From Vision to Competitive Advantage: How Organizations Formulate and Execute Strategy

A vision states where an organization wants to be. A strategy is a set of choices about how to get there: which customers to serve, what to offer them, and which activities to perform differently from rivals. Execution is what turns those choices into behaviour. It includes budgets, structures, targets and hiring, and also the many small decisions made by people who never attended the strategy retreat. Article 06 examined how strategies are formed. The concern here is what happens afterwards.

Kaplan and Norton (1996) located the problem in management systems. Most control systems, they argued, are built around financial measures and targets that bear little relation to progress on long-term objectives, so the emphasis on short-term results "leaves a gap between the development of a strategy and its implementation." The diagnosis suggests a remedy: change what is measured and reviewed, and behaviour will follow the strategy.

Separating formulation from execution is useful because it distinguishes two failures with different remedies: a strategy can be wrong, or a sound one can be left on paper. The separation also has critics. Article 06 discussed Henry Mintzberg's objection that realized strategies are rarely intended plans carried out as written. Roger Martin (2010) went further, arguing that the habit of treating senior executives as a "choosing brain" and everyone else as "choiceless arms and legs" is itself a cause of failure. In his alternative, choices cascade down the organization, each level making the choices that belong to it. An upstream choice works when it is explained, when the next choice downstream is named, and when it is revisited once feedback arrives.

These critiques do not abolish the distinction, but they change what "execution" should mean. It becomes less the faithful delivery of a plan and more a continuing process of translating, testing and revising choices across levels. The tools discussed below look quite different depending on whether an organization uses them to enforce a plan or to learn from it.

Writing on strategy execution often opens with a number: 70, 80 or 90 percent of strategies are said to fail in implementation. Carlos Cândido and Sérgio Santos (2015) traced these figures back to their sources.

The 90 percent figure, quoted by Mintzberg (1994) and by Kaplan and Norton (2001), goes back to two pieces by Walter Kiechel published in 1982 and 1984, the first in Fortune, based on interviews with consultants, executives and theorists. Cândido and Santos found no research instrument or method described. Another estimate of about 90 percent (Judson, 1991) rested on Gray's (1986) questionnaire of 300 planners and executives; recalculating from Gray's own results, Cândido and Santos put the implied failure rate at 51 percent. The 70 percent figure has a stranger history. Charan and Colvin (1999), writing in Fortune about several dozen chief executives who had failed, estimated that in about 70 percent of those cases the real problem was bad execution, not bad strategy. That is a claim about why some failed CEOs failed. According to Cândido and Santos, Kaplan and Norton (2001) misread it as a finding that 70 percent of strategies fail.

Their conclusion is careful. Implementing a new strategy is widely accepted to be difficult, but "the true rate of implementation failure remains to be determined," because most estimates rest on evidence that is "outdated, fragmentary, fragile, or just absent." They still judge that failures are considerably more common than would be desirable. Even Sull, Homkes and Sull (2015), whose research is discussed below, open by saying that "studies have found" two-thirds to three-quarters of large organizations struggle to implement their strategies, without naming the studies.

No replacement figure is offered here, because none of the sources supports one. Two practical points follow. "Failure" needs a definition: an abandoned initiative, a target missed by 10 percent and a strategy delivered as planned but unprofitable are different outcomes. And anyone who quotes a failure rate to justify an expensive execution programme should be able to say where the number came from.

Michael Beer and Russell Eisenstat (2000) studied the problem from inside companies. In their method, organizational fitness profiling, a senior team appoints a task force of respected managers one or two levels below it, the task force interviews colleagues about what blocks the strategy, and the senior team hears the findings in a structured meeting. Of the profiles conducted in 12 companies, covering more than 150 units, the authors examined 12 in depth from four companies. Six barriers came up most often. They called them "silent killers" because they "are rarely publicly acknowledged or explicitly addressed."

What stands out is where the list places responsibility. At least three of the six barriers start with the senior team, and an ineffective senior team appeared in every profile examined. Structure, systems and HR policies were mentioned, but less often. The sample is small and drawn from companies willing to put their leaders through the process, so the counts are not estimates of how common each barrier is. What the study shows is a pattern that a careful diagnostic can bring into the open.

