(FORTUNE Magazine) – When consultants show up to root around inside your company, several things can happen. They may perform dazzlingly, saving you millions of dollars and pointing you in the direction of lucrative new markets. They could also run amuck, costing you tons of money and pushing your company to the brink.

In Dangerous Company: The Consulting Powerhouses and the Businesses They Save and Ruin, James O'Shea and Charles Madigan, both Chicago Tribune veterans, deliver some excellent inside stories about these sorts of extremes. One chapter describes how Guinness installed its Bain & Co. consultant as finance director. After paying off people to prop up the stock, the Bainie turned state's evidence, testified against his client, and sent the CEO to jail. Amazingly, the fellow continued to find work as a consultant when the trial was over.

In sharpest contrast, another chapter describes how Sears CEO Arthur Martinez employed A.T. Kearney consultants to help him engineer one of the most stunning corporate turnarounds in recent memory. Part of his strategy, we learn, was to aim the outsiders at specific targets. The Kearney crew was put to work finding savings with Sears' suppliers, for example, while Martinez used his own team to drive growth. "If the people at the top of the organization can't create a clear line of sight to where they want to go, then they are a little bankrupt personally," he says.

The problem here, however, is with the book's treatment of a third scenario of consultant employment: reality. The vast majority of companies who use consultants don't end up anything like Guinness or Sears. Few consultants are bold enough to install their own employees as corporate officers. If clients let them, they deserve whatever they get. And it's the rare executive, faced with a mess like the one that existed at Sears a few years ago, who has the sophisticated turnaround skills of an Arthur Martinez. Most people need more help than he did, and for them, these stories just don't add up to much.

So how does this third kind of consulting engagement--the one that most companies encounter--tend to unfold? Clients get some good recommendations, pay more for them than they thought they would, ignore some parts, and adopt others. In the end, it's often hard to gauge whether the modest results are due to the consultants' efforts or to unrelated market forces.

That's reality. Unfortunately, it doesn't make for very compelling reading. Take the chapter on Gemini Consulting and two of its "transformation" clients, which is basically unreadable. It meanders from client histories to Gemini's troubles and back again. Surprisingly, we learn almost nothing at all about how the clients' businesses actually changed and what the results were. "It is clear," the authors write, that one client "is on the way to becoming what it thinks it wants to become." Huh?

Much of the book, in fact, falls victim to this sort of imprecision. While it should be obvious that most readers are deeply curious about the companies that consultants "save and ruin," we learn more than we ever cared to know about the inner workings of the various "consulting powerhouses." Partners blather on about their philosophies of doing business, but readers care more about their advice, which often ends up sounding pretty much the same. In the McKinsey chapter in particular, we hear almost nothing about clients but plenty about how warmly McKinseyites regard their own talents. Consultants sometimes draw better than they talk; maybe this tale could have used a few diagrams.

The book does have some bright spots. The authors clearly have spent a lot of time in courtroom basements, burrowing through suits filed against consultants (who always settle long before any trial ensues). In the course of looking into one case against Andersen Consulting, they've turned up a document proving definitively that consultants sometimes get paid according to how many jobs they manage to eliminate for their clients.

The most useful information in Dangerous Company comes in the last two pages, a checklist of issues to consider when bringing consultants aboard. Still, simply selecting the right firm isn't the hardest part of any of this. "The bad thing about consultants is, they think they have all of the answers," Martinez says. "The good thing is, sometimes they have some of the answers." Too bad no one's written a book about how to pick out the right ones.