Over the past decade we have seen Indian companies undergoing significant transformation programmes. Infosys [Get Quote], a company founded in 1981, has transformed itself to become a world leader in IT services. ICICI Bank [Get Quote] has transformed from being a leading Indian project finance company to a rapidly globalising financial institution.

While these are illustrations of some of India's successful transformations, what is striking is not the size of the prize for those who get it right, but the ignominious fate of those that don't (or don't even try). For instance, between fiscal years 2002 and 2004, 31 companies dropped off the ET list of the top 100 companies in India.

Why do some chief executives succeed in dramatically improving the performance of their companies, whereas others flounder in a mire of failed initiatives?

At McKinsey, we've supported hundreds of major transformation efforts over the years, whether born of sudden crises, the gradual loss of previously dominant market positions, or the threat of intensifying global competition.

Over the past couple of years we've interviewed more than a thousand senior managers -- as well as 35 professors from leading business schools -- in an effort to distil what distinguishes a successful change effort.

The exercise not only reinforced our conviction that a leader's ability to create, sustain and channel organisational energy is the crucial catalyst, it identified six key issues typically addressed by those companies that have made a significant performance leap. These are:

1. Programme design

While it's always good to release pent-up entrepreneurial ability and desire, programmes should be carefully organised around discrete, manageable chunks visibly unified by a common logic and purpose.

At the broadest level companies need to articulate a transformation "story" -- a big-picture ambition that captures the imagination internally. Next, they should roll it out in chapters, or in what Julio Linares, executive chairman of Telefonica de Espana, calls "waves" -- blocks of work that between them deliver performance and transformation goals over a given period.

Examples might be boosting geographical reach, achieving operational excellence or building a high quality service business. Last come the individual initiatives that turn a change agenda into reality on the front line.

In our experience, poorly designed programmes only generate incremental performance improvements whose effect is likely to be temporary; so, it can be worth delaying action for a few weeks to ensure the overall architecture is clear and robust.

2. Performance and 'health'

We often hear managers complain that the pressure to achieve quarterly results crowds out attention to longer-term goals -- especially in a turnaround. But an effective change programme has to plant seeds for the future as well as harvest more aggressively the fruit that's already there.

In general we find that companies assess performance gaps more readily than they diagnose problems with their "health" (defined as the ability to perform well year after year after year).

That's why top teams and boards need to monitor variables such as brand strength, employee retention rates, customer satisfaction and external relationships (with, say, the press and regulators), as well as the financial numbers.

Barclays admits to having neglected what it calls its "franchise" health in the past. Inspired by the fresh perspective of CEO John Varley, the UK bank now monitors its personal-retail banking business by tracking a number of non-financial metrics, including customer attrition, staff resignations and various corporate social responsibility indices.

3. Aspirations and pace

Efforts to transform companies are often undermined by failure to translate a long-term vision into a series of "mid term" objectives that really stretch those charged with effecting change. Easily attainable goals spark little energy and falling short of them creates no sense of the danger that inspires urgent action.

Alan Lafley, chairman and chief executive of P&G, is especially conscious of the difference between incremental business building and true transformation, having led the consumer products business to new levels of performance in recent years. "You can get used to being a player without being a winner," he observes, a state of mind that all too easily erodes into complacency.

But Lafley warns that it is counterproductive to overpromise, and in turning around the group he has tended to set internal goals higher than those announced externally.

Even at a fast pace it may take five years or more to realise the original vision -- but we have seen some companies, not least a once notoriously-inefficient European postal system, make progress in half that time. Ambitious targets played a major part in that turnaround.

4. Making change stick

Successful transformations are typically rooted in operational change that quickly takes physical shape -- essential if customers and employees are to be convinced that a company will not revert to its former (bad) ways.

Joe Tucci, responsible for the recent dramatic turnaround at EMC, first sought to establish an "unshakable fact base" that would challenge those who doggedly held to the belief that the information storage business was recession proof -- and thereby resisted change.

But he also set out to capitalise on the operational impact of moving to attack the "mid-tier" storage market -- a strategic reorientation that signalled a clear break with the past.

Other companies encourage "lean" thinking, which began as an automotive-inspired and assembly-oriented methodology but which today is as established in many financial processes, service industries and parts of the public sector. "Lean" projects depend not just on rolling out new techniques but on changing basic ways of working.

5. Employee behaviour

Ultimately, change sticks when it captures individual hearts and minds. In our experience there are four key elements here: people must know what they need to change and really want to do it; they need systems and structures consistent with these desired new behaviours; they must have the skills and competencies required to behave differently; and they require role models to guide them.

N R Narayana Murthy, chairman and chief mentor of the Indian consulting and information services company Infosys, has acknowledged the importance of role modelling at all levels of his company. As he observes, "It is best to remember Mahatma Gandhi who said 'Be the change you want to see in the world'".

6. Leadership capacity

Many change programmes fail because the number of challenges and opportunities exceed the ability of the current crop of leaders to exploit them. While strong individuals are obviously required to drive a transformation, better leaders (and more of them) tend to emerge in the wake of successful change. The two are mutually reinforcing. Poorly performing top teams, moreover, breed competing agendas and turf politics.

"People in the company must understand that you are a group that works well together," observes Corrado Passera, chief executive of Italy's Banca Intesa. "The way to fail in a transformation is to have managers at the top who are fundamentally reluctant to push through change."

To conclude, each transformation is different and we would certainly not advance any magic formula for effecting change. But our research and experience has taught us that all six issues discussed above need to be addressed at some point in a transformation, and that getting them right helps release and channel the leader's energy.

Gautam Kumra is a partner based out of McKinsey & Company's New Delhi office.