Donald Sull, Rebecca Homkes and Charles Sull (2015) reached a related conclusion by a different route: a survey of 7,600 managers in 262 companies across 30 industries, mostly large firms in volatile sectors, a third of them based in emerging markets. Their respondents tended to equate execution with alignment, meaning objectives cascaded down the hierarchy and rewards tied to them. The data pointed elsewhere. Vertical relationships worked well: 84 percent of managers said they could rely on their boss and their direct reports all or most of the time. Lateral ones did not. Only 9 percent said they could rely on colleagues in other functions and units all the time, and about half most of the time. Understanding was thinner than the volume of communication suggested. Only 55 percent of middle managers could name even one of their company's top five priorities, and only 11 percent believed that all their company's priorities had the financial and human resources they needed.

When execution stalls, the authors observed, managers reach for the tool they know best and tighten alignment further, so that companies become "trapped in a downward spiral in which more alignment leads to worse results." The survey measures managers' own perceptions, and the respondents were nominated by their companies, so it says little about small firms or public agencies. Read with Beer and Eisenstat, though, it shifts attention from unmotivated staff and missing KPIs towards the senior team and the boundaries between units. Sull and colleagues themselves list the Balanced Scorecard among the tools managers use to tighten alignment up and down the hierarchy.

The Balanced Scorecard is usually taught as a four-box template. Its history shows what it was meant to do. Measuring non-financial performance was not new: in the 1950s a General Electric staff group recommended measuring the company's decentralized divisions on one financial and seven non-financial metrics (Kaplan, 2010). The scorecard itself grew out of a multi-company research project on performance measurement that Kaplan and Norton ran in 1990 with the Nolan Norton Institute, and it was introduced in a 1992 Harvard Business Review article. Its original structure kept financial results as the ultimate outcome and added three perspectives proposed as drivers of long-term value. Each answered a question: how the organization should appear to customers, which internal business processes it must excel at, and how it will sustain its ability to change and improve.

The shift towards execution came from users. By 1996 Kaplan and Norton reported that none of the more than 100 organizations they had studied or worked with had introduced its first scorecard intending to build a new strategic management system, yet the executives in each found that the scorecard gave a focus to goal setting, business planning, capital allocation and review. They described four management processes built around it: translating the vision into agreed objectives and measures, communicating and linking those objectives to departments and individuals, integrating business and financial planning, and feedback and learning. Larry Brady, then president of FMC Corporation, put the point bluntly in an interview the authors quoted: the scorecard "should be the core of the management system, not the measurement system."

Later work added strategy maps (2000 onwards), which Kaplan (2010) says every scorecard project now builds before choosing any measures, and in 2008 a closed-loop management cycle whose final stage tests and adapts the strategy itself. Kaplan (2010) used Robert Simons's distinction between diagnostic control systems, which check performance against targets, and interactive ones, which senior managers use to debate data and assumptions face to face. Used diagnostically, a scorecard is a broader variance report. The chief executive of Mobil's US marketing and refining division described the alternative: "In the past we were a bunch of controllers sitting around talking about variances. Now we discuss what's gone right, what's gone wrong."

Every arrow in Figure 1, which follows Kaplan's (2010) own example, is a claim about cause and effect, and Analog Devices had shown that the chain can break at the customer link.

Hanne Nørreklit (2000) argued that the scorecard's causal claims are weaker than its presentation suggests. As Bukh and Malmi (2005) summarize her critique, she argued, first, that causality requires a time dimension the scorecard does not specify; second, that some proposed links, such as customer satisfaction leading to financial performance, do not hold in general; and third, that the four perspectives are not independent of one another. Her conclusion, as quoted by Bukh and Malmi, was that the scorecard "makes invalid assumptions about causal relationships, leading to the anticipation of performance indicators which are faulty, thus resulting in dysfunctional organizational behavior and sub-optimized performance." She also questioned the model of management control the scorecard assumes.

Bukh and Malmi accept part of this. Kaplan and Norton never defined causality, and their examples read more like means-and-ends logic. But the authors argue that in practice the links are treated as hypotheses about one firm at one time, not as general laws, and that firms they had worked with built time lags into their targets: so many new customers this year to reach a revenue level next year. Both positions point to the same discipline. A strategy map is useful when its arrows are treated as bets to be tested, and misleading when they are treated as proven.

