Chapter 3: Reading the context
This introduction explains why requirements alone do not determine the right technical choice and identifies the system, team, and company contexts that also shape it.
After this chapter you will be able to
- Context beyond identical requirements. Explain why two teams with the same requirements and candidate solutions could reasonably make different choices: requirements define the problem, while context affects which solution will work well in each situation.
- Three layers of decision context. Describe the system, team, and company components of decision context and the kinds of circumstances each includes.
- Existing system decision context. Explain why a decision that changes an existing system cannot be treated as a blank-canvas design exercise, and when to consider the system context.
- Engineering 5W questions. Describe the five questions to ask about an existing system: what exists and its condition, where a component is used and a change may lead, when it was built, why it has its current design, and who owns or understands it.
- Assumptions before system changes. Explain why engineers should investigate system context before committing to a significant change, using the identity-management extraction example.
- Component design and change condition. Explain what to establish about a component’s existing design and which parts appear easy or difficult to change.
- Existing solutions across teams. Explain why inspecting your own component is not enough and how to check whether another team already has a usable solution.
- Actual component consumers. Explain why identifying a component’s current consumers requires checking evidence beyond documentation. Describe how usage and dependency tracing, service records, and production traffic can reveal hidden or unexpected dependencies.
- Coupling and blast radius. Explain blast radius as the directly and indirectly affected system parts and organizational boundaries. Distinguish a component’s size from its coupling, including dependencies that cross team or company boundaries.
- Decision process by impact. Explain how the size and certainty of a change’s blast radius affect who should be involved and how much care the decision needs.
- Historical constraints on design. Explain how the technology, team, business, and organizational circumstances at the time a component was built help assess its design choices. Distinguish a choice that was reasonable then from one that was poor even by the standards of the time.
- Historical design constraints. Explain how to investigate why a system was designed as it was by looking beyond its age to the constraints and reasoning at the time.
- Revisiting expired assumptions. Explain when an earlier decision becomes a candidate for revisiting, using the GWT example to distinguish its original constraints from supposed framework advantages.
- Component ownership and knowledge. Distinguish the person accountable for a component’s reliability and evolution from people who know its behavior or the reasons behind its design. Explain that these may be different people.
- Concentrated knowledge risk. Explain what bus factor measures and why a bus factor of one on a load-bearing component is a risk worth naming.
- Expert verification of findings. Justify checking assumptions and findings with someone who knows the software deeply, even after consulting documentation, research, or AI.
- Primary and secondary sources. Distinguish original evidence from material that summarizes or interprets it, and explain why distance from the original source can reduce trust.
- Evidence suited to each question. Explain why different sources warrant different levels of trust for questions about current behavior, implementation, and historical reasoning.
- Verification of weaker leads. Describe how to use a less reliable source to start an investigation, then check a consequential finding against more primary evidence.
- Team context and decision outcomes. Explain why a technically sound technology choice can still fail or underperform, and identify the team factors this section says can influence its outcome.
- Depth of technology experience. Distinguish hands-on operational experience from brief experimentation when assessing whether a team knows a technology.
- Transferable operational knowledge. Explain how experience with comparable technologies can ease a transition, especially when they share operational concerns.
- Technology choices and hiring. Explain how choosing a technology constrains future hiring and why a compelling technical case may not justify a stack with a narrow talent pool.
- Hiring feasibility factors. Describe the factors and questions this section uses to assess whether a team can hire for a proposed stack before committing to it.
- On-call models and technology fit. Explain how team-owned, shared, and dedicated operations rotations change which technologies are realistic choices.
- Team preference as a factor. Explain why team preferences should be considered in a technical decision without automatically determining it. Describe when a team's experience with a technology can be a legitimate reason to choose it.
- Preference disguised as fit. Explain how enthusiasm, familiarity, or career interests can make a preferred technology seem objectively right and lead to selective reasoning, even when the team has good intentions.
- Binding team constraints. Explain why the binding team-context constraints differ by decision, and illustrate how one or two forces can outweigh otherwise favorable conditions.
- Team context in framing. Justify considering team context while framing a decision rather than after narrowing the technical options.
- Four company context forces. Describe how scale, industry, business context, and technology strategy each affect what counts as a good or realistic decision.
