Harvard Business School Confidential
Page 9
Within business. Whenever there is a conflict for resources—such as what data to analyze first, what project to do first, what topic to discuss first, and how to avoid digression during a meeting—it is useful to ask, “What is the big picture?” and “What is the central issue?” This often helps prioritize data, projects, and discussions.
Setting Expectations
This is a true story: When I was a new consultant, my boss asked me to get some data for him in a week’s time. Eager to please, I told him I could give it to him in half that time. Unfortunately, it took longer than I thought. After three days, I told him it would take me another two days. But I was confident I could finish before the original deadline he’d given me. I was expecting that he would be OK with that since I would still be ahead of what he originally asked for. But he was furious. Because I had told him to expect the data in three days, he had planned on getting the data early. His plans would need to be totally revamped because of my delays. He was upset and so was I.
From then on, I never forgot the importance of setting the appropriate expectations. Two rules I now abide by at all times:
Better be safe than sorry. It is better not to be overaggressive in setting expectations. Be a pleasant surprise rather than an unpleasant disappointment.
Work often takes longer than seems likely up front. In my case, I find the 1.5-times factor very accurate. That is, if I think a piece of work should take Y number of days, the actual time needed is usually Y multiplied by 1.5. This factor is not scientifically proven and may be specific to my own style. You may want to find your own factor.
Minding the 80/20 Rule
There is a rule of thumb that has worked time and again for me. This is the 80/ 20 rule, which says 80 percent of the total output is or can be generated by roughly 20 percent of the total input. For example, the following observations are probably true:
80 percent of your company’s sales come from the top 20 percent of your customers.
80 percent of your company’s revenues come from 20 percent of your sales force.
80 percent of your payroll goes to 20 percent of your staff.
The 80/20 rule was first developed by Vilfredo Pareto (1848–1923), a French-Italian sociologist, economist, and philosopher. While researching economic conditions in his native Italy, Pareto determined that 20 percent of the population owned 80 percent of the land. Subsequently, while working in his garden, he discovered that about 80 percent of his peas came from just 20 percent of his plants. Based on these and other observations, he determined that for any series of elements under study, a small fraction of the number of elements usually accounts for a large fraction of the effect. Over time, Pareto’s observation became generalized as the 80/20 rule.1
Naturally, 80 and 20 are not rigid numbers. These are estimates that mean that the bulk (usually over 60 percent) of results can be attributed to a small fraction (usually less than 40 percent) of the input.
The 80/20 rule is extremely helpful for prioritization. It means that the top 20 percent should be given the most attention–it can be the top 20 percent of the customers, 20 percent of the data for analysis, 20 percent of the staff who are getting the highest pay.
An example from my early consulting experience solidified my belief in this rule: I was working for a relatively unsophisticated client in China. I found that its sales department evenly distributed sales staff time among the customers, so each customer got the same amount of time and service from the sales staff. As a result, each salesperson was assigned 10 customers and instructed to serve all of them equally well. While the company overall was profitable, revenues and profitability were not growing. When we interviewed the major customers, we found that they felt my client’s sales staff was “all right” but not “stellar” in their service. They did not feel my client valued them. Therefore, even though they continued to buy from the client, they were also giving business to other companies that were giving them “similar or better service.” When we interviewed the small customers, they were very impressed by my client valuing them as if they were a big customer. However, although they wanted to buy more, their size did not allow them to increase their orders by much.
In addition to ineffective customer service, such allocation of resources is also inefficient from a customer profitability perspective, as shown in Figure 5.1.
In the end, the study recommended a reallocation of resources—80 percent of the sales force time should be spent on the 20 percent of major or growing customers.
Another example of the 80/20 rule came up in my recent study on a property investment company, as shown in Figure 5.2.
Figure 5.1 Return on Sales Effort
Low-Hanging Fruit
Sometimes quick results are better than big results. This can be due to a variety of reasons: the need to build credibility for yourself, your team, or your project; a business crisis such as the need for immediate cash; morale; or too much time and effort needed to get big results. In such cases, instead of applying the 80/20 rule, which focuses on the biggest impact, it may be more effective to focus on capturing the “low-hanging fruit.” As the term implies, it means exploiting the opportunities that are easiest to capture. For example, I was once involved as the consultant for reengineering a state-owned petrochemical company in one of the Asian countries. The company was experiencing explosive growth and found that it could not hire enough qualified people to support that growth. The goal of the project was to redesign key processes and functions to make better use of the available human resources. The company had about 20 offices and refineries across the country. When planning the project, we had a few options. Here is a simplified summary of the options we considered:
Figure 5.2 Rental Returns
Options Key Considerations
1. Start with the largest offices and refineries 80/20 rule
2. Start all at the same time To shorten the total duration of project Find “low-hanging fruit” that can help build credibility for project
3. Start with a small refinery that has outdated processes and a significant manpower shortage and then roll out to rest of company
In the end, we chose approach 3. This is because some senior client executives were skeptical of the project and the involvement of expensive external consultants. By quickly achieving results in a small refinery, we built credibility that made it feasible to get support for a rollout to the rest of the company. We were also able to gather some best practices that were useful as we rolled the project out.
