Level 3 of the Retailer S&T Tree

Now we will briefly describe Level 3 of the left side of the S&T tree. Recall that Step 2.1 was focused on changing to a consumption-driven mode of operation and that the steps in Level 3 need to explain how to build and sustain this DCE. Step 3.1.1, as shown in Table 34-4, explains how to build the DCE.

After reading part 1 of the PA, we point out that it is a huge mistake to push inventory into the shops. The result is shortages of some SKUs and surpluses of other SKUs. After reading part 2 of the PA, we point out that holding the inventory back in the supply chain is more effective because the forecast is much more accurate at the distribution center (DC). Keeping more of the inventory at the DC would result in fewer cross-shipments such as between regional DCs (RDCs). Part 3 is about placing daily orders with the suppliers. Most retail executives would point out that the suppliers would not agree to this. In reality, most suppliers are struggling with handling huge orders and spikes in demand. Daily orders and frequent replenishment is a win-win solution for both the retailer and its suppliers.5 After reading part 5 of the PA, we point out that inventory targets should be adjusted frequently because market conditions change often.

When more than one tactic is listed, they are listed in the order in which they are implemented. These tactics need to be implemented in order to achieve Step 2.1, but they are not sufficient. How to achieve Step 3.1.1 is explained in the three steps (4.11.1, 4.11.2, and 4.11.3) that are below it in the S&T tree. The SA in this step points out a reality regarding getting stakeholder buy-in and support of a new initiative. We have to ensure that the first step in the next level of the S&T tree results in a significant and quick impact on the performance of the company in order to get this buy-in and support. The three steps below 3.1.1 are focused on implementing internal pull distribution (Step 4.11.1), the TOC solution for replenishment, keeping correct inventory levels (Step 4.11.2), and dealing with suppliers (Step 4.11.3).

Step 3.1.1 is focused on building the DCE by ensuring existing SKU availability. Step 3.1.2 is focused on sustaining this edge by further protecting and improving inventory turns. The last step under 2.1, Step 3.1.3, is focused on continuing to build the DCE by improving TPS by changing the product portfolio.

TABLE 34-4 Step 3.1.1 of the Retailer VV S&T tree ( E. M. Goldratt used by permission, all rights reserved. Source: E. M. Goldratt, 2008).

General Overview of the VV S&T Tree Structure

Let us now discuss the S&T tree generically to Level 3. The Level 1 strategy was focused on achieving both growth and stability. Level 2 has steps for achieving both base and enhanced growth. In the Retailer S&T tree, the SA of Step 2.1 focused on building the DCE and sustaining it. In other VV S&T trees, the SA of Step 2.1 is slightly different from the one in the Retailer S&T tree in that it also addresses the need to capitalize on the DCE. This component was not needed in the Retailer S&T tree because it is enough for the customers to know that availability is now remarkable. In the other S&T trees, actions are typically needed to market and/or sell more effectively in order to capitalize on the DCE that has been built. Thus, Level 3 of each VV S&T tree consists of steps that are focused on building, capitalizing on, or sustaining the DCE.

Levels 4 and 5 of the Retailer S&T Tree

Next, we will look at the first step of Levels 4 and 5 of the Retailer S&T tree in order to understand an S&T tree better. Step 4.11.1, as shown in Table 34-5, explains how to improve the inventory turns through the implementation of the pull distribution solution of TOC.

Notice that the number of PAs tends to increase, the lower we are in the S&T tree. It is important to note that the PAs are not written as a bullet list, but rather as a presentation of cause-and-effect logic. The SA in this step also reinforces the SA of Step 3.1.1.

TABLE 34-5 Step 4.11.1 of the Retailer VV S&T tree ( E. M. Goldratt used by permission, all rights reserved. Source: E. M. Goldratt, 2008).

