In general I found this a very useful book as my experience has been that many agile implementations stumble when they start working more corporate or financial issues. Part of the reason is that it is easy for many to say “The agile stuff is for the IT or product development folks - it doesn't affect me” when in fact you can stifle a new way of working by not looking at what needs to be changed at higher levels. In many ways the book represents an agile approach to finance, and serves as a good primer for a discussion many change agents will get into.

The book talks about the problems with traditional financial thinking and works to address some of those issues. Starting with an assumption that “Today, there is so much more VUCA out there: Volatility, Uncertainty, Complexity, and Ambiguity”.

The assertion is means organizations need to be built to respond to VUCA but there is a problem in that there is a traditional budgeting process in the way. The following problems are noted with this process:

• “Weak links to strategy: as the strategy and the budget are developed in isolated processes, facilitated by different functions without much mutual respect and contact.
• A very time-consuming process: Budgeting consumes a scary amount of time and energy, both when made and when followed up.
• Stimulates unethical behaviors: The gaming, lowballing, and hidden agendas that normally would not be accepted are seen as normal and unavoidable in a budget regime.
• Assumptions quickly outdated: Many, and sometimes most, of the budget assumptions turn out to be wrong.
• Provides illusions of control: Most of the controls the budget offers are nothing but illusions of control.
• Decisions are made too early: Decisions about activities, projects, and spending are typically made too early, without fresh enough information to make the right decision.
• Decisions are made too high up: Lack of autonomy forces decisions upstairs, often making them worse, not better.
• Often prevents the rights things from getting done: “I can't do the blindingly obvious, because it's not in my budget!”
• Often leads to the wrong things being done: The flipside is people doing what they shouldn't, because it is in the budget - “Spend it or lose it!”
• The world ends December 31: The budget year creates shortsightedness and a start/stop rhythm, which is often artificial from a business perspective.
• A language ill-suited for performance evaluation: “Hitting the budget number” is a narrow and often meaningless way of defining performance.”

The reason this is the case is that the budget process serves goals:

• Good targets
• Reliable forecasts
• An effective resource allocation

And these goals conflict with each other. For example, the sales manager knows that the forecast number will come back as a sales target, and maybe with a bonus attached? We shouldn't be surprised if a lower number comes up. Is this a good forecast number?

What is needed is a radical change in thinking. One idea is to separate the 3 concerns above so that one does not drive the other.

In addition changes needed to be made “both in leadership beliefs and behaviors and in management processes. On the leadership side, we need to be more values-based than rules-based. This does not mean having no rules. It simply means that the stronger our values are, the fewer rules are typically needed. There also has to be more autonomy. Bringing all decisions nine floors up takes too long in a VUCA world and doesn't necessarily improve them. Often it is the other way around. We also need more transparency. As discussed, transparency can be a very effective control mechanism. This should be good news for the many managers who are afraid of leaving traditional management because they are afraid of losing control. The fear might be deep and heartfelt even if much of this control is nothing but an illusion of control. Finally, it is about focusing on the internal motivation instead of leaving it all to the simpler but less effective external motivation, as discussed in the previous chapter. On the management process side, the traditional budget typically needs to go or at least be radically changed. Relative targets should replace absolute targets where possible and where it makes sense. The management process rhythm should be more event-driven, based more on business than on calendar rhythms, in everything from target setting to forecasting and resource allocation. A broader and richer holistic performance evaluation instead of only narrow and mechanical measurement should also be introduced.”

Driving leadership principles:

• Purpose - Engage and inspire people around bold and noble causes, not around short-term financial targets.
• Values - Govern through shared values and sound judgment, not through detailed rules and regulations.
• Transparency - Make information open for self-regulation, innovation, learning, and control; don't restrict it.
• Organization - Cultivate a strong sense of belonging and organize around agile and accountable teams; avoid hierarchical controls and bureaucracy.
• Autonomy - Trust people with freedom to act; don't punish everyone if someone should abuse it.
• Customers - Connect everyone's work with customer needs; avoid conflicts of interest.