- Context-dependent force weighting. Explain why the four forces must be considered together but need not receive equal attention for every decision.
- Strategic role of capabilities. Distinguish differentiating capabilities from enabling ones, and explain why the same capability can have a different role at different companies.
- Strategy-driven technology trade-offs. Describe how a capability’s strategic role changes the appropriate investment depth, choice of standards, and tolerance for novelty.
- Technology strategy guardrails. Explain how technology strategy, derived from business strategy, narrows the options for individual technical decisions through supported technologies, cloud choices, and shared standards.
- Changing technology priorities. Describe how technology strategy establishes the current priority among speed of change, reliability, and cost efficiency, and why that priority can change with business circumstances.
- Strategy as decision navigation. Explain how teams use explicit or implicit technology strategy to guide decisions while retaining autonomy, and distinguish useful guidance from a strategy that is too loose or too prescriptive.
- Startup speed and learning. Explain why startup technical decisions favor shipping a minimum useful solution quickly over investing early in architectural depth or scale.
- Scale-up expanding decision space. Describe how product traction and organizational growth change a scale-up's technical choices and decision process.
- Enterprise durability and guardrails. Explain why enterprise decisions put greater weight on established standards, operational maturity, and long-term investment, while not requiring heavy process for every choice.
- Local scale-pattern reading. Apply the three archetypes as patterns for identifying the dynamics affecting a particular decision, rather than labeling an entire company as one type.
- Industry-specific decision forces. Describe the four dimensions industry can shape in a technical decision: maintainability versus immediate performance, regulatory exposure, ship cadence, and calendar constraints.
- Three layers of context. Explain how system, team, and company context shape the data team's ingestion decision. Describe why the legacy ETL and its limited expert knowledge create migration pressure, while team capacity and company strategy favor lower-effort options.
- Context before option selection. Distinguish framing an evaluation from selecting its winner. Explain which ingestion options the assessment deprioritizes, where evaluation begins instead, and what might make a more manual approach worth reconsidering.
What it builds on
- When the problem is ready for solution discussion (chapter 2) · The distinction between understanding the problem and being ready to discuss solutions.
- Phase 1: Framing the decision (chapter 1) · Framing the decision before evaluating options
Section by section
Introduction
This introduction explains why requirements alone do not determine the right technical choice and identifies the system, team, and company contexts that also shape it.
- Context beyond identical requirements: Explain why two teams with the same requirements and candidate solutions could reasonably make different choices: requirements define the problem, while context affects which solution will work well in each situation.
Why it matters: It keeps a decision from treating a solution that meets the requirements as automatically suitable for the organization. - Three layers of decision context: Describe the system, team, and company components of decision context and the kinds of circumstances each includes.
Why it matters: These three lenses help a reader look beyond requirements without treating context as one vague consideration.
Key terms: requirements, context, decision framing, system context, team context, company context
3.1 System context
This section explains why decisions that change an existing system must account for its constraints after the problem and requirements are understood.
- Existing system decision context: Explain why a decision that changes an existing system cannot be treated as a blank-canvas design exercise, and when to consider the system context.
Why it matters: A solution that fits the requirements still has to work within the system it will change.
Key terms: existing system context, historical complexity
3.1.1 The 5W questions
The section introduces five questions for investigating an existing system before a significant decision and explains how AI can help with the investigation if crucial findings are verified.
- Engineering 5W questions: Describe the five questions to ask about an existing system: what exists and its condition, where a component is used and a change may lead, when it was built, why it has its current design, and who owns or understands it.
Why it matters: These questions help you find context that may change how you approach a proposed change. - Assumptions before system changes: Explain why engineers should investigate system context before committing to a significant change, using the identity-management extraction example.
Why it matters: A change that looks straightforward from its visible uses may be complicated by dependencies the team has not discovered. - AI-assisted investigation limits: Explain how AI-assisted analysis changes the cost of initial system exploration and why crucial findings still need verification.
Why it matters: You can use AI to investigate more cheaply without treating its findings as unquestionable evidence.
Key terms: 5W questions, system context, assumptions, internal submarine usage, AI-assisted analysis, initial exploration, verified
3.1.2 What exists in the system today?
This section explains how to inspect a component’s current condition and check for existing solutions elsewhere in the organization before making a change.
