Key Takeaways
Pair agile innovation with bureaucracy instead of declaring war on it.
Balance is the point. Agile helps when a company must discover what customers value and how to deliver it. Bureaucratic rules and clear authority remain useful for stable work that depends on safety, consistency, and predictable execution. Replacing every process with experimentation can create as much damage as freezing the business in old routines. The goal is to make operations and innovation reinforce each other, not to make every employee an agile team member. The authors estimate that agile teams commonly involve 10 to 50 percent of employees, while most people continue running the business.
Evidence favors careful use. The authors reviewed 70 third-party research reports: 76 percent found agile innovation outperformed traditional approaches, while 10 percent found no benefit. The enterprise-level evidence was much thinner, so results support experimentation, not a guarantee of success. Agile is a tool for improving performance, not a strategy or a cover for layoffs.
The argument is strongest when read as a design problem rather than a culture war. Reliability and experimentation solve different kinds of uncertainty, and mature organizations need both. This resembles the broader idea of organizational ambidexterity: firms must exploit proven capabilities while exploring new ones. The difficult question is where to draw the boundary. Too much separation leaves innovations stranded; too little exposes essential operations to needless variation. The book’s insistence on balance is useful precisely because it refuses to make autonomy a universal virtue. The evidence base also deserves calibration: studies of agile teams are not interchangeable with evidence about whole-enterprise transformations.
Give small teams customer problems, not instructions for building solutions.
Start with the work. An agile team is a small, multidisciplinary group, usually three to nine people, with the skills and authority to solve a customer problem. Leaders clarify what matters and why, while the team determines how to proceed. A product owner ranks the team’s backlog, a prioritized list of customer needs and opportunities. The team works in short cycles called sprints, producing something testable and gathering feedback instead of waiting for a polished final product.
Brian’s team shows the method. In the book’s composite story, a snack company team makes early prototypes, tests them with consumers, and changes its priorities based on what people actually like. In six weeks, it has tested seven prototypes. The point of a sprint is not to pressure people into working faster. It is to shorten the distance between a guess and useful evidence.
The customer-facing loop changes who gets to define reality. Instead of relying solely on executives’ forecasts, teams can observe how people respond to actual offerings. This echoes experimental design: a modest prototype can test a consequential assumption more cheaply than a polished launch. But customer input is not automatically a compass. Current customers may not represent future markets, and reported preferences can differ from actual behavior. A strong team therefore treats feedback as evidence to interpret, not as an order to obey. Its advantage lies in repeatedly testing assumptions while retaining responsibility for the business outcome.
Remove enterprise waiting time before celebrating faster agile teams.
Speed is a system property. Flow efficiency is the share of total elapsed time spent doing the work, rather than waiting for approvals, funding, testing, or another department. The authors report that many companies operate at only 15 to 20 percent flow efficiency. If a team makes its active work 20 percent faster, the enterprise may gain only 3 to 4 percent overall when waiting dominates the timeline.
The app still waited. A financial-services company used agile to build a well-received mobile app, but annual funding, traditional security testing, and a six-to-nine-month integration queue delayed release. The team improved its own work while the larger system remained slow. Cutting support staff without changing the processes can make queues longer still, while assigning people extra projects creates costly multitasking.
This is a useful correction to the habit of measuring visible activity instead of end-to-end delivery. Queueing theory helps explain why: when work arrives faster than a constrained process can handle it, the queue grows, and elapsed time can rise sharply even if individual tasks improve. Local productivity is therefore a poor proxy for customer value. The practical implication is not to abolish every control. Some reviews protect security and quality. Instead, leaders should identify which waits add meaningful risk reduction and which persist because nobody owns the handoff. The book’s emphasis on flow directs attention to the whole path.
Set a sustainable pace of change, then adjust from evidence.
There is no universal dose. The right level of agility varies by company, industry, and activity. Too little change leaves a business unable to respond; too much can overwhelm operations and create confusion. The authors call the desirable balance the golden mean, where the benefits of agility outweigh its costs. Because leaders cannot reliably forecast the right destination in advance, they should test changes, measure results, and adjust the pace.