Kaplan's own account supports that caution. He described most studies of where the scorecard succeeds or fails as "ad hoc correlations," and his claim that leadership may be "both necessary and sufficient" for success was offered as "a bold hypothesis" drawn from successful implementations. Evidence collected from successes cannot show how often the same practices fail. He also acknowledged that for many years an executive's description of the learning and growth perspective as "the black hole of the Balanced Scorecard" was apt. Companies measured employee satisfaction, turnover and absenteeism, but none had metrics linking employee capabilities to the strategy. For a human capital audience this is the central point. The perspective that represents people sits at the root of every causal chain, and it has had the weakest measures.

Where the scorecard asks what to measure, the 7-S framework asks what has to fit together. It was set out by Robert Waterman, Thomas Peters and Julien Phillips, then McKinsey consultants, in an article whose title states its argument: "Structure Is Not Organization" (Business Horizons, 1980). Their target was the reflex of answering every strategic problem by redrawing the organization chart. Structure, they argued, is one of seven interdependent elements: strategy, structure, systems, style, staff, skills, and what they called superordinate goals, later renamed shared values. Systems are the procedures through which work actually flows, such as budgeting, reporting and rewards. Style is how senior managers spend their time and what they signal.

The execution logic is simple. A change in one element that the others do not support tends to be absorbed or reversed. Strategy and structure can be changed on paper quickly, but in the authors' view the pace of real change depends on all seven elements. Teaching treatments often divide the seven into "hard" elements that can be changed by decision (strategy, structure, systems) and "soft" ones that change slowly (style, staff, skills, shared values). The original authors argued that the softer elements could be observed, and even measured, if managers took them seriously.

Applied to the evidence in Section 3, the framework explains why a cross-unit strategy rarely succeeds through a new structure alone. If a bank wants its corporate and retail divisions to serve the same family businesses jointly, a new reporting line will achieve little while bonuses reward each division's own book (systems), division heads compete for the chief executive's attention (style), and relationship managers lack the skills to sell outside their own products. The framework's limits are just as clear. It lists interdependencies without saying which configuration works, offers no measures and no sequence, and describes fit at a single moment while strategies change. A strong emphasis on fit can also become an argument for leaving everything as it is.

Comparison of the two frameworks. The "blind spot" and "failure mode" rows are this article's interpretation, not claims made by the frameworks' authors.

Bangladesh's power sector offers a real, dated example of a strategic target that was met while the system needed to serve its purpose lagged behind. The sector is a group of public agencies and private producers, not a single firm, so what follows illustrates the measurement problem discussed above rather than accounting for the sector's decisions.

The headline goal was electricity access for every household. On 21 March 2022, inaugurating the 1,320 MW Payra coal-fired plant, Prime Minister Sheikh Hasina announced that every house in the country now had access to electricity. The Daily Star reported that coverage had risen from 47 percent in 2009, and generation capacity from 4,942 MW in January 2009 to 25,514 MW in February 2022. Officials qualified the claim: coverage meant that anyone who applied could be connected, through the grid or a solar home system. By the measures chosen, the vision had been achieved.

Four months later, after global energy prices rose following Russia's invasion of Ukraine, the government halted all diesel-fired plants to reduce spending on power generation. It accepted a shortfall of 1,000 to 1,500 MW and announced area-wise load-shedding of one to two hours a day (The Business Standard, 18 July 2022). The plants existed. What was short was fuel the government judged it could afford. A December 2024 report by the Institute for Energy Economics and Financial Analysis (IEEFA), drawing largely on data from the Bangladesh Power Development Board (BPDB), documented the longer pattern.

Over the same five years BPDB still recorded a combined loss of Tk 236.42 billion. Industries ran more than 3,000 MW of their own gas-fired captive generation because grid supply was unreliable, and load-shedding occurred on at least 23 days in nine months of FY2023-24. Peak demand in 2024, at 17,200 MW, was already below the projections of the Integrated Energy and Power Master Plan adopted only a year earlier.

