Business-IT Alignment

## What the value chain decomposes Michael Porter introduced the value chain in his 1985 book Competitive Advantage as a way to decompose a company into the discrete activities it performs to design, produce, market, deliver, and support its product or service. The model splits activities into two categories: - **Primary activities** that touch the product directly — inbound logistics, operations, outbound logistics, marketing and sales, and service. - **Support activities** that make the primary ones possible — firm infrastructure, human resource management, technology development, and procurement. Each activity has a cost and, ideally, contributes a **margin**: the difference between what customers will pay and what it costs the company to perform all the linked activities. ## The mechanism — overlay systems onto the activities they support Enterprise architects reuse this decomposition for a specific reason: 'align IT with the business' is meaningless until you can point at which piece of the business a given system actually touches, and whether that piece is where the company competes or just where it has to show up. The mechanism is to take the organization's value chain and overlay every major application, platform, or IT capability onto the activity or activities it supports: - A CRM and a real-time pricing engine map onto marketing/sales. - A warehouse management system maps onto outbound logistics. - A payroll system maps onto HR, a support activity. This overlay produces a view that separates systems into two buckets: - Those supporting activities where the company **differentiates** or competes on cost. - Those supporting activities that are necessary **overhead** shared by every competitor in the industry. ## Why it exists — where technology spend should concentrate The reason this exists is to fix a specific, recurring failure of IT investment: treating every system request with the same governance and the same appetite for customization, so that a commodity payroll system gets bespoke integrations while an actual differentiator — say, a recommendation engine — gets starved because it competes for budget on equal footing with things that create no competitive advantage. Value chain mapping forces a conversation about where technology spend should concentrate: - Activities identified as sources of **competitive advantage** justify custom build, deep IT-business partnership, and faster iteration. - Activities that are **pure support** justify buying an off-the-shelf SaaS product, standardizing, or outsourcing entirely, because differentiating there produces no return. ## The trade-off The trade-off is that this decomposition is a simplification of a much messier reality. 1. **Designed for manufacturing-era, largely linear businesses.** Value chains were designed for manufacturing-era, largely linear businesses; a platform business or a two-sided marketplace does not decompose cleanly into inbound-logistics-to-service sequences, and analysts have proposed alternatives like value networks or value shops for those cases. Applying the classic value chain rigidly to, say, a social media platform can misclassify genuinely strategic capabilities (like the recommendation algorithm or the ad-matching engine) as 'technology development,' a support activity, understating their importance. 2. **A point-in-time snapshot.** The other cost is that the mapping is a point-in-time snapshot; what counts as a differentiating activity shifts as competitors catch up — an online checkout flow was a differentiator for early e-commerce and is now table stakes for everyone, so systems that once justified premium investment should eventually be reclassified as commodity and cost-optimized. ## Failure modes 1. **A one-time diagram rather than a living input.** The most common production-grade failure mode is treating the value chain exercise as a one-time architecture diagram rather than a living input to portfolio decisions. Teams draw the chain, get sign-off, and never revisit it, so two years later the 'differentiating' bucket still contains systems that have become commodities, and genuinely new differentiators (a fraud-detection model, a personalization pipeline) never get mapped in and so never get prioritized funding. 2. **Too coarse a grain.** A second failure is mapping at too coarse a grain — labeling all of 'marketing and sales' as differentiating lets low-value marginal systems ride on the coattails of the genuinely strategic ones, defeating the purpose of doing the exercise at all. ## Where it shows up A concrete, widely cited real-world pattern: retail and logistics companies (Walmart's investment in supply-chain visibility and cross-docking systems is the textbook example) used value-chain thinking to identify inbound/outbound logistics as their competitive battleground and poured disproportionate IT investment there — building proprietary systems years ahead of competitors — while running HR and finance systems on largely standardized, vendor-supplied platforms, because no customer ever chose Walmart for its payroll system.