Mark Allen learned this physically. The triathlete initially trained at a punishing speed, failed to finish his first Ironman, and kept getting sick. Coach Phil Maffetone helped him train near a sustainable aerobic heart rate of about 155 beats per minute. Allen felt slow at first, but built endurance and eventually won the 1989 Ironman World Championship. The business lesson is to pursue demanding progress without confusing exhaustion with improvement.
The training analogy makes a complex systems principle tangible: adaptation requires enough stress to stimulate learning, but not so much that the system breaks down. In organizations, the equivalent of injury can be lost expertise, fatigued employees, or weakened service quality. Small pilots make it possible to observe these costs before scaling. Still, a sustainable pace should not become a polite excuse for delay when a business faces an urgent threat. The relevant measure is not whether change feels comfortable, but whether the organization can absorb it while maintaining essential performance and learning from the consequences.
Leaders must practice autonomy before asking others to trust it.
Change begins at the top. Agile leadership is less about assigning tasks and more about clarifying priorities, developing capable decision makers, and removing barriers. Leaders still set direction and remain accountable for results, but they stop prescribing every method. Their own team should also work differently, with transparent priorities, short decision cycles, and space for candid feedback.
Henk Becker changed his habits. At Bosch Power Tools, the executive began asking for feedback, listening more, and replacing reflexive explanations of why work could not be done with questions about possible solutions. He gave up his office and assigned parking spot, and stopped requiring polished presentations when teams already had useful working information. Such gestures alone cannot transform a company, but Becker paired them with new routines, including short leadership stand-ups and broader participation in strategy discussions.
The leadership shift is a redistribution of information and decision rights, not a retreat from responsibility. Front-line teams often see customer friction and process delays before senior leaders do, so autonomy can improve both response time and the quality of local decisions. Yet delegation without clear priorities can simply transfer confusion downward. Trust is most effective when teams know the intended outcome, the limits of their authority, and when escalation is appropriate. Becker’s example also highlights a practical test of sincerity: employees judge a new management philosophy by repeated behavior, especially when results disappoint, rather than by the vocabulary leaders adopt.
Fund learning in stages, and move money when evidence changes.
Planning remains essential. Agile planning treats forecasts as hypotheses to revisit, not contracts to obey regardless of new evidence. Dell’s Management Model combines a multiyear ambition with a ranked list of initiatives, then reviews priorities during the year. Its central agenda usually contains fewer than a dozen active initiatives, limiting the drag of multitasking.
Budget for discovery. Agile funding resembles venture investment: provide enough resources to test an opportunity, then expand, change, or stop funding as results emerge. A financial-services company expected customers to want car recommendations alongside an online search tool. Testing showed that search mattered, but recommendations did not, so a team was reassigned rather than spending money to finish an unwanted feature. For durable customer needs, persistent teams can remain in place and adapt their work without seeking fresh project approval each time.
This approach treats budget allocation as a sequence of decisions under uncertainty rather than a yearly contest to defend estimates. It can reduce the sunk-cost trap, the tendency to continue funding a weak initiative because money has already been spent. But flexible funding needs guardrails: teams still require transparent outcome measures, financial oversight, and a clear way to stop work that no longer merits investment. Otherwise, adaptability can become a label for shifting priorities without accountability. The book’s emphasis on both evidence and results makes the distinction clear: learning is valuable when it improves a decision, not when it excuses an endless experiment.
Redesign decision rights and talent systems, not just the org chart.
Structure is only one component. An operating model also includes accountabilities, decision rights, leadership routines, talent practices, and ways of coordinating work. Copying another company’s chart cannot reproduce the culture and technical conditions that made it work. Spotify, for example, designed its squads and related groups for its own engineering teams, and warned that others should not simply transplant the model.
