

His book is a guide to performing this kind of analysis.

Introduction

  • The essence of strategy is identifying the one or two critical issues in a situation: the pivot points that multiply the effectiveness of effort
  • Focus effort on these pivot points
  • Good strategies:
    • Honestly acknowledge challenges and provide a way to overcome them
    • Include a set of coherent actions
  • Bad strategies:
    • Skip over problems
    • Accommodate a multitude of conflicting objectives rather than choosing one or two
    • Cover up their shoddy thinking with ambitious platitudes, like a quarterback who only says “let’s win”
  • The kernel of good strategy:
    • Diagnosis of the challenge
    • Guiding policy
    • Coherent action

Part I: Good and Bad Strategy

  • Strategy is the application of strength against weakness
  • Strategy creates new strengths through subtle shifts in viewpoint

Chapter 1: Good Strategy is Unexpected

  • Most organizations have multiple conflicting objectives that amount to little more than spend more and try harder
  • They spread resources to placate internal interests rather than concentrate resources to focus on pivot points
  • Creating a good strategy means saying no to a variety of interests

Chapter 2: Discovering Power

  • Success isn’t just about what your organization does, it’s also about blocked or failed competition
  • For example, in setting cold war strategy, it made little sense to match Soviet capabilities. The best method was to build on our strengths in ways that were aimed at their weaknesses.
  • Act so as to impose exorbitant costs on the competition using your relative advantages

Chapter 3: Bad Strategy

  • Bad strategy has four hallmarks
    • Fluff: Unnecessarily long words
    • Failure to face or define the challenge
    • Mistaking goals for strategy. You can’t just state your desires, you need to plan to overcome obstacles.
    • Bad strategic objectives. Good ones are means to an end, bad ones fail to address key issues or are impracticable.
  • Bad strategy has lots of goals but no policy or action
  • Fluff
    • “Our fundamental strategy is one of customer-centric intermediation” is fluff, it’s just another way of saying that you’re a bank
  • Failure to face the problem
    • If you fail to identify and analyze the obstacles, you don’t have a strategy. You have a wish list.
    • You must offer a theory of why things haven’t worked in the past
    • International Harvester failed to face its problem: high costs
    • DARPA has a strategy with specific policies around obstacles, such as retaining managers for only four to six years to stop inertia and talent stagnation
  • Mistaking goals for strategy
    • “We want to grow revenues at more than 20% per year with margins at 20% or more” is mistaking goals for strategy
    • Trying harder isn’t enough. The cult of the offensive in the First World War led to pointless and wasteful charges into machine guns under the misguided theory that victory could be achieved by simply having more enthusiasm and commitment
    • Strategy requires a point of leverage or a key strength to exploit, some sense of the one or two most critical opportunities to increase growth
  • Bad strategic objectives
    • Dog’s Dinner (Bad) Objectives
      • Good strategies focus on one or two objectives, bad ones sweep together 143 items on a to do list that encompasses the wish list of every department or constituency in your company
    • Blue Sky (Bad) Objectives
      • A simple statement of the goal with no attention to how it will be achieved or the difficulties of achieving it
      • Underperformance is not a challenge; the challenges are the reasons for the underperformance
      • LAUSD example of blue sky objectives: Asking for transformational leadership from principals amidst a stultifying bureaucracy and a history of managerial incompetence
    • Examples of good objectives:
      • Dividing customers into three tiers based on importance and setting specific objectives for each tier, like maximizing prime shelf space with the top tier and getting some shelf space with the bottom tier
      • Linking small stores together to maximize their purchasing power and fight back against a large chain

Chapter 4: Why so Much Bad Strategy

  • Bad strategy isn’t miscalculation, it’s avoiding the hard work of crafting a good strategy
  • Good strategy means making priority choices
  • Sources of bad strategy:
    • Avoiding choices
    • Vision templates
    • New Thought: The Power of Positive Thinking
  • Inability to choose
    • Opting for all the options on the table or a consensus with something for everyone means that nobody sharpens their arguments or analysis
    • Major choices involve reshaping the firm. You can only reshape in one way.
    • Scarcity creates a need to focus, focus is the core of strategy
    • Any strategy involving focus will disappoint some constituencies inside the company; often this is hard to do unless the company is under threat
  • Template style strategy
    • These are a company vision that inspires people to change and empowers people to accomplish that vision
    • All this inspiration isn’t the same as making choices to focus on the right objectives; therefore it shouldn’t be confused with strategy
  • New Thought
    • This is idea that positive thoughts help you achieve outcomes. Examples include The Power of Positive Thinking, Tony Robbins, Christian Science
    • A focus on “shared vision”
    • Success doesn’t come from New Thought. Ford didn’t have a unique vision of building a car for the masses nor was his vision particularly shared with the assembly line workers; what he had was the engineering skill to make it happen