- Evidence about the existing component: Describe how code-analysis and AI tools, together with operational and change-history evidence, can quickly build a picture of a component.
Why it matters: Several quick sources of evidence give you a better starting picture than familiarity with the code alone. - Component design and change condition: Explain what to establish about a component’s existing design and which parts appear easy or difficult to change.
Why it matters: The same proposed change can have different implications depending on the design and condition of the code it touches. - Existing solutions across teams: Explain why inspecting your own component is not enough and how to check whether another team already has a usable solution.
Why it matters: Checking beyond your team can reveal an existing solution before you commit to building another one.
Key terms: code quality analyses tools, AI insights, operational metrics, existing design choices, test-covered, tech debt, outdated dependencies, horizontal axis, reuse, global channel
3.1.3 Where is the component used?
This section shows how to find where a component is actually used, map a change’s blast radius, and match the decision process to that impact.
- Actual component consumers: Explain why identifying a component’s current consumers requires checking evidence beyond documentation. Describe how usage and dependency tracing, service records, and production traffic can reveal hidden or unexpected dependencies.
Why it matters: A change can affect users and processes that the documentation does not show. - Coupling and blast radius: Explain blast radius as the directly and indirectly affected system parts and organizational boundaries. Distinguish a component’s size from its coupling, including dependencies that cross team or company boundaries.
Why it matters: A small component with many dependents may have a much wider impact than a large, isolated one. - Decision process by impact: Explain how the size and certainty of a change’s blast radius affect who should be involved and how much care the decision needs.
Why it matters: Assessing impact helps avoid both over-processing contained changes and under-processing consequential ones.
Key terms: consumers, hidden dependencies, production traffic, use and dependency tracing, blast radius, coupling, organisational boundaries, cascading impact, lightweight process, one-way door decision
3.1.4 When was it built?
This section shows how the circumstances in which a component was built can explain puzzling design choices and change how an engineer approaches them today.
- Historical constraints on design: Explain how the technology, team, business, and organizational circumstances at the time a component was built help assess its design choices. Distinguish a choice that was reasonable then from one that was poor even by the standards of the time.
Why it matters: Historical context helps you judge an existing design against the constraints its creators actually faced. - Historically imposed complexity: Explain how the frontend layout example changes the assessment of an apparently complex codebase and the decision to modernize it.
Why it matters: Knowing where complexity came from can change whether you treat it as a warning or an opportunity to simplify.
Key terms: when question, technology context, team's constraints, CSS grid, custom properties, current CSS primitives, modernization candidate
3.1.5 Why is it the way it is?
This section shows how the original reasons for a design help explain its current shape and reveal when it may be worth revisiting.
- Historical design constraints: Explain how to investigate why a system was designed as it was by looking beyond its age to the constraints and reasoning at the time.
Why it matters: A design that looks arbitrary today may have been a reasonable response to its original circumstances. - Revisiting expired assumptions: Explain when an earlier decision becomes a candidate for revisiting, using the GWT example to distinguish its original constraints from supposed framework advantages.
Why it matters: Checking whether the original reason still holds keeps an old decision from becoming an unquestioned constraint.
Key terms: ADRs, historical constraint, Conway's Law, original constraints, decision context assumptions, Google Web Toolkit (GWT)
3.1.6 Who owns it, who knows it deeply, and who has the historical context?
This section shows how to identify the people who can explain a component, recognize concentrated knowledge as a risk, and check findings with someone who knows the software deeply.
- Component ownership and knowledge: Distinguish the person accountable for a component’s reliability and evolution from people who know its behavior or the reasons behind its design. Explain that these may be different people.
Why it matters: Finding the right person for each question makes the component’s context easier to establish. - Concentrated knowledge risk: Explain what bus factor measures and why a bus factor of one on a load-bearing component is a risk worth naming.
Why it matters: A decision can expose an existing knowledge risk even if that risk does not determine which architecture to choose. - Expert verification of findings: Justify checking assumptions and findings with someone who knows the software deeply, even after consulting documentation, research, or AI.
Why it matters: A conversation with a knowledgeable person can catch omissions before they shape a decision.
Key terms: owner, historical context, who knows it deeply, bus factor, load-bearing component, assumptions, findings, who knows it
3.1.7 Understanding whether your findings are solid enough
The section explains how to judge the reliability of findings about an existing system and verify weaker sources before relying on them.