Two more steps in Level 4 are needed to achieve Step 3.1.1. Step 4.11.2 is focused on keeping the correct inventory levels through the implementation of the Buffer Management (BM) solution of TOC (which provides an effective priority management system), and expediting and adjusting for peak demand (as explained in the steps in Level 5). Step 4.11.3 is focused on how to deal with the suppliers in order to achieve much more improvement in the inventory turns and the bottom line (NP, ROI, and cash flow).

Step 5.11.1, as shown in Table 34-6, is the first step that needs to be implemented in retail. Because the level of sales changes over time in retail, we must ensure that we can prove that the increase in sales is the result of our initiative.

Level 5 is the lowest level of the Retailer S&T tree. Therefore, it provides the details that are required to be implemented in order to achieve Level 1 of the S&T tree.

TABLE 34-6 Step 5.11.1 of the Retailer S&T Tree ( E. M. Goldratt used by permission, all rights reserved. Source: Modified from E. M. Goldratt, 2008).

Need for Lower Levels of an S&T Tree

It is possible that for some steps in an S&T tree, Level 6 or even Level 7 needs to be written. At this time though, Level 6 has not been written for any VV S&T trees, although we believe for some steps it would probably be quite useful to write Level 6. Another level needs to be written in an S&T tree to explain how to implement the step above only if it is not clear in that step how to achieve it. In the Retailer S&T tree, there are no Level 5 steps under Step 4.12.1 or Step 4.12.3, which are both under Step 3.1.2. In addition, there are no Level 5 steps under Step 3.1.3.

Details Regarding the Structure of an S&T Tree

This section summarizes what we have covered about the structure of an S&T tree with additional content now that an S&T tree has been shown and explained. The S&T tree presents the logic for how to achieve a high-level strategy. Can we agree that a strategy is an answer to the question "What for?" In other words, what is the objective? And a tactic is the answer for "How?" Brushing our teeth is an action. Can we ask what for? Yes. Can we ask how? Yes. Trying to put a strategy up high in the S&T tree and tactics all lower in the S&T tree does not make sense. In other words, it is flawed to think that strategy is for top management and tactics are for lower levels of management in the organization. Every action that we take has a strategy and tactic. Therefore, we have a number of strategy and tactic combinations or pairs, which we refer to as steps. In actuality, each tactic is an action.

The S&T tree is read from the top down. Level 1 is the top of the S&T tree, which is only one step. Level 2 is the level below Level 1, etc. Each level of the S&T tree corresponds to a level of management in most cases. The strategy and tactic plan provides a more detailed explanation as we progress down through the layers of management. Level 1 is related to the Chairman, while Level 2 is for the Board of Directors (including the CEO). Level 3 is for the executive vice presidents (EVPs). Level 4 is for functional departments, while Level 5 is for the head of the department in the function. Level 6 is for the managers, while Level 7 (which may not ever need to be written) is for the individual employees.

We must ensure that responsibility and authority are aligned. The managers are responsible for delivering on the strategy of their assigned step. They also have the authority to change the tactics in the step for which they are responsible. For example, if the PAs are not facts of life in their company, the tactics need to be changed. We have to be careful not to have idiotic Draconian rules in the tactics. In other words, we should always check to see if the tactic we intend to implement is a logical derivative of the PAs. For example, if a tactic in the S&T tree states that 25 percent of the projects should be frozen, this rule should not be blindly applied, but rather modified in some cases to achieve the logic of the step. In some cases, it would make sense to freeze more or less of the projects in a particular company.

The S&T tree also has a sequence from left to right in a level. A step that is to the right of another step in the same level of the S&T tree cannot be implemented before the step to the left of it on the same level has started being implemented. The timing of when to start implementing a specific tactic needs to be based on logic. In some S&T trees, there is content in the step that points out when to start implementing the tactics.

