Business strategy
This chapter examines strategic analysis, choice and implementation. You must connect analytical models to a firm's resources, competitive environment and stakeholders, then evaluate whether a proposal is suitable, acceptable and feasible.
Core rules
- Strategy is the long-term direction of the whole organisation. It may be deliberate or emergent; Mintzberg argues that intended strategies are reshaped by experience.
- Analysis tools include SWOT, PESTLE, Porter's five forces, the value chain and core competences. SWOT is only a summary: pair strengths with opportunities (SO) or weaknesses with threats (WT) to generate options.
- Ansoff's matrix gives four growth directions: market penetration, product development, market development and diversification. Familiarity can reduce uncertainty, but actual risk depends on capabilities, commitment and the particular market.
- Porter's generic strategies are cost leadership, differentiation and focus. A firm stuck between them may lack a clear competitive advantage.
- Implementation requires structure, culture, leadership and resources. Lewin's unfreeze–change–refreeze and Kotter's change steps explain why strategy often fails in execution.
- Evaluation criteria are suitability, acceptability and feasibility. Decision trees use expected values but ignore risk attitude and qualitative factors.
Worked example
A retailer is deciding whether to open a new outlet. Success has probability 0.6 and an operating payoff before the initial outlay of $200,000; failure has probability0.4 and an operating loss before the initial outlay of $50,000. The outlet costs $80,000. Expected value = (0.6 × $200,000) + (0.4 × −$50,000) = $120,000 − $20,000 = $100,000; net EV = $100,000 − $80,000 = $20,000. This is positive, but the downside is a $50,000 loss plus the $80,000 cost. The break-even probability p solves 200p − 50(1 − p) − 80 = 0, giving p = 0.52. Reasoning: the estimated financial expected value is positive when the stated success probability exceeds52%, with the other supplied payoffs fixed. Survival constraints, estimation error and risk preferences can justify a different decision.
Traps
Diversification involves new products and new markets, but actual risk should be evaluated rather than inferred mechanically from a matrix label. Cost leadership is not the same as low price with negative margins. SWOT lists without links earn little. Stakeholder mapping uses power and interest, not wealth and status. Expected value ignores risk attitude and non-financial factors. Strategy is not the same as an objective; implementation and evaluation must be addressed.
Testing the business model
A business model explains how an organization creates value for customers, delivers that value and obtains enough resources to continue. A strategy chooses where and how to compete. Neither document becomes reliable simply because its boxes are filled. A useful model identifies assumptions that can be tested and shows how a change in one part affects the others.
Begin with a specific customer segment and a concrete problem or desired outcome. A value proposition describes that outcome rather than merely listing product features. A channel makes the offer discoverable, purchasable or deliverable. Customer relationships explain how customers are acquired, supported and retained. Revenue streams describe how payment arises, while cost structure identifies important fixed, variable and step costs. Key activities, resources and partners explain how the promised outcome can actually be delivered. A partner name is not itself a capability, and an attractive proposition is not evidence that the organization can deliver it profitably.
Use a canvas as a connected hypothesis map. A subscription service needs recurring benefits and retention, not only a monthly invoice. A two-sided platform needs participation on both sides and a way to handle the initial coordination problem. A low advertised price can attract customers but destroy contribution if delivery, support or refunds are omitted. Acquisition cost and repeat purchasing need consistent definitions and time periods. Do not compare an optimistic lifetime revenue estimate with only one month of acquisition spending while ignoring service costs and customer departures.
An inexpensive pilot can test the riskiest assumption before major investment. State who will participate, what behavior counts as evidence, what success threshold was chosen in advance and what would make the team revise the model. Actual bookings or purchases often provide different information from expressions of interest. A minimum viable product still needs appropriate safety, honesty and data protection. The word minimum does not justify selling an unsafe or misleading offer.
For example, a hypothetical trial sells two hundred boxes at ten currency units each. Direct fulfillment costs six per box and the campaign costs one thousand. Revenue is two thousand, contribution before the campaign is eight hundred and the trial result after that campaign is a loss of two hundred before other costs. This may still provide valuable learning, but it is not evidence of immediate commercial profitability. Distinguish learning value, current cash requirements and long-run economics. A pivot changes a substantial assumption in response to evidence and should be followed by another test, rather than by a claim that the replacement model is now guaranteed.