- Primary and secondary sources: Distinguish original evidence from material that summarizes or interprets it, and explain why distance from the original source can reduce trust.
Why it matters: Knowing where a claim came from helps you decide whether to rely on it or check further. - Evidence suited to each question: Explain why different sources warrant different levels of trust for questions about current behavior, implementation, and historical reasoning.
Why it matters: Matching evidence to the question prevents an account of what was intended from being mistaken for what runs today. - Verification of weaker leads: Describe how to use a less reliable source to start an investigation, then check a consequential finding against more primary evidence.
Why it matters: Checking a lead before acting can reveal dependencies that an apparently useful diagram missed.
Key terms: primary source, secondary source, runtime evidence, AI analyses of the code, direct testimony, primary documents, leads, primary evidence, runtime logs, API credentials
3.2 Team context
This section explains why a technology choice must be judged in the context of the team that will adopt and operate it.
- Team context and decision outcomes: Explain why a technically sound technology choice can still fail or underperform, and identify the team factors this section says can influence its outcome.
Why it matters: Considering the people who must use and support a technology helps avoid choices that look good technically but struggle in practice.
Key terms: team context, team knowledge, talent pool, technology preference
3.2.1 Team knowledge
This section explains how to assess a team’s relevant technology knowledge by looking at hands-on operations, comparable experience, and expertise across the organization.
- Depth of technology experience: Distinguish hands-on operational experience from brief experimentation when assessing whether a team knows a technology.
Why it matters: It helps you avoid treating a quick trial as evidence that the team can run a technology. - Transferable operational knowledge: Explain how experience with comparable technologies can ease a transition, especially when they share operational concerns.
Why it matters: Related experience can make an unfamiliar option less unfamiliar, but the operational overlap matters. - Organizational expertise sharing: Describe how to find and share relevant experience from colleagues’ earlier roles and from elsewhere in the organization.
Why it matters: Looking beyond the immediate team can reveal useful experience before a decision is made.
Key terms: direct experience, under load, recover, comparable technologies, self-hosted solutions, managed services, shared operational concerns, earlier roles, peer learning session, team autonomy
3.2.2 Talent availability
Technology choices shape future hiring, so their technical benefits must be weighed against the talent a team can realistically recruit and develop.
- Technology choices and hiring: Explain how choosing a technology constrains future hiring and why a compelling technical case may not justify a stack with a narrow talent pool.
Why it matters: A stack that works well technically can still make it difficult to build the team needed to sustain it. - Hiring feasibility factors: Describe the factors and questions this section uses to assess whether a team can hire for a proposed stack before committing to it.
Why it matters: The same stack can be a practical hiring choice for one team and a serious constraint for another.
Key terms: talent pool, niche technology, technical case, hiring case, remote-first organisations, background match, ramp-up
3.2.3 On-call setup
This section explains how the people responsible for incidents constrain which technologies a team can operate reliably.
- On-call models and technology fit: Explain how team-owned, shared, and dedicated operations rotations change which technologies are realistic choices.
Why it matters: A tool that is easy to build with may still be a poor choice if the incident responder cannot operate it. - Shared operational capability: Justify checking whether the whole responder group can debug a proposed solution, rather than relying on implementation convenience or one expert.
Why it matters: Checking operational capability can expose risks that are easy to miss when judging a tool by implementation productivity alone.
Key terms: team-owned, shared rotation, Network Operations Center (NOC), runbooks, operational stack, shared operational stability, local implementation productivity, debug, managed service
3.2.4 Team preferences
This section explains how to account for team preferences in technical decisions without mistaking enthusiasm or familiarity for evidence of fit.
- Team preference as a factor: Explain why team preferences should be considered in a technical decision without automatically determining it. Describe when a team's experience with a technology can be a legitimate reason to choose it.
Why it matters: Taking preferences seriously without giving them a free pass makes the decision more realistic. - Preference disguised as fit: Explain how enthusiasm, familiarity, or career interests can make a preferred technology seem objectively right and lead to selective reasoning, even when the team has good intentions.
Why it matters: A persuasive technical argument can still be shaped by a choice the team wanted to make from the start. - Questions that test fit: Describe the section's questions for checking whether a favored option fits the team's context rather than merely feeling familiar, impressive, or fashionable.