Chapter 5: The Kernel of Good Strategy

  • The kernel of good strategy has three components:
    • Diagnosis that explains the challenge and highlights its most important aspects
    • Guiding policy that forms the overall approach to overcome the obstacle
    • Coherent actions that reinforce each other to accomplish the guiding policy
  • The diagnosis:
    • Finds patterns in the problem
    • Suggests that some aspects of the problem are more critical than others
    • Defines a domain of action
    • Example: Kennan defines the cold war as a long-term struggle against the Soviet Union without any possibility of a negotiated settlement
  • The guiding policy
    • Outlines the overall approach to handle the situation
    • Rules out many potential actions
    • Includes a sense of “why” you’re engaged in the specific action
    • Plays to a particular advantage you have
    • Creates advantage by:
      • Anticipating the actions of others
      • Reducing complexity and ambiguity
      • Exploiting the leverage that comes from concentrating effort on a pivotal point
      • Creating actions that are coherent and coordinated, meaning that they build on one another rather than cancel one another out
    • Example: Kennan’s policy of containment; keeping the USSR contained until its own internal contradictions brought it down
  • Coherent actions:
    • Coordination, by itself, is a source of advantage

Part II: Sources of Power

  • A good strategy works by harnessing power and applying it where it will have the greatest effect
  • The fundamental sources of power:
    • Leverage
    • Proximate objectives
    • Chain-link systems
    • Design
    • Focus
    • Growth
    • Advantage
    • Dynamics
    • Inertia
    • Entropy

Chapter 6: Using Leverage

  • Focus, channelled at the right moment onto a pivotal objective can create a cascade of favorable outcomes
  • Sources of strategic leverage
    • Anticipation
    • Pivot points
    • Concentrated application of effort
  • Anticipation
    • Consider the habits, preferences, and policies of others together with inertias and constraints on change
    • Focus on the predictable downstream effects of changes that have taken place
    • Example: Toyota building hybrid engines because they foresaw that higher oil prices would create demand for greater fuel economy
  • Pivot Points
    • Pivot points magnify the effects of focused energy; a small adjustment can unleash pent up energy
    • Example: Seven & I Holdings
      • In Japan, customers want variety, so they have hundreds of food varieties that constantly change
      • In China, customers want service, so their stores are spotless
  • Concentration
    • Returns to concentration happen when focusing on fewer things generates large payoffs
    • Threshold effects are in evidence when you need to hit a certain critical mass to see a return
    • In advertising, companies often “pulse” ads so they overcome thresholds any one moment
    • Dominating a very small market can be better than having some presence in many
    • Most organizations can only focus on a few critical problems at a time
    • Example: Harold Williams ran the Getty Trust with a clear focus. Instead of just buying more art, he decided to transform the study and practice of art with catalogues, educational efforts, and research.

Chapter 7: Proximate Objectives

  • A proximate objective names a target that the organization can hit or even overwhelm
    • Example: Engineers can’t work without a specification, so set something determinate as a goal even if you’re not exactly sure what you need
    • The US decided that the space race was about the moon, which was the objective we were better placed to achieve
  • Proximate objectives are simple and defined. Leaders construct them this way even if the organization’s true challenges are complex and ambiguous
    • Example: Chess masters are trying to win the game, but they’re thinking about more proximate objectives like “improving their position” when they make specific moves
  • Proximate objectives cascade downwards. The top level defines goals and then proximate objectives for itself. These proximate objectives become the goals of the next level and the next level sets proximate objectives based on these goals.

