#NoEstimates advocates are on a Snipe hunt that will allow them to spend their customers money without having to tell that customer - with some level of confidence - how much money will be needed and how much will likely be spent, how long it will take to spend that money, and what will likely result from the passage of time and spending of that money in the end.
In case you weren't in the Boy Scouts in the 1950's and 1960's in the Texas Panhandle, a snipe hunt - a fool's errand - is a type of practical joke that involves a group of people making fun of credulous newcomers by giving them an impossible or imaginary task. The novice Scouts were sent on a Snipe Hunt, out into the woods and along the Canadian River in search of a mythical creature, with a bag and a flashlight.
The search for a method of making decisions in the presence of uncertainty - reducible (event-based, Epistemic) and irreducible (statistical variances, Aleatory) in the underlying work processes) - produces the same outcome. An empty bag at sunrise.
Let's start with some principles of managing other people's money. This way we can have a principled based discussion rather than a personal anecdote based discussion since anecdotal evidence isn't actually evidence of anything.
Anecdotes are a sample of one drawn from an unknown population.
- All project work contains underlying statistical processes. The project activities have natural variances. These are irreducible and labeled as Aleatory. These uncertainties are always present and must be dealt with in any management of the work. The only way to address irreducible uncertainty is with margin. Schedule margin, cost margin, technical performance margin.
- As well there are probabilistic events that occur on all projects. These are called epistemic uncertainties. There is a probability that some event will occur that will disrupt the progress of the projector impact the performance of the products or services produced by the project. These outcomes are reducible in that there can be work on the project to address the undesired outcomes when they occur. Redundancy, buy down, Plan B's are typical of reducible uncertainty strategies.
We need to know something about both reducible and irreducible uncertainties before they turn into Issues. If we haven't taken action - margin and buy-down - before it appears, we're going to be late, over budget, and our gadget is not likely to work as needed.
To discover the impacts of our actions made now on outcomes in the future when making decisions now in the presence of uncertainty - both reducible and irreducible - we need to make estimates of the outcomes of those actions.
Without these estimates - of the outcomes in the future - for the decisions we need to make today, we're on a Snipe hunt for the expected beneficial results of spending our customer's money. It that simple and that hard.
