Conclusion
Looking back over these chapters, you’ll notice that there are many suggestions but no recipes for success. No project is an average project. Each has its own unique challenges and circumstances. Learning how to recognize the variance between expectations and reality, and adjust for them, is a key part of success.
There are many suggestions and warnings here based on my experience and preferences. These are intended as a starting point. There is no destination called “perfect estimation.” Feel free to go beyond my suggestions. You are likely to discover things I haven’t yet. (And please let me know what you discover!) Feel free to do things that seem appropriate for your circumstances, even if I’ve warned against them. Just do them with eyes wide open, aware of the potential problems. You’ll likely find other means to keep those problems in check.
Start with examining what needs you are trying to fill. As in software development, if you don’t know what you’re trying to achieve, you’re unlikely to do so. Also, as in software development, realize that you’re unlikely to know all the needs at the start. Keep your eyes open for needs that you overlooked or that have arisen since the start.
Take a deep breath and look around. See the options you have. Note the alternatives to “what you’ve always done” and to “the way we do things around here.”
Start with understanding what you want. No, not the first thing that comes to mind. If you had what you wanted, what would it do for you?
The real need is never the estimate. Usually it’s some decision that needs to be made. Sometimes it’s reassurance that things are generally OK, or a warning if they are not.
Sometimes it’s not your need that needs to be addressed. Your need might be that someone else is satisfied. This complicates things as it requires understanding their need.
There may be multiple competing needs. These needs might not be satisfied with the same approach to estimation. They may have different needs for accuracy, for precision, for distribution of error, or other considerations. You may need to estimate in more than one way.
If you do need an estimate that is both precise and accurate, realize that this is a high art that requires skill and practice. And it requires accurate and appropriate historical data. Even then it’s always possible that something you didn’t, and possibly couldn’t, anticipate can make your estimate completely wrong.
Most of the time you need little precision and not even very much accuracy. Most of the time you need situational awareness so that you can prepare for the future and notice when conditions change. This requires attention more than correctness. It requires some smaller, nearer estimates to sense the problems in the larger, more distant ones.
The differences between your estimates and reality give you clues as to what assumptions you’ve made that aren’t holding true. Estimates are not the same thing as plans. Estimates inform plans, but as the situation changes, or your understanding of the situation changes, plans generally need to change as well.
The information power of incorrect estimates is so powerful that it’s beneficial to quit thinking of estimates as predictions and instead think of them as hypotheses. An estimate, properly framed, is an experiment. If the results don’t come out as expected, you’ve learned something. This feature of estimates may be powerful enough to convince you that you want more estimates, rather than fewer.
This reframing should also defuse a lot of the angst and blame surrounding estimation. Where it doesn’t, yet, realize that change is a process, not an event. As Esther Derby suggests in 7 Rules for Positive, Productive Change [Der19], ask yourself “why might reasonable intelligent people act this way?” Others may be viewing the situation differently from you. Where can you find common ground? Where can you have dialog to understand each other’s point of view?
Estimates with value still take thought and effort. I hope that this book will help guide your thought and reduce effort that doesn’t contribute to meeting your needs. I would like it if you would let me know how it has helped, and even where it has not helped. In the latter case, perhaps I can offer something I neglected to include.
For now, go forward and estimate with all the confidence you can. Use those estimates for your greatest benefit. Prosper to the best of your ability.