As previously described, plans are the forward-looking tactics by which resources and attention are expected to be consumed, and are the basis of providing credible commitments to others. The first step in building plans which will support customer needs is to clearly establish what the goals and priorities are for each effort. This section is intended to offer some suggestions on how to improve your planning and goal setting.

• We must grapple with systems, people, circumstances, and resources, and our own shortcomings. Somehow we must advance, we must refine, and we must improve. - Atul Gawande, in Better.

- Mark Twain

It is difficult, yet highly desirable, for a manager and an employee to collectively generate and agree on a challenging and meaningful set of customer-driven goals and candidate personal improvements during each appraisal period. If the collective goals are too prescriptive or ambitious, their pursuit will be frustrating. If goals are not clear, or sufficiently comprehensive, opportunities to demonstrate value and get attention will be left on the table.

Too often, goals are ambiguous. When this is the case, accomplishments will probably be equally ambiguous. To make things worse, prior feedback may have been provided along lines that were not actionable (for example, 'increase customer satisfaction', 'find a way', or 'improve productivity'), rather than about specific behaviors that could really lead to such outcomes (for example, 'strive to improve the first-time quality of your written communications so that it meets the organizational checklist'). Managers and employees can't alter behaviors from routinely utilizing abstract slogans as goals, to establishing clear and actionable stretch targets, overnight. However, such a transition is an important behavioral change to continue to strive for, if real coaching, meaningful feedback, and rewarding results are to be expected from performance appraisals.

This is why it is important to refine your own personal and work goals into SMART goals, since unless the attributes of such SMART goals are specified, understood, accepted, and committed to, there will continue to be wide range of interpretations around what work actually needs to be done, and whether a particular goal's target expectations are achieved or not.

Individuals can only effectively manage between 5 and 9 tasks at any one time. Therefore, once a list of goals is identified, they must be planned so that there is an orderly and efficient approach to how the goals will be pursued. But one of the main challenges of this planning is dealing with the fog. Steve McConnell describes a cone of uncertainty as a recognizable pattern in many software projects (his company, Construx, provides the above graphical representation of this situation, and offers training approaches to reduce this uncertainty, on their Web site). Their depiction describes the reality that effort estimates (and therefore time estimates) have large variability early on in the project. When considering the underlying reasons for this from Steve's writings, it becomes clear that this is true for more than just software projects:

Research has found that the accuracy of the software estimate depends on the level of refinement of the software's definition. The more refined the definition, the more accurate the estimate. The reason the estimate contains variability is that the software project itself contains variability. The only way to reduce the variability in the estimate is to reduce the variability in the project itself.

It is everyone's job to work to reduce this variability, by striving for clarity in their own goals, and ensuring that they proactively manage the flow of required information they need to perform their tasks productively, and maintain a good connection with their customers about what results are expected, what roadblocks exist, and when goals are likely to be achieved. None of this is easy.

People will generally respond to challenging goals if they:

• believe the goals are worthwhile; this means that we they believe in the importance of the goal itself, and can draw power from their ability to contribute to it
• believe the goals are achievable; this means they have considered how they will need to approach the goal to be successful
• have confidence that mistakes (while undesirable) will be embraced as opportunities for learning, rather than punishment; this is usually all about how much trust they have in the environment in which they are working

Goals should also be made personal; goals that just say 'support xyz' do not satisfy this requirement, since they are neither measurable nor meaningful in terms of what specifically is expected from the individual in providing the support (knowledge? encouragement? doing the work?). If you are stuck for stretch goals, consider these suggestions on ways to generate ideas which could contribute to enhancing productivity. It is also worth considering the impact of these time-wasters in working to improve your own work efficiency.

Organizational goals are also important, but until they are translated into something specific that each person will do, they are often not very useful for motivating performance. Well-written goals can themselves help to motivate action, both for the individuals and the organization:

Schedule targets and priorities also play an important function in goal-setting. Making commitments and being accountable to such plans are a necessary part of work life. The above diagram reinforces the need to carefully consider the relative priorities of activities, utilize iterative development techniques, and come up with a way that provides flexibility for when plans change, so that one is not always constantly re-planning.

There are two effective strategies to support dynamic project tasks and priorities, and enable improvements to accommodating emerging work:

• First, all goals should be considered at two possible levels of accomplishment - a 'minimum acceptable' level, and an 'awesome' level. The key concept here is essentially to separate 'essential' levels of performance from 'desirable' levels of performance. This is hard, because some customers may believe that everything is necessary, and want it yesterday. But when realities are factored in, this position is usually no longer defensible, and priorities can be established. It is better to determine these levels in advance, rather than when time pressures arise. This allows these priorities to be factored into your planning.

• The second strategy is to develop plans which deliver capabilities incrementally. Plans should be traceable to the goals they are intended to support. You identify the increments by answering this question: what is the best achievable investment that can quickly be agreed to, and that will return the most value to the project and customer as soon as its produced, using available resources? Typically, what this means is to pursue attainment of all 'minimum level' achievements, from your above list, before pursuing any 'awesome' achievements. This takes discipline, and usually will only be achieved if you are accountable to someone else to make sure this is done.

A wise reader will recognize that such successive incremental steps may not get you where you need to go to meet the schedule targets that have been established, due to constraints that arise, insufficient resources, or other limitations. When constraints are the roadblock, it's always helpful to have more than one goal available to work on at any one time. Then, when a roadblock is encountered on one task, and if the individual is unable to remove the roadblock themselves, they will already be enabled by designated priorities to immediately select other work (though you should obviously notify your work cell leader and manager when this has occurred, and when their help is needed in removing a roadblock).

The better an operational context that can be provided, such as the above, the more likely that people will be able to make appropriate decisions themselves, rather than needing a lot of coordination and communications with others. Multi-tasking is one of the most cited reasons for inefficiency during a development effort, and planning can reduce the impact of these context changes. By structuring goals and planning using the above guidelines, the effects of this multitasking can be minimized, while enabling increased utilization of people's capabilities and capacity.

Once plans are in place and baselined, techniques such as painless schedules (from Joel on Software) can be used to track and report progress. This approach is an excellent, excel-based approach to tracking estimates and performance on tasks. Joel's team, Foghat software, has actually built some very advanced capabilities that support forecasting the future, using Monte Carlo simulation techniques, that are appropriate for medium and large teams. But when deliverables are primarily produced by single individuals, Excel is a very acceptable alternative. By using the spreadsheet, and tracking the time spent on each task, you can get a handle on how your time is being spent on each task, how effective you are in estimating your work, and how your planning approaches may need to change with time.

To demonstrate these ideas in action, I implemented functionality to define goals, to plan and track tasks, and to track time spent on this web site, and used them to produce the content in this series of articles. This information has already been helpful in my own planning for new content, and should be useful to others as a concrete example of how to perform such planning, tracking, and goal setting.