Each step in the S&T tree includes some or all of the following statements in this order: necessary assumption(s), strategy, parallel assumption(s), tactic(s), and sufficient assumption. The NAs explain why the step is necessary (as part of the group of steps on this level that correspond to the step in the level above) to achieve the higher-level corresponding step in the S&T tree. Therefore, there are NAs listed in each step except for the step that is Level 1 at the top of the S&T tree. The NAs need to be convincing that the action must be taken by pointing out the damage of not taking the action and/or the benefits of taking the action. The sequence of NAs is in order of the concerns that people will have. The NAs in Level 4 are focused on what is currently being done and the need to do it differently. The NAs in Level 5 are about the difficulty in doing the tactic in Level 4-the tactic we already agreed to do.

One way to better understand what an NA is results from an explanation of necessity-based logic and a visual aid. In a conflict, we understand from the Evaporating Cloud (EC) tool of Thinking Processes (TP) that a need is what is desired from a want. In other words, an action that we want to take is really focused on achieving some need. The connection between the need and the want is a necessary assumption, the explanation why this want will result in the need being achieved. It can be read as follows: "In order to achieve the , we must because of ." Likewise, in an S&T tree, the NA provides the connection between one S&T pair or step and another in a S&T tree as shown in Fig. 34-3. It can be read as follows: In order to achieve Step 1, we must achieve Step 2.1 because of the NA of Step 2.1.

FIGURE 34-3 How the NA connects one step of the S&T tree to another.

The NAs are always generic. If the NAs we want to address are not generic, then we can devote a whole step in the S&T tree to exceptions. A step in an S&T tree can be just for a specific case. Another possible approach is to include exceptions in the PAs and the resulting tactics within a step already in the S&T tree.

A strategy is what we want to achieve. It is not stated as an action, but rather as being current reality. The parallel assumptions are for checking that the tactic will achieve the strategy. The PAs need to tell us how to do it; they are the most important part of the S&T tree because they provide the logic. The PAs explain the whole logic of the tactic-why the tactic has a real chance of making the strategy a reality. There are no surprises allowed in a tactic. The PAs must explain why the tactic is needed and will result in the strategy being achieved.

The PAs are written using cause-and-effect logic. They are written in the order that makes logical sense to reach the conclusion regarding what the tactic(s) must be. They are not written based on the sequence of the tactics. They are written so that the cause-and-effect logic is clearly presented. An entity (cause or effect) must be just one sentence. However, we can have more than one sentence in a PA when the next sentence is a comment. It is better to keep the comment as part of one PA; it will be shorter because some content will not have to be repeated. In addition, we can put both a cause and effect in a PA. In some cases, we have freedom where to put the assumption in the S&T tree; if we have a long explanation before a conclusion, it is better to put it in PAs instead of in the NAs.

A visual aid with explanations will be useful for further explaining the logic of how the PAs are written. Figure 34-4 shows a generic example with respect to sufficiency-based logic. Within the TP of TOC, we read sufficiency-based logic as follows: If A and B and C, then D is the unavoidable result. The oval shape represents a logical "and."

FIGURE 34-4 Sufficiency-based logic example.

FIGURE 34-5 Logic within the PAs.

Figure 34-5 shows an example of how the PA is written. As stated earlier, the PA presents cause-and-effect logic, not a bullet list. If the logic were the same as presented in the previous generic example of sufficiency-based logic, then the PAs would appear as shown in Fig. 34-5. Often, we would verbalize D as starting with a word such as "Therefore, . . ." to indicate that it is an effect or a conclusion that results from the previous statements of A, B, and C.

Every tactic must get results and is written as an action. The tactics are written in the order in which they are implemented. Note: The logic explaining each tactic must be clearly presented in the PAs. We can write, "Vast experience shows that . . ." in a PA if it will be explained lower in the S&T tree. The tactics in Level 5 explain exactly what to do.

It is not possible to come up with an SA that proves the steps below are sufficient to achieve the step above them. The only real test is reality. The solution was to have the SA highlight a fact that, if not dealt with by steps of the corresponding lower level group of that step, would result in sufficiency not existing. The SA must be something that is common sense, but is typically ignored; that if ignored will not result in sufficiency. They are "Confucius says" statements, which are generic. A step has an SA only if there is another level of the S&T tree written below that step.