Why it matters: These questions help expose weak reasoning without rejecting a preference that genuinely matches the context.
Key terms: team preference, team's experience, technical judgement, selective reasoning, motivated reasoning, fit, prior experience, competent engineer outside this team, big tech companies
3.2.5 Understanding which forces are binding
This section explains how to identify the team-context forces that actually constrain a decision and account for them when framing it.
- Binding team constraints: Explain why the binding team-context constraints differ by decision, and illustrate how one or two forces can outweigh otherwise favorable conditions.
Why it matters: An option the team could build may still be one it cannot realistically operate or absorb. - Team context in framing: Justify considering team context while framing a decision rather than after narrowing the technical options.
Why it matters: Considering the team early prevents an apparently attractive option set from overlooking operational and change capacity.
Key terms: binding constraint, team context, on-call rotation, framing, option set
3.3 Company context
This section explains why company context changes which technical decisions are realistic and introduces four forces whose influence varies by decision.
- Company and team context: Distinguish team context from company context when explaining why the same technical decision may work differently in two organizations.
Why it matters: A technically sound choice still has to fit the organization adopting it. - Four company context forces: Describe how scale, industry, business context, and technology strategy each affect what counts as a good or realistic decision.
Why it matters: These forces explain why evaluating a choice on technical merits alone is insufficient. - Context-dependent force weighting: Explain why the four forces must be considered together but need not receive equal attention for every decision.
Why it matters: Identifying the forces that matter most keeps the decision focused on its actual context.
Key terms: team context, company context, scale, industry, business context, technology strategy, guardrails, four forces, company scale, industry context
3.3.1 Business strategy
This section shows how a capability’s role in business strategy should shape technology decisions, even when its immediate requirements are clear.
- Strategic role of capabilities: Distinguish differentiating capabilities from enabling ones, and explain why the same capability can have a different role at different companies.
Why it matters: You cannot infer where to differentiate merely from how visible or critical a capability is. - Future capability headroom: Explain why a technology choice should account for where the business intends to differentiate, beyond what satisfies the initial requirements.
Why it matters: A choice that works now may constrain a capability the company expects to compete on later. - Strategy-driven technology trade-offs: Describe how a capability’s strategic role changes the appropriate investment depth, choice of standards, and tolerance for novelty.
Why it matters: The best technology choice is the one that advances the capability’s strategic role, not an abstract technical winner.
Key terms: differentiating capabilities, enabling capabilities, strategic role, capability headroom, initial requirements, strategic context, investment depth, choice of standards, tolerance for novelty
3.3.2 Technology strategy
This section explains how technology strategy shapes technical decisions through organizational guardrails and changing priorities without taking decisions away from teams.
- Technology strategy guardrails: Explain how technology strategy, derived from business strategy, narrows the options for individual technical decisions through supported technologies, cloud choices, and shared standards.
Why it matters: A technically workable option may still impose avoidable costs across the organization. - Changing technology priorities: Describe how technology strategy establishes the current priority among speed of change, reliability, and cost efficiency, and why that priority can change with business circumstances.
Why it matters: A trade-off that made sense last year may not fit what the business needs now. - Strategy as decision navigation: Explain how teams use explicit or implicit technology strategy to guide decisions while retaining autonomy, and distinguish useful guidance from a strategy that is too loose or too prescriptive.
Why it matters: Useful strategy gives teams a common direction without replacing their judgment. - Early strategy alignment: Justify checking a proposed choice against technology strategy while framing the decision rather than waiting for approval.
Why it matters: An early alignment check can prevent costly rework and a late loss of confidence in the proposal.
Key terms: technology strategy, guardrails, shared standards, priorities, speed to market, change agility, reliability and resilience, cost efficiency, navigation system, meaningful autonomy
3.3.3 Company scale
The section shows how startup, scale-up, and enterprise dynamics change technical trade-offs, and why those dynamics must be read in the specific decision context rather than assigned as fixed company labels.
- Startup speed and learning: Explain why startup technical decisions favor shipping a minimum useful solution quickly over investing early in architectural depth or scale.
Why it matters: It helps you avoid building expensive foundations before you have proved that customers need the product. - Scale-up expanding decision space: Describe how product traction and organizational growth change a scale-up's technical choices and decision process.