Build the conditions around teams. Bosch Power Tools spread changes across strategy, organization, leadership, processes, and culture over several years. Its structure changed, but so did coaching, feedback, career paths, and team accountability. Walmart’s Hiring Helper provides a concrete example of talent practices changing through experimentation: after testing the tool in two markets, the company reported a 20 percent improvement in hiring fit, alongside lower early attrition and absenteeism.
The operating-model view helps explain why structural imitation so often disappoints: the same chart can produce different behavior when incentives, authority, skills, and technical dependencies differ. This is a classic systems-design problem, where a change in one component can create pressure elsewhere. A useful implication is to prototype not only team structures but also staffing, evaluation, and escalation rules. The Walmart figures are promising, though they are company-reported results rather than a general causal estimate. Any organization adopting a similar tool should test for fairness and unintended effects, not just hiring efficiency.
Work backward from customer needs to processes and modular technology.
Solve the experience, not the department. Customer needs should shape the service or product, which should then shape the processes and technology required to deliver it. Teams improving internal systems must consider both their direct users and the external customers those users serve. Modular processes and software let teams make changes with fewer dependencies on the rest of the organization.
RBS redesigned mortgage applications. The bank began with a paper form averaging 66 pages and reorganized work around the customer journey of getting a mortgage. Cross-functional teams tested prototypes in two-week cycles, changed both the digital experience and the work of mortgage advisers, and trained staff as the new process was introduced. The bank reported that 90 percent of applications became paperless and average application-to-offer time fell from 23 days to 11. The software mattered, but it worked because the surrounding process changed too.
The sequence matters because digitizing a poor process can make an inefficient experience faster without making it better. Starting from the customer journey gives technology teams a reason to ask which steps are necessary, rather than automating every inherited rule. Modularity then helps separate changes and limit the risk of disrupting unrelated services. This approach is especially valuable in complex institutions, although regulation and safety may constrain experimentation. Those constraints should shape the test, not automatically block it. The RBS example illustrates an important measure of success: whether the whole service improves, not simply whether the software ships.
Borrow Amazon's design logic, not its org chart or culture.
The system mattered more than a template. Amazon built small, autonomous two-pizza teams for innovation, pushed teams to reason from customer needs, and used modular software architecture to make services easier to change or replace. Its written PR/FAQ proposals began with a description of the customer benefit, followed by questions about how the idea would work. This forced teams to explain the problem before asking for resources.
Components reinforced one another. Microservices helped teams release changes independently and reverse unsuccessful ones. That architecture supported experiments that contributed to offerings such as Amazon Marketplace and Amazon Web Services. But the company also had demanding deadlines, a confrontational culture, and weaknesses the authors do not recommend copying. The lesson is not to replicate Amazon’s teams or management style. It is to understand how customer focus, technical design, team authority, and funding practices fit together in that particular system.
Amazon makes the book’s systems argument concrete: a practice succeeds partly because other practices make it workable. Team autonomy is easier to sustain when architecture reduces dependencies and proposals start with a customer problem. This is a more valuable lesson than copying a team size or vocabulary. It also invites scrutiny: a system can produce impressive innovation while imposing substantial human costs. Performance alone does not settle whether a model is desirable. Organizations should decide which outcomes matter, including employee experience, and adapt mechanisms to their own context rather than treating commercial success as proof that every feature is wise.
Analysis
The book’s central contribution is to shift the question from whether agile is good to what kind of work it is good for, and what surrounding systems allow it to succeed. Its recurring examples make one distinction especially memorable: teams may work quickly while the company remains slow. That gap turns agile from a team-level method into an enterprise design challenge involving funding, technology, talent, leadership, and operating controls.
The authors are persuasive in rejecting two shortcuts: universal conversion and imitation. Their alternatives are coherent: anchor change in customer outcomes, build small teams with real authority, shorten feedback cycles, track constraints, and scale only as the organization learns. The cases from Bosch, Dell, RBS, and Amazon show different configurations rather than one universal blueprint. The Mark Allen story supplies a compact analogy for pace: intensity without recovery is not durable progress.