Read through the frameworks above, the sector tracked an internal measure (megawatts installed) and an access measure. The outcome that matters to a factory or a household, reliable supply, and the financial cost of paying for idle capacity were not tied to the same objectives. In 7-S terms, strategy and generating capacity moved quickly, while the systems for fuel procurement, demand forecasting and financing, and the grid itself, did not keep pace.

Three cautions apply. A power system needs reserve capacity, and the right margin is a technical judgment; IEEFA's own roadmap aims for about 36 percent by 2030, not zero. The 2022 crisis was triggered by an external price shock. And IEEFA states that its mission is to accelerate the transition to a diverse, sustainable energy economy, so its reform proposals are one position in a contested debate, although the figures it reports come largely from BPDB itself. The figures above end in 2024; later data were not reviewed for this article. None of this shows that anyone intended a poor outcome. It shows how a strategy judged mainly by one indicator can be executed on its own terms while serving its purpose only in part.

Constructed teaching case. Halda Valley Hospital is fictional, and all people, figures and events below are invented for teaching. Any resemblance to a real hospital is unintended.

Halda Valley Hospital is a 250-bed private hospital in Chattogram. In January 2025 its board adopted a new vision: to become the city's hospital of choice for cardiac care, so that patients would no longer need to travel to Dhaka for angioplasty. It spent BDT 38 crore on a cardiac catheterization laboratory and a 12-bed cardiac intensive care unit, and recruited two interventional cardiologists from Dhaka.

Eighteen months later the laboratory performs about 40 percent of the procedures in the business plan. Of the 14 nurses trained for the cardiac unit, three have left for hospitals in Dhaka and the Gulf. On two occasions, heart-attack patients arriving at night were transferred to Dhaka because no full catheterization team was on call. Referrals from district physicians depend mostly on one cardiologist's personal contacts. The executive committee is divided. The chief financial officer proposes a scorecard of 22 KPIs, cascaded to every department and linked to bonuses. The medical director wants a marketing campaign with discounted angioplasty packages. The HR director argues that the binding constraint is the cardiac nursing team and the absence of a funded 24-hour on-call rota.

A slimmer scorecard built on the strategy map logic of Section 5 might look like this:

Illustrative scorecard for the fictional case. The measures are examples, not recommended clinical or financial targets.

What the case is designed to show. The three executives are not simply arguing about tactics. Each proposal reflects a different theory of why the strategy is stalling: weak accountability, weak demand, or a missing capability. A useful committee discussion would identify which theory the evidence supports before spending the money, and would say what new evidence would change its mind. Notice also that the problem first surfaced at the HR director's level, in nurse retention and rota design, which is exactly where the scorecard's "black hole" sits.

Statements 1, 5 and 6 reflect the barriers Beer and Eisenstat found at the top, statements 2, 3, 4 and 10 draw on Sull, Homkes and Sull, statements 7 and 8 test whether a scorecard is used interactively, and statement 9 asks the 7-S question of fit. The most informative use is to compare your ratings with those of colleagues at other levels. Both studies in Section 3 suggest that senior managers are often the last to learn how the strategy is understood below them.

The studies discussed point in a consistent direction, even though none can say how often strategies fail. Execution tends to break at the top of the organization and between units, and the causal chains that are supposed to connect people to profits are assumptions until tested. The most widely used execution tools, by contrast, are strongest at vertical alignment and measurement. The Balanced Scorecard can be used to test strategy as well as enforce it, and its developers moved it in that direction, but the interactive use they describe depends on senior teams willing to have their own assumptions questioned.

That leaves an organizational question open. Tools that measure the organization below the senior team are easier for senior teams to adopt than processes that examine the senior team itself, which may explain why Beer and Eisenstat's barriers stay silent. This is an interpretation, not a tested finding. Competitive advantage comes from the strategy an organization actually realizes, and whether its leaders are prepared to diagnose themselves may matter as much as which framework they choose.

Select an organization you know well, or can research through dated public sources, that announced a significant strategic change within the last five years. You may disguise its name.