PLAN B
It is obvious that many factors are beyond our control, both in our personal lives and in business. Plan B means to have a backup plan, to have a plan for the worst-case scenario, to have a plan to turn to when your assumptions turn out to be wrong. It can be completely different from the original plan or it can be a moderation, change, or adaptation of the original plan should that plan become infeasible. While this seems logical and natural, the reality is that Plan B is often either forgotten or not done properly. This can be because of negligence, wishful thinking that Plan A will work just as planned, or lack of resources.
It is true that Plan B, when developed, is very often not used. But Plan B is like car insurance. Many people have the insurance but never claim enough to cover the premium paid. This is how insurance companies can make money. However, you do want the insurance just in case you get into an accident. The chance of a major accident may be slim. But you want insurance just in case it happens. This is the rationale behind insisting on having a Plan B.
I keep a story and an adage in mind to remind me of Plan B. The story (which is true): There’s a U.S. television reality show called The Apprentice, hosted by the high-profile real estate tycoon Donald Trump. Each season, a group of contestants including business school graduates, entrepreneurs, lawyers, and so on have to solve many business challenges under a tight time frame. Contestants are eliminated until the final two. Then there is a final challenge and a winner is selected fr
om the two. The winner will get an apprenticeship from Trump.
In the season finale, the two finalists were asked to each plan a big charity fundraising activity. One finalist was an HBS graduate with a doctorate from another leading school, who had been outstanding throughout the competition. He was definitely the front-runner going into the finals. He was asked to plan an outdoor charity event within a very limited time. While he was planning, somebody informed him that the weather forecast indicated possible rain on the day of the event. However, working under tremendous time pressure, he ignored the information. The day came and rain was pouring down. He had no Plan B. He had to scramble to change everything in his Plan A. Needless to say, he paid a high price for the mistake.
In addition to this story, I find Murphy’s Law (an oft-quoted Western adage) useful as a reminder of the need of a Plan B. Murphy’s Law has many versions, but the one I go by whenever I plan is “What can go wrong will go wrong,” especially when you least expect and can least afford it. Of course, I don’t mean you have to have a Plan B for everything that “can go wrong.” That would be too arduous and too costly. The factors to consider in determining whether a Plan B should be done and how much effort should be put into it can be expressed in a qualitative formula:
With a “qualitative formula,” hard data and numbers are not necessary. It is just an aid to a qualitative decision. The formula is self-explanatory:
The larger the possible impact if certain assumptions or parameters do go wrong, the bigger the need for a detailed Plan B. Using the story of the would-be apprentice, it should have been obvious that if it rained, it would be impossible to use the outdoor venue and setup and to do any of the outdoor events planned. So the impact would have been major and a Plan B for bad weather should have been developed to a significant level of detail.
The higher the probability for certain assumptions and parameters to go wrong, the bigger the need for a detailed Plan B. Again, using the same story, given the weather forecast, the probability of rain was high and a Plan B was called for.
The smaller the effort for Plan B, the bigger is the argument to have a more detailed plan B. If a detailed Plan B requires millions of dollars and significant time and energy to plan, it may be natural to take the risk with only a very broad Plan B. One option is not to have a specific Plan B, but to assign resources that can be deployed when necessary. A client of mine always appoints a “Navy Seals” team for its major projects. The team is ready to be deployed to solve any unplanned problems for major projects.
The final decision on the need for Plan B is a judgment based on putting all three factors together.
ASS-U-ME—TRITE BUT TRUE
I learned one thing early on in my career: if in doubt, ask. For many different reasons, including overconfidence, lack of confidence to ask, or simply lack of time, many people have a tendency to make unfounded assumptions. Whenever I make assumptions, I always remind myself of some words of wisdom I learned from my superiors very early on in my consulting career: “Do you know what happens when you assume? You make an Ass-(out of)-u-(and)-me.” Needless to say, these words of wisdom came after I made some poor assumptions in some consulting projects. You can imagine the damage to your credibility if you are questioned on your assumptions and you cannot defend them, as in this exchange:
CLIENT (OR YOUR SUPERIOR): I see in your strategy study you assume inflation of 5 percent a year in the next few years. What makes you say that?
YOU: That’s the historical inflation rate.
CLIENT: Why did you assume it will be the same in the next few years?