The evidence deserves the same empiricism the book advocates. Its review of 70 reports finds broad support for innovation and team-level agile, but the evidence for whole-enterprise transformations is younger and less academically mature. Many cases are descriptive, so they illuminate mechanisms more clearly than they establish causal effects. Agile also shifts costs rather than abolishing them: more autonomy can raise coordination demands, while fewer approvals can increase variation and risk.
A useful application is therefore diagnostic, not ideological. Identify whether the work is predictable or uncertain, where elapsed time is lost, which decisions can safely move closer to customers, and what outcome would justify the investment. Preserve controls that protect essential reliability, redesign those that merely create queues, and measure customer, financial, and employee consequences together. The book’s most practical test is whether the whole operating system becomes more capable of learning without losing its ability to deliver reliably.
Review Summary
Doing Agile Right receives mostly positive reviews for providing practical insights on implementing Agile methodologies in organizations. Readers appreciate its balanced approach, real-world examples, and emphasis on customizing Agile practices to fit specific business needs. The book is praised for its clear explanations and guidance on scaling Agile across enterprises. Some criticisms include a lack of detailed implementation strategies and occasional repetitiveness. Overall, it's recommended for business leaders and those interested in understanding Agile transformation at an organizational level.
People Also Read
Glossary
Agile enterprise
Balanced innovation and operationsAn organization that uses agile methods to improve both innovation and the reliable operation of its existing business. It does not turn every function into an agile team. Instead, it adapts structures, leadership, planning, funding, talent practices, processes, and technology so innovation and operations support one another.
Agile at scale
Many agile teamsThe expansion of agile teams across more projects, functions, or large initiatives, while much of the surrounding organization may continue to operate bureaucratically. It can improve the teams’ innovation work, but traditional funding, approvals, and operating processes may limit improvements in end-to-end delivery.
Flow efficiency
Work time versus waiting timeThe proportion of total elapsed time that people spend actively working on an innovation, rather than waiting for funding, approvals, testing, integration, or other dependencies. Low flow efficiency means that faster team execution may produce little improvement in the time customers wait for a finished solution.
Backlog
Ranked work opportunitiesA prioritized list of customer needs, problems, or opportunities for an agile team or leadership group to address. Unlike a fixed task list, a backlog is continually reordered as evidence changes. Teams generally focus on its highest-value items first, helping limit multitasking and low-value work.
Sprint
Time-limited work cycleA short, defined period in which an agile team produces a testable piece of work, commonly in two-week cycles and generally under a month. Sprints create frequent opportunities to gather customer feedback, learn whether assumptions hold, and adjust the team’s priorities or approach.
Product owner
Customer-value decision makerThe person accountable for ensuring that an agile team’s work creates value for customers and the business. The product owner works with customers, leaders, and stakeholders to identify and rank opportunities, while the team decides how to carry out the work.
Dell Management Model
Adaptive strategy and funding cycleDell’s approach to setting a long-term ambition, identifying initiatives that could close the gap between that ambition and expected performance, and revisiting priorities during the year. A small, ranked corporate agenda helps leaders concentrate resources while allowing new opportunities to be added when evidence warrants.
Microservices
Independent software componentsSmall software services designed to perform specific functions and connect to other services through defined interfaces. Teams can develop, test, deploy, or replace these components with less dependence on a single large system. The approach can make experimentation faster and make unsuccessful changes easier to reverse.
PR/FAQ
Customer-focused proposal documentAn Amazon proposal format that begins with a mock announcement describing the intended customer benefit, then answers questions about how the proposed solution would work. It is used to clarify the customer problem and examine an idea before resources are committed to building it.
FAQ
What's Doing Agile Right about?
- Agile Transformation Focus: The book explores how organizations can implement agile methodologies effectively, balancing them with traditional structures to avoid chaos.
- Good vs. Bad Agile: It distinguishes between successful agile practices and common pitfalls, providing insights to achieve sustainable success.
- Practical Roadmap: Offers guidance for leaders and teams, supported by real-world examples and empirical evidence across various industries.
Why should I read Doing Agile Right?
