Use by government-sponsored manufacturing consultants (e.g., Switzerland’s CCSO) to help small companies develop strategic plans for restructuring (in one case, the tool helped the consultant to understand that the company’s initial strategic plan was unlikely to succeed until management agreed to reduce the amount of variation that it allowed in its process);

Use by large software vendors (e.g., EDS) to help a company decide to not relocate its plant from one foreign country to another (because the expense of closing the "gaps" created by the move were likely to be too expensive);

Use by a large manufacturing company (General Motors) to decide if a joint venture plant was ready to be opened (they decided on delaying the opening because the tool helped to surface differences of opinion in how to manage the workforce);

Use by a small manufacturing company (Scantron) to decide if their best practices needed improving (the tool helped them to discover that while they did indeed have many best practices, they needed to involve the workforce more closer with the supplier and customer-base, an action they subsequently took);

Use in a large technology change effort at a large manufacturing company (Hewlett-Packard) to help identify the workforce and organizational changes needed for the new production technology to operate correctly (resulting in a substantial improvement in "ramp-up" time when the new product and production process was introduced).

Use by a redesign effort of a maintenance crew (at Texas Instruments) to determine that the team-based approach they had envisioned needed several important improvements prior to startup

Use by a strategic planning committee at a large manufacturing company to identify areas of misalignment among elements of a new strategic plan (in this case between quality and throughput time).

Use by a manufacturing division manager to verify his current business strategy that had been given to him by his group manager. As a consequence of using TOP Modeler™, he discovered that he had agreed to a business objective of "new product development" without having the authority over the necessary people, skills, and other resources to deliver on that objective. He went back to his group manager to renegotiate these resources.

1) The tool has a variety of different users and uses, many of which we did not initially foresee. This means that there doesn’t appear to be a single expected use and benefit from the tool. For some users, they find this frustrating since they are used to simulation packages which provide a finite and well-defined set of functionalities. We find that the most successful users of TOP Modeler are those who begin their process of using TOP Modeler by first trying to identify the many ways in which it might be useful in their organization and then slowly trying different pilots to see which ways provide the greatest benefits for that organization at that point in time.

2) TOP Modeler™ is not currently coupled with a training class or consulting service to help you determine how best to use it in your organization. We have found that some users are better at determining how to integrate TOP Modeler into their organization than others. We have also found that some users don’t want to actually interact with the tool, but just to have someone use the tool to conduct an TOP analysis of their organization. In both these situations, we suggest that the organization assign someone the responsibility of learning not just the tool but also the knowledgebase and alternative ways of using the tool in their organization. We have found that people best suited to accept this responsibility are people who: a) have a deep understanding of their organization (including its strategy as well as its operation), b) have a basic understanding of the principles of organizational design and human behavior, c) have experience successfully translating externally-derived information about possible process improvements (from books, consultants, site visits) into organizational use, d) have the responsibility and authority to suggest process improvements, and e) have good group process and presentation skills. We have also found that successful users explicitly have discussions about alternative ways to use and diffuse the tool with managers and workers in the organization. These alternatives may range from use by a single engineer on an as-needed basis to wide diffusion to all workers to use as preparation for continuous process improvement teams and team meetings. These discussions about how to use TOP Modeler™ in the organization should be facilitated by the identified person since they are often quite value-laden. For example, we found one manager who refused to allow workers unsupervised access to TOP Modeler because, in his words, "they might begin to believe that they know more about organizational design than I know and question my authority; I don’t want that; that could lead to chaos". We found that having these discussions early on helps to clarify expectations so that these latent values do not come up later to kill the pilot experiences for seemingly unrelated reasons.

3) When you conduct a pilot experience, don’t be surprised if, at the end, users report that TOP Modeler™ did not "change" their ideas on the way they would improve their organization. We found several examples where discussions (such as among managers and/or sociotechnical or organizational consultants) would be contentious and frustrating about a particular topic (such as problems in strategy implementation). TOP Modeler tool would then be used to help the participants reformulate their strategy in greater detail. Outside observers would then report that the discussion were now more productive, leading to a quick consensus and action plan. Yet, when asked if TOP Modeler™ helped, the participants would report "No". Upon further examination, we found that some users believe that the knowledgebase of TOP Modeler cannot publicly provide any value-added to them since it would then suggest that they could not perform their own work adequately without it; and no one who is proud of their ability to redesign organizations wants to believe in any dependence or help from a machine! We have found that you MUST act on this issue in 2 ways:

First, it is very important to identify a role for people who believe that they are expert organizational designers (be they managers or consultants) since they will quickly sabotage otherwise. Ways of doing this might be to integrate TOP Modeler™ into an existing organizational change process which is controlled by the experts, helping experts understand how the tool can be used to more quickly and widely disseminate their personal knowledge, and more clearly identify the boundaries between the expert’s organizational change efforts and those of the users of TOP Modeler.