Management Processes

• Rhythm - Organize management processes dynamically around business rhythms and events, not around the calendar year only.
• Targets - Set directional, ambitious, and relative goals; avoid fixed and cascaded targets.
• Plans and forecasts - Make planning and forecasting lean and unbiased processes, not rigid and political exercises.
• Resource allocation - Foster a cost-conscious mindset and make resources available as needed, not through detailed annual budget allocations.
• Performance evaluation - Evaluate performance holistically and with peer feedback for learning and development, not based on measurement only and not for rewards only.
• Rewards - Reward shared success against competition, not against fixed performance contracts

Based on these core ideas the book takes you through the journey a couple of organization took to revamp their processes. For example, the idea of investment allocation: “Investment Management Capital allocation without annual budgets was almost a no-brainer. The starting point was the relative RoACE, with profitable investments as one of several levers available for the business units to improve their performance. We created three investment categories:

1. Small projects were treated more or less like operating costs, even if accounting-wise they classified as investments.
2. Medium-size projects were decided case by case. The hurdle rate (the “interest” that future cash flows are reduced with because a dollar tomorrow is less valuable than a dollar today) was meant to be the lever to pull if we needed to take down the total investment level in this category. In practice, this mechanism was never used explicitly. When money was tight, there were simply fewer proposals coming forward.
3. Major strategic projects had never been managed through annual budgets. These projects could be acquisitions, new plants, or other major capital commitments. Such decisions seldom fitted any autumn decision scheme. They were made as needed, using the best and most recent information we had on commercial assumptions, financial capacity, and the project itself.”

In terms of performance evaluation, one of the problems with is that the process serves different and conflicting purposes, just like the budget does. These purposes are:

• Feedback and development
• Reward
• Legal documentation

Additional ideas on the problems:

• “According to Washington Post, 10 percent of the Fortune 500 companies have now abolished the traditional annual performance appraisal, including Microsoft, Accenture, Deloitte, and Expedia.
• Fifty years of research almost unanimously discounts individual bonus as an effective way of motivating and driving performance in knowledge organizations
• A bonus can undermine the interest in the job itself and reduce the value of the task it pays for, even though the intention is the opposite.
• You can kill interest by rewarding people for something they used to do without a reward because they thought it was fun, the story also reminds us that incentives do not create any lasting and sustainable change in behavior unless you keep paying up.
• David Sirota sees it like this: ”The main question for management is not how to motivate, but rather how management can be deterred from diminishing or even destroying motivation.
• By the way, if bonus is meant to motivate, how come the biggest dose is needed at the higher manager and executive levels? Is this where we find the most boring jobs? I just don't get it!
• Team or collective bonuses are very different, as they are designed with a different purpose: hindsight reward for shared success. Note: Individual bonus trying to push behavior. Group behavior looking back rewarding success
• Free riders are real. They exist in every company and in many teams. But they are still a small minority. Again, we cannot design our management processes based on minorities.
• The Institute of Leadership & Management (ILM) recently did a study on the use of bonuses in British companies. It seriously questioned the business value of the £37bn companies spend on bonus annually, as only 13% of respondents said the bonus made them work harder. Note: Numbers. Bonus don't work. And are expensive
• Tough times are used as an excuse for increasing bonus levels, to motivate executives for all the difficult decisions ahead. As if it isn't their job!
• A survey by the U.S. compensation and benefits consulting firm William M. Mercer sums it all up when concluding that “most merit or performance-based pay plans share two attributes; they absorb vast amounts of management time and resources, and they make everybody unhappy.”

On changing the process: “A key lesson learned came as quite a surprise to us: Don't design everything up front. The issues and challenges might pop up in the most unexpected areas. Design to 80 percent and jump, and sort out the issues as they occur”

On the “balance scorecard” approach: “A scorecard can protect and reinforce a command-and-control regime, or it can do the opposite. Too many companies are in the first category. We aim to be in the second.”

On setting a strategy “So next time you listen to someone presenting their strategy, ask what they have said no to. If they can't answer, there is no strategy.”

On characteristics of a good KPI “Do they measure progress toward strategic objectives? Do they measure real performance? Is there a good mix of leading and lagging indicators? Do they address areas where we want change or improvement (or is monitoring sufficient)? Are the KPIs perceived as meaningful at the level they are used? Can data be collected easily?”

On comparing metrics “One thing must always be in place for benchmarking and league standings to work. They must be seen as fair and relevant. If this is not in place, forget it” and “One solution is “indirect benchmarking”: comparing how well each unit improves their own performance. This gives everybody a more common and fair starting point”. In other words an indirect benchmark compares improvements not in comparison to others.