Figure 34-6 will be helpful for clarifying the logic. The SA is based on sufficiency-based logic, as described earlier. One way to read it is as follows: If Steps 2.1 and 2.2 are achieved, then Step 1 will be the unavoidable result because the SA of Step 1 was a fact of life (assuming reality verifies that sufficiency was actually achieved). Another way to read it is: If Step 2.1 and Step 2.2 are achieved and the SA of Level 1 is a fact of life, then Step 1 is the result.

When preparing to write the next level of the S&T tree, we think of the how (the titles) and the why of the steps and the sequence of the steps. The next level below a step must have a minimum of two steps for it; otherwise, the content should be within the step itself. When creating the steps, we do not artificially break the content into two parts, but rather divide it logically into two or more parts. When I was writing Level 5 of the Retailer S&T tree, I had to figure out how to write the steps in Level 5 below Step 4.11.2 of Keeping Correct Inventory levels. I realized the way to logically divide it in Level 5 was to have one step explaining Buffer Management (BM), the mechanism that automatically ensures correct inventory levels, and to have another step specifically focused on adjusting for peak demand, explaining how to adjust the mechanism for upcoming expected spikes in demand that are due to a sale or other planned or known event. Later, expediting was also added as a step.

FIGURE 34-6 Visual aid for SA connection.

It is much easier to write the SA for a step after we have completed writing the steps below that step in the S&T tree; then, we have enormous intuition. The SA does not have to address all the steps below it, but it is better if it does. It is especially important to address the first step (the one to the left) of the corresponding steps below. It is better to take a known phrase/quote and change it to what we need. It makes the SA verbalization more interesting. An NA and SA can be essentially the same.

Note: We decided not to have sequence assumptions in the S&T tree. Therefore, we have to deal with any sequence that needs to be pointed out in another way. To do so, we write within a step that this step is dependent on another specific step in the S&T tree. An example of this is in Step 3.1.2 of the RRR S&T tree. This S&T tree starts with focusing on becoming a more reliable supplier by reaching at least 99 percent due date (on time) performance. It states as a note in Step 3.1.2 that the reliability selling offer should not be given the green light to start being marketed until after 99 percent due date performance has been achieved.

Key Concepts Regarding Creation of S&T Trees

We have been using S&T trees to guide TOC VV implementations for years now. VV projects are holistic TOC implementations. These VV projects consist of multiple, synchronized TOC applications (subsystem implementations), such as operations, distribution, project management, marketing, and sales. VV projects are consulting projects in which the target net profit in four years or less will be significantly higher than the best possible profit that top management of the company believes is achievable. S&T trees can be used to achieve any strategy of any system (such as any type of organization or for a person). Once an S&T tree has been developed, it is necessary to first verify that the strategy, which is one of a number of strategies we have, applies to the system (person or organization) for which the S&T tree is written. The assumptions really are facts of life. We need to validate that they are facts of life for a particular system. If they are not facts, then the corresponding strategy or tactic does not apply to the given system.

The S&T tree provides all the strategies and tactics needed to achieve the strategy in the step in Level 1. The S&T tree provides answers to the three questions in TOC regarding how to manage effectively: 1. What to Change?

2. What to Change to?

3. How to Cause the Change?

The S&T trees that were developed to guide VV implementations provided the first significant application of TOC to answer the third question above. The development of these S&T trees clarified the steps for successfully implementing TOC. The guidance provided in Level 5 of the VV S&T trees is very useful for implementing TOC in an application/subsystem (such as a function). Level 5 provides a simplified approach to implementation that ensures that only the injections (solution elements) required are implemented and provides information regarding the sequence of implementing those injections.

The project network/plan for implementing the VV project consists only of the tactics in the lowest level of the S&T tree. In other words, a network showing all the dependencies (both task and resource) will be created based on all the tactics in the S&T tree at the lowest level. Specifically, if a step does not include a Level 5 while others do, then we will have tactics in the project plan from both Levels 4 and 5. This project plan in VV implementations is created using the Critical Chain network (which includes all dependencies and appropriate buffers) and is managed using critical chain software (from Realization Technologies).