The second action you must take up front is to agree on the basis for evaluating the success of TOP Modeler’s use. We have used a variety of different metrics for success. One metric is: did the user learn anything? We have found, in a systematic study of 30 users, that 80% reported learning something about TOP integration or their own organization from using it. Another metric is: did the user learn what they had expected to learn? We found that only 50% reported yes to this. Another metric is: could the user generate action plans from its use? We have found in that same study, that 100% of the users were able to generate action plans as a result of using TOP Modeler™. Another metric is: "were the action plans implemented?" . We have found that, of those cases we know of, about half report implementing the action plans. Therefore, it is clear that the metric you select will affect how successful the pilot will be.

4) TOP Modeler™ is NOT an expert system; it does not tell you what to do. Instead, it provides you with different ways to obtain a prioritized list of features (called "gaps") that need improvement. TOP Modeler™ does not tell you how to fix these features. In fact, it has been intentionally designed to NOT tell you how to fix these features since there may be many ways to fix them. Some users find this frustrating since they hoped TOP Modeler™ would help them fix their problems. We found that the most successful users were those who, after deciding on a list of prioritized features (helped in part by your interactions with TOP Modeler™), decided to turn to either their workforce or selected people with knowledge and foresight and brainstorm suggestions about how to improve these individual features. So, for example, two companies which both found from their analysis that they needed to provide workers with more decisionmaking authority over customer relationships implemented changes very differently: in one company, they began customer site visits and in another company they gave workers the authority to contact customers after product delivery to determine customer satisfaction.

5) TOP Modeler™ does not eliminate what you might call "infeasible" solutions from its analysis. This means that TOP Modeler™ will inform you, for example, that your reward system needs improvement, even though you may have decided that you want to not include changes to the reward system in your organizational redesign efforts. TOP Modeler™ was intentionally designed to expose you to ALL the features that need improvement so that you become fully aware of the costs of not considering changes to certain features. Some users find this frustrating. We found that the most successful users treated this capability of TOP-Modeler™ seriously by reassessing the features they had previously decided not to be open for change. In so doing, their eventual change process became much more systemic. So, for example, a manager who originally had said that performance rewards would not change, realized that he had some rewards he could allocate under the right situations: time off, bonuses, and recognition for team-based performance.

6) TOP Modeler™ can not do all the translation into a specific site. This means that even though each definition was the results of thousands of consensus-building hours among U.S. manufacturing industry and academia, there is still some abstraction to the terms. This abstraction is intentional to allow for specific interpretations in each individual site. We have provided you with an "annotation" capability which allows you to add additional specification for your site. As long as the additional specification does not violate the initial definition, we highly encourage it since it contributes to a better understanding of the terms. But it is important to not violate the initial definition. For example, we have found that some users’ definition of quality is quality through rework, not quality first-time through. The definition of quality in TOP Modeler™ explicitly states quality as first-time through since that is the goal to be achieved. Additional specifications such as what first-time through means in a particular plant might be useful and do not violate the initial definition.

7) We are often asked the extent to which the TOP Modeler™ knowledgebase is specific to just batch-oriented manufacturing (because of such terms as "products") or can be applied to other manufacturing sectors (such as flow- or process-based manufacturing) as well as non-manufacturing operations (e.g., new product development, software engineering, health care). Since the Feature Sets and relationships among features are derived from basic research that is not specific to batch manufacturing, we believe in the generalizability to other sectors. We have used students to test out some of these claims and found, that the concepts hold in other sectors although much translation might be needed. So, for example, in applying the knowledgebase to process-based manufacturing, the concept of minimizing throughput time at individual workstations is best viewed as minimizing throughput time for specific segments of the process. If you have any questions about specific translations, do not hesitate to email us.

8) TOP Modeler™ does not provide cost information. As a result, features are prioritized not based on cost to fix but rather based on the number of business strategies adversely affected. TOP Modeler™ was intentionally designed this way because our experience indicates that cost of fixing and not fixing a problem are often difficult to assess accurately and quickly. The whole point of TOP Modeler™ is SPEED – i.e., quickly try out new and different organizational designs and see how they might help or hurt you. We believe that submitting this brainstorming and "what-ifing" to cost analysis will slow the process and unnecessarily yield a more conservative, less participative, and less innovative redesign. If cost analysis is necessary, we suggest waiting until after TOP Modeler™ has been used, generate several alternative plans, then begin the process of arguing over the numbers.