Why it matters: Growth opens up valuable architectural options while making both uncoordinated choices and unnecessary process more costly. - Enterprise durability and guardrails: Explain why enterprise decisions put greater weight on established standards, operational maturity, and long-term investment, while not requiring heavy process for every choice.
Why it matters: It helps distinguish governance that manages substantial risk from process that merely slows a smaller decision. - Local scale-pattern reading: Apply the three archetypes as patterns for identifying the dynamics affecting a particular decision, rather than labeling an entire company as one type.
Why it matters: A local reading prevents a company-wide label from hiding the constraints that actually govern your choice.
Key terms: startup, product-market fit, learning speed, managed cloud services, SaaS, premature optimization, scale-up, foundational trade-offs, decision space, cargo-culting
3.3.4 Industry norms
This section shows how industry norms shape technical trade-offs and how engineers can uncover norms that are rarely written down.
- Industry-specific decision forces: Describe the four dimensions industry can shape in a technical decision: maintainability versus immediate performance, regulatory exposure, ship cadence, and calendar constraints.
Why it matters: These forces help explain why a sound technical choice in one industry may be a poor choice in another. - Game and safety-critical trade-offs: Explain why game development may favor immediate player experience over long-term architectural cleanliness, while healthcare and aviation put greater weight on demonstrable correctness and careful change.
Why it matters: The contrast makes clear that architectural priorities depend on the consequences of a decision in its domain. - Release timing trade-offs: Distinguish how time pressure shapes crypto and Web3 releases from how peak business periods shape releases in ecommerce and travel.
Why it matters: Timing can justify either shipping sooner or holding a change back, depending on the industry's risks. - Finding unwritten industry norms: Describe why company strategy documents may not reveal industry norms and how engineers can discover the relevant constraints and failed approaches.
Why it matters: Looking beyond written strategy can expose constraints that otherwise remain invisible in a decision.
Key terms: maintainability versus immediate performance, regulatory exposure, ship cadence, calendar constraints, player experience, re-validation, re-certification, safety-critical, technical debt, peak windows
3.4 Reading context: a worked example
The section uses a data-ingestion decision to show how system, team, and company context narrow the options worth evaluating without choosing a solution.
- Three layers of context: Explain how system, team, and company context shape the data team's ingestion decision. Describe why the legacy ETL and its limited expert knowledge create migration pressure, while team capacity and company strategy favor lower-effort options.
Why it matters: These forces explain why a technically possible build may be a poor fit for this particular team. - Context before option selection: Distinguish framing an evaluation from selecting its winner. Explain which ingestion options the assessment deprioritizes, where evaluation begins instead, and what might make a more manual approach worth reconsidering.
Why it matters: Framing focuses evaluation effort without mistaking an initial preference for a final decision. - Different context different framing: Explain why another team facing the same ingestion question might favor extending existing platforms rather than adding a managed service.
Why it matters: It shows why an option that fits one organization need not fit another with the same requirements.
Key terms: system context, team context, company context, legacy ETL, bus factor, context assessment, managed SaaS solutions, build-vs-buy, connector coverage, existing ecosystem
3.5 Summary
Where this is used later
- Screening against key requirements (chapter 4) · System context describes what already runs and must interoperate.
- Qualifying candidates (chapter 4) · Candidate qualification follows the narrowing of options demonstrated here.
Concepts explained in this chapter
- company scaleCompany scale is an organization's size and stage of development as factors shaping its technical decisions.
- company archetypeA company archetype is a recurring organizational type, such as a startup, scale-up, or enterprise, with characteristic decision dynamics.
- decision makingThe selection among technical alternatives in light of context, priorities, constraints, and uncertainty.
- guardrailA boundary or guideline that constrains choices while allowing teams to make decisions within it.
- technical strategyA long-term direction for an organization's technology choices that establishes operating boundaries and priorities.
- differentiating capabilityA differentiating capability is a business capability whose superiority to competitors creates customer value and competitive position.
- enabling capabilityAn enabling capability is a business capability needed to operate but whose superiority over competitors does not strongly differentiate the business.
- organizational contextCompany context is the organizational forces that shape which decisions are realistic at a particular stage of a company's life.
- team contextThe experience, capabilities, and interpersonal dynamics of the people who will own a result.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.