The VV S&T trees also include strategies and tactics for ensuring that significant negative branch reservations (NBRs) are trimmed, thus ensuring that any significant, potential negative consequences of implementing the solution are prevented. In addition, these S&T trees include strategies and tactics for ensuring that significant obstacles that may block or delay implementation are overcome.

The VV S&T trees are written to follow the plus buy-in process of TOC.6 The steps of the plus buy-in process are: 1. Agree on the very ambitious objective we desire to reach-a pot of gold.

2. Agree that reaching the pot of gold at the top of the cliff is much more difficult than we originally thought.

3. Agree that there is a direction for the solution, an anchor on the cliff against which a ladder can be leaned.

4. Agree on the solution details.

5. Overcome unverbalized fears, such as the potential NBRs of success.

The huge pot of gold is the strategy in Level 1. The strategy points out what the pot of gold is and that we can do it. It must be a strategy to which all would readily agree. Next we show how steep the cliff is (Step 2)-these are the PAs summarized in the tactic of Level 1. We show that we have nothing to hang on to in order get up the cliff. The PAs show how impossible it is to reach the pot of gold.

Step 2 of the plus buy-in process shows why it is not going to be easy to reach the pot of gold. The last PA of Level 1 usually shows why there is hope, but is not always in an S&T tree. The tactic states what will be done to reach the pot of gold, after establishing why this is the case in the PA's. The pot of gold may be a target they wanted to reach before, but decided was not possible to achieve. Step 2 is for generating credibility, to bring insight to the executives about how well we understand the problem of reaching the pot of gold. They will be more inclined to listen to us because they will assume we might have found a solution. The PAs establish the parameters of the solution-what the solution has to address.

The NAs in Level 2 of the S&T tree are the anchor for the ladder-the third step of the plus process. The rest of Level 2 is the silhouette of the anchor. Level 2 is about how to meet the needs of the stakeholders, such as those of the external market. We consider the purpose and essence of the organization to write this level.

Level 2 below the NAs and Level 3 (which is how to build, capitalize, and sustain the DCE in the VV S&T trees) of the S&T tree are the ladder (Step 4 of the plus buy-in process)-the major rungs of the ladder and major ways in which to break your legs (Step 5 of the plus buy-in process). First, we start with showing the rungs of the ladder to climb, and then we think of major ways in which to break our legs, especially ways that can happen when we are successful.

Level 3 provides the method for achieving the strategies and tactics, but no clue how to do it. In the VV S&T trees, Level 3 explains how to build, capitalize, and/or sustain the DCE. In the VV S&T trees, Level 3 typically starts with an implementation of one of the logistical solutions of TOC (Pull Distribution, Critical Chain Project Management, or the TOC production solution of Drum-Buffer-Rope/BM). Then, if needed, a marketing or sales solution of TOC is implemented to capitalize on the DCE achieved through the logistical solution. Finally, to sustain the DCE, we are focused on how to ensure that the performance does not deteriorate as sales increase.

Every time we go down a level in the S&T tree, we are detailing the steps of the ladder. Level 4 is the level in which we are making the switch to TOC through the "golden assumptions." These types of assumptions of Level 4 explain what we can do. In other words, we are moving from theories into practice (the actions move from being a direction to being practical actions). In reality, we tend to call everything above the golden assumptions strategy and everything below them tactics. This is why people have the belief that the strategy is for the high level of the organization and the tactics are for the low level. This explains how the S&T tree concepts fit with conventional views of strategy and tactics.