- Expert Insights: Authored by experienced consultants, the book is backed by extensive research and case studies.
- Applicable to All Levels: Provides valuable guidance for both senior executives and team members on adopting agile practices.
- Avoid Common Mistakes: Helps readers recognize and avoid pitfalls in agile transformations for smoother transitions and better outcomes.
What are the key takeaways of Doing Agile Right?
- Agile Requires Commitment: Agile is not a quick fix; it requires time, experimentation, and commitment to achieve results.
- Balance is Key: Successful organizations balance agile practices with necessary bureaucratic structures for innovation and efficiency.
- Leadership's Crucial Role: Active involvement and support from leadership are essential for successful agile transformations.
What are the best quotes from Doing Agile Right and what do they mean?
- "If you and your team are not having fun with agile, you’re not doing it right.": Highlights the importance of engagement and enjoyment in the agile process.
- "Agile isn’t a goal in itself; it’s a means to an end.": Emphasizes focusing on business outcomes rather than just implementing agile for its own sake.
- "The challenge is not to replace bureaucracy with agile everywhere but to find the right balance between the two.": Suggests a balanced approach to agile transformation.
How does Doing Agile Right define agile leadership?
- Empowering Teams: Agile leaders empower teams to make decisions and take ownership of their work.
- Continuous Learning: They foster a culture of continuous learning, encouraging experimentation and adaptation.
- Collaboration and Trust: Building trust and collaboration among team members is essential for agile leadership.
What are the common pitfalls of agile transformations mentioned in Doing Agile Right?
- Misunderstanding Agile Principles: Adopting agile terminology without understanding its core principles leads to ineffective implementations.
- Overemphasis on Structure: Changing organizational structure without addressing cultural issues can hinder agile effectiveness.
- Top-Down Control: Maintaining a command-and-control approach undermines the self-management aspect of agile.
How does Doing Agile Right suggest scaling agile effectively?
- Start Small and Iterate: Begin with a few agile teams and gradually scale up based on what works.
- Create an Agile Enterprise: Transform the entire organization to support agile practices, aligning processes, structures, and cultures.
- Focus on Customer Needs: Prioritize initiatives that address customer needs and feedback to remain responsive and relevant.
What is the role of planning and budgeting in an agile enterprise according to Doing Agile Right?
- Adaptive Planning: Planning is a flexible process that adapts to changing circumstances, allowing quick pivots.
- Continuous Budgeting: Budgeting is ongoing, adjusting based on current priorities and needs for effective resource allocation.
- Prioritization of Initiatives: Focus on funding critical initiatives that align with strategic goals to avoid inefficiencies.
How does Doing Agile Right address the integration of agile and traditional bureaucratic structures?
- Complementary Approaches: Agile and bureaucratic structures can coexist, supporting both innovation and operational efficiency.
- Harmonizing Processes: Successful integration involves harmonizing processes across agile teams and traditional functions.
- Cultural Shift: A cultural shift is required, valuing both agile and bureaucratic elements, with leaders modeling this change.
What are the key metrics for measuring agility as discussed in Doing Agile Right?
- Comprehensive Measurement: Measure inputs, activities, outputs, and outcomes to understand agility.
- Customer Feedback: Prioritize metrics reflecting customer satisfaction and engagement.
- Continuous Improvement: Use metrics to drive continuous improvement, identifying areas for enhancement.
How does Doing Agile Right define agile budgeting?
- Outcome-Based Funding: Link financial resources to specific customer outcomes for accountability and value delivery.
- Continuous Adjustment: Flexible budgeting allows ongoing adjustments based on performance and priorities.
- Scenario-Based Funding: Use projections of potential outcomes to inform resource allocation decisions.
What is the significance of the agile operating model in Doing Agile Right?
- Holistic Framework: Encompasses structure, leadership, culture, processes, and technology for a truly agile enterprise.
- Alignment with Strategy: Aligns with strategic goals to ensure agility contributes to overall success.
- Customization: Tailor the operating model to the organization's unique context and needs for effective practices.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.