Chapter 8: Chain-Link Systems

  • Systems have a chain-link logic when performance is limited by the weakest subunit, or link
  • Improving a single link does not improve the results of the system as a whole unless the weakest link is improved
  • If all the links are weak, improving one of them won’t show results until the others are also improved
  • Merely creating more pressure for higher profits wouldn’t solve the problem because multiple areas under the control of different teams each need to be improved before the results show up in increased profits
  • Therefore, a sequential strategy works best in which leaders take responsibility for the final results and focus on one link at a time. The payoff doesn’t come until the very end when all links have been improved.
  • Effective chain-link systems have tremendous advantages because they’re hard to duplicate. For example, IKEA integrates all aspects of furniture production and sales into a cohesive experience that’s very difficult to replicate.

Chapter 9: Using Design

  • Many good strategies today are more designs than decisions, in other words you’re building something not making a choice
  • In design problems, getting the right combination of elements that complement each other produces massive gains, being slightly off imposes massive costs
  • Tightly integrated designs are very effective but are harder to create, narrower in focus, and less flexible in the face of change
  • Strategic resources are those long lasting and cannot be duplicated without creating a net economic loss
  • Many upstart companies have a tightly integrated strategy that unseats an incumbent

Chapter 10: Focus

  • Focus is attacking a segment of the market with a system supplying more value to that part of the market than anyone else can
  • To find a company’s focus, look at its distinctive policies and ask what all of those policies are aimed at
  • Example: Crown Cork & Seal has fast response times and great technical assistance so they can do small production runs on short notice. This raises their costs but means that they’re the only bidder for companies that need a fast, small production run. This increases their bargaining leverage and their profits.
  • Many large companies have no strategy and no focus

Chapter 11: Growth

  • Growth by acquisition usually results in paying too much
  • Example: Crown Cork & Seal later grew by acquisitions, which raised revenue but depressed margins. They abandoned focus in favor of illusory economies of scale and suffered for it.
  • Example: Telecom Italia considered buying a telecom operator on another continent. There was no economy of scale because the two companies operated in different areas and was no operational logic because they could have signed a contract to get the other company to carry their traffic. The merger would have been a disaster.
  • Healthy growth is a response to demand for special capabilities
  • It should be accompanied by superior profits

Chapter 12: Using Advantage

  • Advantages are rooted in differences with opponents
  • Identify which differences are most important and maximize their impact
  • Example: A startup with smart scientists is good at making a new kind of fabric; innovation is their advantage. It is not good at creating a clothing line. Once the technology is created it’s time to sell it off to a company that specializes in bringing clothing to market.
  • Press where you have the advantage, avoid situations where you don’t
  • The nature of advantages
    • They either deliver lower cost or greater perceived value
    • Advantages are often particular to a context, such as a location or a buyer persona
    • Sustained advantages have to be hard to duplicate; as with a network effect or a brand
  • Interesting advantages are ones that you can increase with strategic skill, like growing a fruit that you can market more effectively using your special marketing skills. Boring advantages are just a given that cannot be optimized, like a machine that simply produces silver.
  • There are four mechanisms for increasing the value of an advantage:
    • Deepening advantages
    • Broadening the extent of advantages
    • Creating higher demand for advantaged products
    • Strengthening the isolating mechanisms that stop others from replicating the product or advantage
  • Deepening advantages:
    • Advantage is a surplus between buyer value and cost; you can widen advantage by increasing buyer value or lowering cost
  • Deepening advantage can fail when:
    • Management thinks it’s natural or can be created with incentives alone
      • Example: Gilbreth redesigned the process of laying brick to increase productivity. He had to change every stage of the process; merely incentivizing employees who worked on one part or another to work harder wouldn’t have improved the entire system
      • Improvements come from reexamining the details of how the work is done, not just incentives
    • Isolating mechanisms stopping others from replicating the advantage are weak, causing everyone in the industry to attempt to free ride on the development
  • Extending an advantage requires looking at the skills and resources that create the advantage and asking what other markets of customers they could serve
    • Be rigorous when defining your advantage; don’t pretend that it’s something fuzzy like “management”
    • Examples of effective broadening:
      • DuPont uses chemical knowledge from one area to expand to others
      • Disney defines the core values of its brand by looking at past movies and realizes that these can apply beyond the confines of cartoon movies for kids
  • Creating higher demand
    • Example: Redefining pomegranate juice as health food and creating a brand around it
  • Strengthening isolating mechanisms
    • Creating a moving target through constant product developments

