[nswanberg]

Gerald Weinberg discusses this problem in his "Secrets of Consulting", somewhere after proclaiming "it's always a people problem." One is hired as a consultant, and so one thinks the best approach is to recommend a change that likely achieves the best possible outcome for the client.

But in accepting this, the client would have to acknowledge the existence of this problem and all its potential consequences.

If I remember correctly, his recommendation is for a consultant to only improve matters by, say, 10%, making the client look good for hiring a consultant, making the consultant look good for delivering, and, what seemed to me just a happy side-effect, benefiting the client's organization.

The alternative, I suppose, is for the consultant to move on to greener pastures, and for the organization to see a 0% gain.