YOU: (silence because any answer will look bad: I had no time, I was sloppy and did not think hard . . . )
Therefore, it is important to remember that any assumption can be questioned. So, whenever you make an assumption, you must ask yourself— can I defend this assumption? Possible defense includes a credible source (such as government or expert published research data, or interviews with experts or the client’s senior management) or a clear logic (such as the historical rate is studied over 50 years and is a good medium-term average, or sensitivity analysis indicates that inflation has very little impact on our decisions and hence we did not spend too much time on it). If you have a source or logic, others may still disagree. But their disagreement will lead into a discussion designed to refine the assumption, and you will not lose credibility as a result of making it.
Note
1. Ethan M. Raisial and Paul N. Friga, The McKinsey Mind (New York: McGraw-Hill, 2002).
PART II
OPERATIONS
6
PROCESS
THERE IS A PROCESS FOR EVERYTHING
Part II of this book is about HBS’s skills and tools for managing operations. Instead of immediately going through each function, this section starts with the discussion of an overarching operations management tool called process management, which is powerful for thinking-through and optimizing any function. After process management, I discuss some key principles in managing human resources, marketing, sales, and finance that are critical but not covered by process management.
Formally, HBS defines a process as “any part of the organization that takes inputs and transforms them into outputs of greater value to the organization than the original inputs.”1 Informally, I find it useful to think of process a series of steps, supported by the appropriate tools and systems, by which work gets done.
At the functional level, a significant amount of work gets done through processes. Here are some examples of typical processes in various functions:
In human resources: recruitment and hiring, annual evaluation, sick leave application
In finance: expense reimbursement, budgeting
In manufacturing: production, materials planning
In sales and marketing: annual marketing planning, pre-sales, sales, post-sales
In back office operations (as in a bank): loan approval, bad loan collection
In all these examples, a series of steps, employing various supporting tools and systems, need to be taken to complete the work. By being able to see and understand work done as a process and to analyze key processes, ineffective and inefficient steps, policies, tools, systems, and paperwork can be identified and improved where possible. This skill is sometimes referred to as “process reengineering.”
The key tools for understanding and analyzing a process include a set of best practice principles that help question and identify inefficiencies and ineffectiveness, and a process mapping technique that makes it possible to analyze processes using those principles.
BEST PRACTICE PRINCIPLES
This section lists the best practice principles that I have found most useful when analyzing process efficiency and effectiveness. An efficient process is one that is low cost and fast. An effective process is one that delivers the intended outcomes. The first three practices listed here are more related to efficiency and the last two are more related to effectiveness. I start with the efficiency-related principles even though the effectiveness-related ones are more fundamental to a process because my experience has shown that the former are easier to apply and more frequently applied. I have phrased the principles in the form of questions so it is clearer how they can be applied when analyzing a process.
Are the steps in the process done right the first time?
Is waiting time minimized?
Is the weakest link as strong as it can be?
Is the process doing what it is supposed to do?
Is authorization appropriate?
Are the steps in the process done right the first time?
Work should be done correctly the first time, and mechanisms should be built in to help ensure correct action. This will reduce the need for quality inspection, rework, waiting time for rework, wastage, and so on, all of which are low-value-added and increase the time and cost needed to get the process completed. checking and even correction mechanisms designed and built into the process can be very
powerful in ensuring “right first time.” The spelling checker and automated spelling correction in Microsoft Word is an example. other typical examples for white-collar work are templates and checklists. Examples for manufacturing include color coding and workstations designed to prevent workers from making mistakes.
Is waiting time minimized?
Waiting time means time has passed with no value being added to the products or services that need to be produced. For a manufacturing process, waiting time will mean larger work-in-progress inventory. Inventory means the cash invested in making the inventory is not yet ready to be converted into income. Inventory also incurs interest costs because of the money invested in it. outside manufacturing waiting time means increasing the amount of time needed to complete a process. This means low efficiency and poor customer service if the waiting time affects the customer.
Is the weakest link as strong as it can be?
When I took the HBS class on operations, we were required to read a best-selling business novel called The Goal. The book was almost 400 pages long and the first 200 pages were more or less trying to illustrate a single basic point:
capacity of a process = capacity of its biggest bottleneck
A bottleneck is a factor that limits production.1 In other words the weakest link (bottleneck) of a process will limit and hence determine how much a process can produce. A simple example: say the process of publishing this business book has only two steps: writing the book by me and printing the book by my publisher. I can write a book every two years. My publisher can get any finished book printed within one month. However, even though my publisher can print in such a short time, the process is limited by my speed. I am the bottleneck. If I’m the only author on tap, the publisher can put out only one book every two years because of my speed, as illustrated in Table 6.1.