Chapter 13: Using Dynamics

  • Exploit a wave of change by seeing its effects before others do
  • Waves of change are exogenous: They occur because of reasons outside your company
  • Most industries are stable most of the time; without change there may be few strategic opportunities
  • Look for a present effect, understand the forces underlying the effect and the second-order changes that these forces will create
    • Example: The rise of TV was going to create free entertainment (first-order). So (second-order) B movies were out. So the studio system, which was good at producing B movies, was out. How do you make better movies? Independent producers and production companies with studios in the role of distributor.
    • Example: Cisco builds world-beating routers because of their embedded firmware designed by perhaps five people. This let them conquer the market because it harnessed an industry-wide shift from economies of scale to the centrality of one brilliant engineer, which resulted from the centrality of software rather than hardware.
    • Example: Cisco rode key waves: The primary of software, corporate data networking, IP networks, and the explosion of the public internet. Competitors were each attached to their own proprietary networking standards and therefore couldn’t adapt.
    • Example: The rise of software as being more important than hardware. Microprocessors are important because you can customize their uses with software. Software engineers are no less expensive than hardware engineers, but it’s much easier and cheaper to iterate on software ideas. You can discover this sort of contrast by stating the question “Is software cheaper than hardware” in a specific context, such as whether a jet engine manufacturer would rather create fuel monitoring equipment with proprietary hardware or proprietary software and thinking through the process of making improvements to both software and hardware.
  • Mental guideposts for predicting future changes:
    • Rising fixed costs
    • Deregulation
    • Biases in forecasting
    • Assessing incumbent responses to change
    • Attractor states, the future evolution of an industry based on drive for overall efficiency
  • Rising fixed costs
    • Usually creates consolidation because only large players can afford to compete
    • Example: The Photographic Film Industry
      • R&D costs shot upwards when the film industry moved from black and white to color
      • Black and white was well understood; color was not and therefore required vastly more research to perfect
      • This forced smaller players from the market because they couldn’t afford the R&D
    • Other Examples: IBM’s rise was partly because of the escalating R&D costs for computers in the 1960s. The aviation industry consolidated when jet engines replaced piston engines because only the big players could fund the development of new jet engines.
  • Deregulation
    • Regulated industries subsidize some buyers at the expense of others
      • Regulated airline prices helped rural travelers at the expense of urban ones. When deregulation took place, the subsidies diminished quickly.
      • But habits based on the subsidies continued for years because of inertia and the difficulty of getting internal accounting adjusted for the new environment.
  • Predictable Biases
    • These will be present during changes and lead to predictable errors
    • When sales grow rapidly, everyone expects it will continue. But this is unlikely to continue if the good being sold is a durable good. In that case, sales will rise, then fall to the low rate needed to replace purchased inventory.
    • When there is change, everyone expects a battle between the large players. But sometimes a smaller competitor will address the change more effectively.
    • When there is change, most consultants will suggest that you adopt the response of the highest valued companies in the industry. They may miss the wave due to inertia and sticking to their previously successful strategies.
  • Attractor States
    • Attractor states embody the idea that an industry should evolve toward greater efficiency in light of changes in technology and demand
    • Example: Cisco’s vision of “IP Everywhere” was an attractor state. It was more efficient for buyers and eliminated the higher margins associated with multiple proprietary network standards.
    • Attractor states are modified by accelerants and impediments
    • Demonstration effects are an accelerant. Napster showed that digital music was possible, even though it had been for a while
    • Regulation is an impediment. We should have nuclear power but regulations make it impossible to build in the US.

Chapter 14: Inertia and Energy

  • Inertia categories:
    • Routine
    • Culture
    • By Proxy
  • Routine
    • Airline deregulation: Even after deregulation ended, airlines used the same rules of thumb they’d used during the period of regulation
    • Assumptions like “we’ll always make 12% return on capital” may be baked into decision processes but no longer apply in the new environment
    • Fix these by changing the perceptions of top management
  • Culture
    • Even after AT&T lots its position as a regulated monopoly, it’s old noncompetitive routines stayed in place even after management wanted to shift the company to a competitive footing
    • Bell Labs simply refused to create demonstration code; they did basic research
    • Fix this by simplifying the organizational structure. This will illuminate the inefficiencies hidden in the system. It may be necessary to fragment the group structure into smaller pieces.
    • To change group norms, you may need to change the alpha member of the group.
  • Inertia by Proxy
    • Organization may fail to adapt because it wants to preserve old profit streams, thus taking on its customers’ inertia

