Of the three causes to budget changes (errors in estimations, new information obtained, and new mandates), I believe firstly that identifying them early on and addressing these as risks to be mitigated is important. For example, errors in estimation are simply the result of not properly judging the cost benefit analysis of research v. resources that created those estimates. Thomas Sowell argues for a balance between perfect knowledge and enough knowledge to make an informed decision in his book Knowledge and Decisions--an economic classic!
To address the risk of new information it’s important to have processes to gather the right amount of information early in planning but also as the project moves along. Good news doesn’t get better with time. Good PM needs to involve information updates regarding items that may change the project’s scope, risk, or budget. It’s a good practice to leverage technology to assist (google alerts, twitter etc.) and generally employ folks whose hobbies closely match their professional work. These hobbyist employees will be watching for new information because it is their passion. The trick is creating an environment where they’re willing to share what they learn.
With regards to new mandates one of the keys to surviving this is to move as much of an operation to an environment where mandates becomes unlikely. This is one reason why jobs move overseas. It’s not simply the cost of labor, it’s the cost of uncertainty for doing business. Dodd-Frank’s uncertainty to the financial market is huge and a key component of it (the CFPB) was declared unconstitutional. The entire financial system hinges on being regulated, and the current system for regulation just got labeled as defunct. How’s that for uncertainty?
If trying to prevent the changes prior to project execution is impossible then I would simply work through the same problem solving steps I’ve used for most of the things I’ve managed. Identify the problem, define the problem, identify potential solutions, evaluate potential solutions, and execute a final solution. In this process one trick is to select a solution that has maneuver room as more information evolves during the length of the project.
These changes aren’t exclusive to one another. Mandates can require new information for example when BPAs got labeled as bad Nalgene had to retool their formula and research new material.
In the end my plan is to minimize the risk of change early on or deal with it using a proven formula.