On planning “Planning is about: Which actions do we need to take in order to deliver on strategic objectives and KPI targets? What are the expected consequences of these actions, expressed as a forecast, either against KPI targets or in other financial or operational areas where we need to understand what lies ahead (e.g., financial capacity)?” and “Planning is not about target setting, because this lies behind us. Neither is it about resource allocation, because this is handled in a separate process.” Planning is also not about forecasting.

On forecasting “A forecast is not a promise, not something to deliver on. People using that expression have not understood the difference between a forecast and a target. Again, a forecast is what we think will happen, an expectation; a target is what we want to happen, an aspiration.” By example “Assume we are out sailing. The target is clear; we want to reach the next harbor. Our radar screen provides us with forecasts of what lies ahead, and now it tells us that we are on course to hit a rock. That is definitely not a forecast we want to deliver on! On the contrary, we want to do whatever it takes to avoid hitting the rock. We use forecasts in our decision making to help us hit harbors and not rocks. Some would call “on course to hit a rock” a bad forecast. Assuming it is true, it is a good forecast even though it contains bad news. The only thing that can turn a good forecast into a bad one is if we get it too late, leaving no time to respond. Leadership behaviors are often to blame for good forecasts becoming bad ones. What kind of reception do we usually see when people bring good forecasts with bad news? There is seldom applause; on the contrary! Such experiences hardly encourage anyone to do it again, or do it early rather than late, which often means too late.”

On traditional cost control “Instead we want people to ask, “Is this the right thing to do? What is good enough? How is this creating value? Is this within my execution framework?” If in doubt, a good test is to imagine that the answer was yes and the money was spent. If someone later questions your decision, how comfortable will you be in defending it? If not, think again, or discuss with someone.” Instead of asking “do we have budget for this” ask harder series of questions with every cost decision. Every cent. Not just in November and December when budget is running out.

On traditional reporting “We spend less time on number crunching and historical variance explanations.”

On why innovation requires different thinking “Imagine a project assignment where the task is to develop the world's best soup. Obviously, there is quite some uncertainty here: which ingredients, when will it be ready, and at what cost? Innovation, experiments, and creativity are key for success. Then imagine a second assignment. The recipe is ready and the task is to produce 100 identical portions of the soup. Would we apply the same methodology for these two projects? Obviously not! This is where it went wrong for IT many years ago, when opting for the manufacturing or replication approach for software development. This kind of work is not about replication; on the contrary, there is significant uncertainty and all variables—content, time, and cost—can't be closed upfront. At least one must be left open. While Agile should be the process for developing the soup, Lean can be a great concept for the manufacturing process as the uncertainty is gone and the outcome is defined.” Note: do not agree with characterization of lean this way.

On process to get started “I have grouped it as follows: Create the case for change. Handle resistance. Design to 80 percent and jump. Keep the cost focus. Don't start with rolling forecasting only. Involve HR and Agile IT. You can't get rid of command and control through command and control. Do not become a fundamentalist.”

On selling “I often recommend to start with the quality problem, with the separation of the three budget purposes, especially if the organization is dominated by engineers, finance people, and a lot of rational thinking and problem solving. The simple fact that an ambitious target cannot be the same number as a 50/50 expected outcome forecast has a mathematical ring to it. That a cost forecast that doubles as a request for resources seldom will be a good forecast is quite obvious and most people get it right away. Both are simple to explain and can be illustrated easily with concrete examples from actual budget and planning practices in the company. Remind people that separating and improving only can make things better, and that we are not stopping what the budget tried to do for us. Add on some calendar examples from the rhythm problem. The light will most likely go on, and people will see something they have always sensed but not fully understood.”

“Formula for Change” created by David Gleicher and later refined by Kathie Dannemiller. It says little about sequence but highlights critical elements for change to take place. It reads as follows:

organizational_change = dissatisfaction x vision x first_steps

For organizational change to take place, there must be dissatisfaction with the current situation; there must be a vision of something better; and there must be some first tangible and credible steps toward it. The product of the three must be bigger than the resistance to change. If any of the three is low or zero, the resistance will normally be bigger and kill the change.

• Implementing Beyond Budgeting: Unlocking the Performance Potential