Chapter 15: Putting it Together

  • Nvidia’s rise is a story of great strategy: Diagnosis, guiding policy, coherent action
  • Violent video games created a need for 3D power in desktops; users needed maximum computing power to compete in online matches
  • Giving a buyer a competitive advantage leads to fast uptake
  • Nvidia shifted from multimedia to gaming and judged that the more power they could provide the more the market would buy
  • Nvidia broke out of 18 month development cycles in favor of six month cycles, meaning that they were the newest product most of the time
    • This involved creating three development teams working in parallel
    • Bad drivers created development risk, so they took control over drivers
    • They outsourced fabrication; it wasn’t essential to control
  • They sold to computer makers rather than board makers; concentrated buyers helped them grow from a marginal position
  • Nvidia’s speed persuaded others to try to do too much and they failed

Part III: Thinking like a Strategist

Chapter 16: The Science of Strategy

  • Good strategies are hypotheses about what will work
  • Strategy is inductive, not deductive. You don’t already know all the basic principles you need in order to solve the problem.
  • Hypotheses may arise from intuition. The best can be tested without excessive investment.
  • If the strategy requires tight coordination among many different elements, it may be necessary to own each element rather than contract it out

Chapter 17: Using Your Head

  • Asking the right questions can be shockingly novel and useful: What is the purpose of your division or company?
  • Focus on the intersection between the important and the actionable
    • Rise above myopic focus on immediate needs
    • Make a list of the ten most important things you can do and start at the top. Not a to do list, a priority list.
    • Priority lists are helpful because we don’t naturally choose the things that are most important, we often forget our key priorities in the press of the day to day
  • Developing strategic ideas
    • People usually grasp the first insight that comes to mind
    • Initial insights are good, but we’re usually overconfident in them
    • We want to end the uncertainty associated with not yet having a decision
    • Common errors in intuition:
      • Our competence relative to others
      • Cause and effect
      • Probability of a future event
      • Overweighting vivid examples as opposed to broad statistical evidence
    • To avoid these sorts of errors, (1) develop tools that force you to question your judgment and (2) record your judgments so you can analyze them in the future
      • The Kernel of good strategy: (1) diagnosis, (2) guiding policy, (3) coherent action
      • Problem-solution: force yourself to tie your action to a diagnosis by framing strategic analysis in the form of problem-solution
      • Create-Destroy: Force yourself to question prior judgments by actively trying to destroy your initial intuitions
      • Commit to a judgment: Force yourself to decide what is most important and write it down. Don’t merely identify a long list of potential options; if that’s all you do you can always convince yourself that you spotted the right issue. If you write your judgments down, you can evaluate them later and improve your judgment.

Chapter 18: Keeping Your Head

  • Bad strategy follows the crowd, substituting slogans for insights
  • Be independent without being eccentric and doubting
  • Two stories illustrate how to do this:
  • Global Crossing
    • Everyone liked their economies of scale in moving data, everyone believed that data needs were destined to grow exponentially
    • “Analysis” became a comparison of stock market performance rather than thinking through their business model. Global Crossing’s stock did quite well.
    • Their business model was flawed: data hauling was a commodity product, competitors had access to the same technology, capacity was growing even faster than demand, and the cost of the infrastructure was sunk, so it was rational to price at near zero if a bidding war started
    • Complex systems can’t be judged based only on their internal logic, you have to look outside the system to see if it makes any sense. Don’t just listen to the market, test the logic.
  • The Mortgage Bust
    • A result of social herding and the inside view
    • Asset bubbles involve a rising price of an asset, which people borrow against to buy more of the asset. When someone realizes that the asset is overvalued, there’s a run on the asset and prices decline, meaning that none of the loans can be paid back because the borrowers are bust and the collateral is worthless.
    • This has happened many times in American history and all over the world
    • It was totally foreseeable that it would happen again if you took the “outside view” – looking at the system from the perspective of a non-participant and using examples from similar situations to judge the probable outcome
    • But if you followed social herding – abdicating your judgment to the crowd – you would not have predicted it
    • Similarly, if you adopt the inside view – we’re different than everyone else – you wouldn’t have predicted it
    • As another example, we all know that talking on a cell phone increases accident risk by a factor of five in the population as a whole (the outside view) but we all believe that we’re such good drivers that it doesn’t affect us (the inside view)
Mike Pinkel