There are cases when Level 5 does not need to be written. Here are the considerations. Are there any real difficulties to do the step in Level 4? Are there fundamental concepts that must be changed? In these cases, we need to write Level 5. We should also think about typical mistakes that might be made in implementation. We check to see if the conventional way of doing the step in Level 4 would result in mistakes. Level 5 is written if a change in a key belief is needed. Level 5 presents the actions and the implementation issues. Once Level 4 has been validated, management has already agreed to make the required changes. The logic of the how to do it is in Level 5.

The criteria for judging a solution listed below were kept in mind when writing S&T trees.7 These criteria are listed in the order in which they must be considered.

1. Results in excellent benefits.

2. Is win-win-win for all whose collaboration is needed. This is important because collaboration results in an increased probability of success, and a faster and more sustainable implementation. When it is not win-win-win, forces will erode it over time.

3. The risk associated with implementing the solution, multiplied by the corresponding damage, is not small relative to the benefits of implementing the solution. This is about comparing the level of risk with the level of impact. For example, if the risk is small and the potential level of damage is high, while the benefits are huge we need to consider carefully whether we should implement the solution.

4. It is simpler than what we do now. Why is it important? The more complicated it is, the higher the chances of disillusionment of those needing to implement or support the change. If it is complicated, we do not know if it will work (if there is a chance of implementing it successfully).

5. The sequence of implementation is such that each action or cluster of actions leads to immediate, significant results, thus enabling getting everyone on board (their collaboration).

6. Does not self-destruct. If the solution self-destructs (is not sustainable), the company can be in much worse shape than it was before the solution was implemented.

All of the VV S&T trees ensure that the constraint (the factor most limiting the ability of the organization to achieve its goal) is in management's control-it is the rate at which the company can grow. The VV S&T trees are focused on achieving a DCE based on meeting a significant need of the clients. As a result, there is no limit to growth except the rate at which management will choose to grow. The S&T tree includes actions to ensure the constraint does not become internal (such as in production or sales) or a market constraint (the level of demand).

How the S&T Tree Relates to Other Thinking Process Tools of TOC

How does the S&T tree relate to the other tools of the TP of TOC? The S&T does not replace the Current Reality Tree (CRT)-the map of all the cause-and-effect connecting the core conflict or root cause to all the undesirable effects (UDEs) in the system-and EC (conflict resolution tool) at all. The S&T tree does include some of the elements of the CRT to show why we used a different direction of a solution than the conventional approaches. The core conflict is addressed in Level 1 of the S&T tree in the PAs. The tactic of Level 1 states that we will achieve both needs of the core conflict without a compromise. The NAs of Level 2 are the assumptions that we are invalidating underlying the core conflict. The core conflicts of the subsystems are also addressed in the lower levels of the S&T tree.

The S&T does not replace the Future Reality Tree (FRT)-the logical map connecting all the injections (solution elements) through cause-and-effect to the desirable effects (thus ensuring no UDEs of the CRT continue to occur). The S&T tree does include all of the injections that are in the FRT.

All of the assumptions in the S&T tree must be facts of life from the CRT and FRT of the system. In other words, the assumptions must be verbalized as fact based on the current cause-and-effect logic of the system. When writing an S&T tree, it is best to conduct a full TP analysis first before writing any of the S&T tree. The core conflict, CRT, and FRT are invaluable in terms of more quickly and more effectively writing an S&T tree.

The S&T tree does replace the Prerequisite Tree (PRT) because the S&T tree addresses the obstacles and how to overcome them. The S&T tree provides much more logic and content than the PRT for causing the change. Two main advantages of the S&T tree to the PRT are the ability to distinguish between the big picture and various levels of detail and the ability to ensure that chupchik (unimportant details) are not included in the plan. This does not mean that the PRT should not ever be used. It can still be used as an effective tool for figuring out how to reach an ambitious target by determining the obstacles to reaching the target and how to overcome them. The Transition Tree (TRT) is not yet replaced by the S&T tree. The S&T tree might replace it after Level 6 has been written or sequence assumptions have become part of the S&T tree. This does not mean that we would no longer use TRTs, but rather that we wouldn't need to use them when we have an S&T tree.