Prepare a report of 2,500 to 3,000 words that:

1. sets out the intended strategy as stated at the time, and the strategy actually realized since, with evidence for each;
2. builds a strategy map for the intended strategy, states the causal assumption behind each link, and assesses what evidence exists that each link held;
3. analyses the fit among the seven elements of the 7-S framework before and after the change, identifying the element that moved most slowly;
4. identifies, with evidence, where execution broke down or succeeded, using the barriers described by Beer and Eisenstat (2000) and the findings of Sull, Homkes and Sull (2015), and says where the evidence is too thin to judge; and
5. recommends changes to the organization's execution system, including their costs, risks and what would show they were not working.

Use at least five academic sources, including at least three from this article's reference list. Every factual claim about the organization must be traceable to a document, dataset or interview you could produce if asked. Do not invent figures, and do not quote a strategy failure rate unless you can show where it came from.

Assessment rubric (100 marks)

1. Beer, M., & Eisenstat, R. A. (2000). The silent killers of strategy implementation and learning. Sloan Management Review, 41(4), 29–40. https://sloanreview.mit.edu/article/the-silent-killers-of-strategy-implementation-and-learning/
2. Bukh, P. N., & Malmi, T. (2005). Re-examining the cause-and-effect principle of the balanced scorecard. In S. Jönsson & J. Mouritsen (Eds.), Accounting in Scandinavia: The northern lights (pp. 87–113). Liber & Copenhagen Business School Press.
3. Cândido, C. J. F., & Santos, S. P. (2015). Strategy implementation: What is the failure rate? Journal of Management & Organization, 21(2), 237–262. https://doi.org/10.1017/jmo.2014.77
4. Charan, R., & Colvin, G. (1999, June 21). Why CEOs fail. Fortune, 139(12), 68–78.
5. Kaplan, R. S. (2010). Conceptual foundations of the balanced scorecard (Harvard Business School Working Paper 10-074). Also published in C. S. Chapman, A. G. Hopwood & M. D. Shields (Eds.), Handbook of management accounting research (Vol. 3). Elsevier, 2009. https://www.hbs.edu/ris/Publication%20Files/10-074_0bf3c151-f82b-4592-b885-cdde7f5d97a6.pdf
6. Kaplan, R. S., & Norton, D. P. (1996). Using the balanced scorecard as a strategic management system. Harvard Business Review, 74(1), 75–85.
7. Martin, R. L. (2010). The execution trap. Harvard Business Review, 88(7/8), 64–71.
8. Nørreklit, H. (2000). The balance on the balanced scorecard: A critical analysis of some of its assumptions. Management Accounting Research, 11(1), 65–88. https://doi.org/10.1006/mare.1999.0121
9. Sull, D., Homkes, R., & Sull, C. (2015). Why strategy execution unravels, and what to do about it. Harvard Business Review, 93(3), 57–66.
10. Waterman, R. H., Jr., Peters, T. J., & Phillips, J. R. (1980). Structure is not organization. Business Horizons, 23(3), 14–26. https://doi.org/10.1016/0007-6813(80)90027-0

Kiechel (1982, 1984), Gray (1986), Judson (1991), Mintzberg (1994) and Kaplan and Norton (2001) are discussed as reported in Cândido and Santos (2015), not from the original texts. Nørreklit's (2000) argument and the quotation from it are reported as summarized in Bukh and Malmi (2005). The 1992 Kaplan and Norton article, Simons's control framework and the Brady and Mobil quotations are reported as described in Kaplan (2010).

Sources for the Bangladesh power-sector section

• Alam, S. (2024, December). Fixing Bangladesh's power sector. Institute for Energy Economics and Financial Analysis. https://ieefa.org/sites/default/files/2024-12/Fixing%20Bangladesh%E2%80%99s%20Power%20Sector_Dec2024.pdf
• "Electricity now in every house." The Daily Star, 21 March 2022. https://www.thedailystar.net/news/bangladesh/news/electricity-now-every-house-2987601
• "More load shedding to come as govt shuts diesel-run power plants." The Business Standard, 18 July 2022. https://www.tbsnews.net/bangladesh/energy/govt-shuts-diesel-run-power-plants-petrol-pumps-remain-closed-once-week-460222

The Halda Valley Hospital case, its scorecard, the Strategy Execution Diagnostic, the framework comparison table and the bank example in Section 6 are original teaching material for this series, not findings reported by the sources above.

Next in The Human Capital MBA series: Article 08, continuing Module 02 (Strategic Management).