Why 70% of change initiatives fail (or do they?)

In the world of work and people development, commonly held views, models and tools are not always supported by the evidence. This can be a problem if we act on such information in good faith.

​In this article we provide, what we think is, a really good example of the kind of information that gets very authoritatively established and which in fact is incorrect.

As our name Managing Change indicates, our business is in helping people, teams and organisations manage change. In that time we have often heard it said that 70% of change initiatives in organisations fail. In fact if you Google that phrase there are 115,000,000 entries on it. It is repeated by very credible authorities, authors and researchers (including Harvard Business Review and Gallup). So it must be true, right?

In 2011 a study by researcher Mark Hughes looked into this. Following a thorough review of all published material, he found no empirical evidence to support this at all. In fact he didn't even find evidence that half of them fail.

So, where does 70% come from?

In 1993, a book called Re-engineering the Corporation (written by Michael Hammer and James Champy) included this sentence "our unscientific estimate is that as many as 50% to 70% of the organisations that undergo a re-engineering effort do not achieve the dramatic effects they intended". When this began to be misreported the co-author (Hammer) sought to emphasise that their earlier observation had been misrepresented. Despite those efforts, that statistic has been widely repeated in prestigious journals, books and articles ever since. Not only is it misreported, it doesn't even represent what they said; saying that 70% don't achieve the dramatic efforts intended is not the same as saying they failed!

To date there don't appear to be any authoritative or consistent studies that examine the success or failure rate of change programmes. A 2009 McKinsey study suggests that only around 10% of change programmes are considered to be complete failures. Given that change programme outcomes is unlikely ever to be a binary matter of success or failure, the McKinsey work suggests that the remainder probably fall around 30-40% completely or largely successful, 30% to be somewhat successful and the rest more unsuccessful than successful.

This illustrates the need to apply critical thinking to the information we are being given. If the issue seems important and we are going to take some form of action in relation to it, we should always first check where those numbers come from!