Eric Garland, a futurist/strategic analyst, published an indictment of strategic consulting last week in The Atlantic. Although Garland is focused on the sort of foresight services provided to governments and CEOs and boards of large companies, he also mentions general purpose consulting companies such as Booz Allen, Accenture and McKinsey. Garland’s experience has some lessons for CIOs in the market for IT consulting services.

Michael Krigsman cites research by IAG Consulting finding that 68% of IT projects will likely fail. It’s the most pessimistic of estimates that Krigsman has reviewed. IT failure is difficult to quantify because there are so many definitions of failure – does failure mean being late and over budget, or does failure mean the project was aborted entirely? But one thing is for certain: project failure is expensive.

Krigsman invited IT experts and DevOps Cookbook co-authors Gene Kim and Mike Orzen to calculate the annual cost of failed IT projects. The duo pegs the cost of IT failure at $3 trillion annually.

Why do IT project fail? There are a few reasons. The IAG study focuses on planning and requirements gathering, or as Krigsman puts bluntly: “your organization probably does a lousy job setting up projects, which is why they fail.”

But Krigsman has written about another root cause of project failure: denial. Krigsman describes denial as what happens when projects move forward despite unresolved disagreements between stakeholders. “Problems then simmer below the surface only to ‘unexpectedly’ erupt later, usually in more severe form.”

Hiring “yes men” instead of a service provider that will tell you things you might not want to hear is a good way to foster this sort of denial. When hiring a service provider, ask hard questions. Ask for something unrealistic and see how they respond. But most importantly, be open to hearing something you don’t want to hear and be ready to adjust your expectations accordingly.