Many digital consultants open with the same slide. “Seventy percent of digital transformations fail,” usually with a McKinsey or BCG logo parked next to it. The number is everywhere, and it travels well because it’s short, scary, and round.
Here’s what holds up when you trace it. We found no credible study that establishes 70 percent as a measured digital transformation failure rate under a consistent definition. The figure is a paraphrase of a strict success bar, a fusion of two different consulting numbers, and an echo of an older change-management claim that a peer-reviewed review found baseless.
This page rates the claim and the sources behind it the way we rate every number: High, Medium, or Low, based on whether you could defend it to someone who knows the research.
What the claim says, and where it comes from
The claim travels in a few different costumes, and they don’t say the same thing.
Start with the misquote, because it’s the most common. People cite McKinsey for “70 percent of digital transformations fail.” McKinsey does use a 70 percent figure, but for organizational change broadly. Its 2015 article Changing change management states that 70 percent of change programs fail to achieve their goals, with no source note, method, or sample shown behind it.
A 2020 McKinsey article repeats that 70 percent of transformations, “both tech-enabled and conventional,” fall short of their goals. That one carries a footnote, and the footnote points to a 2017 Forbes column and back to the unsourced 2015 article. Its 2021 report Losing from day one leans on John Kotter for the same number. So the McKinsey 70 percent is real and repeated, sourced to a Forbes column, its own earlier writing, and Kotter, rather than to a study that measured it.
McKinsey’s actual digital-transformation measurement is a different number. Its 2018 Global Survey on digital transformations found that only 16 percent of respondents said their digital transformations both improved performance and sustained the improvement, with another 7 percent improving but not sustaining. The “fewer than 30 percent succeed” line that often travels with it describes McKinsey’s broader transformation research, not the digital survey specifically. Either way, McKinsey did not measure a 70 percent digital failure rate.
BCG did publish a 70 tied to digital. Its 2020 report, Flipping the Odds of Digital Transformation Success, says 70 percent of digital transformations fall short of their objectives. Read the breakdown and it softens: 30 percent fully succeeded, 44 percent created some value but missed their targets, and 26 percent created little or nothing. So 44 of those 70 points delivered something. The scores came from executive self-assessment plus BCG’s own work with 70 client companies, on a composite scale rather than from audited outcomes.
A widely cited version appears in a 2019 Harvard Business Review article by Behnam Tabrizi and co-authors. It opens by stating that 70 percent of digital transformation initiatives don’t reach their goals, and that of the 1.3 trillion dollars spent in a year, an estimated 900 billion was wasted.
Those figures trace to a 2018 Forbes Technology Council column by Steven ZoBell, then chief technology officer at the work-management software firm Workfront. ZoBell took the year’s projected 1.3 trillion dollars in digital transformation spending, applied the 70 percent, and arrived at the 900 billion. No primary study sits behind the 70 percent in either piece.
Trace the round number back far enough and it predates digital entirely. In 1993, Michael Hammer and James Champy wrote in Reengineering the Corporation that somewhere between 50 and 70 percent of reengineering efforts didn’t achieve the results intended. They called it an “unscientific estimate.”
By 1995 Hammer was walking it back, writing that there’s no inherent failure rate for reengineering. The number kept going anyway. Mark Hughes later traced the broader “70 percent of change fails” narrative across five separate published sources and found no empirical evidence behind any of them. The digital claim echoes that older narrative, but a direct lineage hasn’t been demonstrated. Hughes debunked the broad narrative; he didn’t prove the digital figure descended from it.
The Reliable Confidence Score
| Claim or Source | Figure | Reliable Confidence | What It Really Means |
|---|---|---|---|
| “McKinsey says 70% of digital transformations fail” | 70% fail | Lowtwo figures conflated | Mixes two different McKinsey claims. McKinsey does use “70% of transformations fall short” in change-management work, footnoted to a Forbes column and its own unsourced 2015 article. Its 2018 digital survey is separate and measured success: 16% improved and sustained. Neither is a measured 70% digital failure rate. |
| McKinsey Global Survey, digital transformations (2018) | 16% improved and sustained (digital); under 30% succeed (overall) | Mediumself-reported | Survey of 1,793 respondents, 1,521 of whom had been part of a digital transformation, GDP-weighted. Perception against a strict success bar, not audited outcomes. |
| BCG, Flipping the Odds of Digital Transformation Success (2020) | 70% fall short of objectives | Medium“fall short,” not “fail” | Real published figure from 825 executive self-assessments plus 70 BCG client engagements, on a composite scale. But 44 of those 70 points created some value. “Fall short” got rounded up to “fail.” |
| HBR, Tabrizi et al., Digital Transformation Is Not About Technology (2019) | 70% miss goals; $900B of $1.3T wasted | Lowtraces to a vendor column | The most-cited home of the digital 70%. The figures trace to a 2018 Forbes column by a software-firm CTO, who applied the 70% to the year’s $1.3T spend forecast to get the $900B. No primary study sits behind the rate. |
| Hammer & Champy, Reengineering the Corporation (1993) | 50% to 70% don’t achieve results | Lowauthors’ own caveat | An early instance of the broad change-failure number, not a proven source of the digital claim. The authors called it an “unscientific estimate,” and it described reengineering, not digital. Hammer rejected the fixed-rate reading in 1995. |
| Hughes, Journal of Change Management (2011) | No empirical basis for any 70% rate | Highpeer-reviewed | Reviewed five published sources of the 70% claim, including a McKinsey article and Kotter. Across all five, Hughes found no valid and reliable empirical evidence for the rate. Returned to the finding in 2022. |
The Big Takeaway
The honest answer to “do 70 percent of digital transformations fail” is that nobody has measured it cleanly. What we have is one strict success bar (McKinsey), one “fall short” band that includes partial wins (BCG), and one opening-line assertion with no study named (HBR), all echoing a 1993 guess about a different practice.
That doesn’t make digital transformations easy. McKinsey and BCG both land well under half on full success, and that’s worth taking seriously. The problem is the specific, scary, round number that gets quoted as if someone counted. It’s folklore wearing a lab coat.
No one has cleanly measured a 70 percent digital transformation failure rate. The figure fuses McKinsey’s broad change-failure number with the authority of its separate digital survey, and echoes a 1993 estimate the authors called unscientific.
Why the numbers don’t agree
Three things keep these figures from lining up.
First, definitions. McKinsey counts a transformation as a success only if it both improved performance and made that improvement stick. BCG scored projects on a composite 1-to-10 scale across targets met and value created. “Fail” means something different in each, and most people quoting the 70 percent never check which definition they’re importing.
Second, measurement. The McKinsey and BCG figures both lean on executives rating their own transformations, which is perception data. BCG adds its own client engagements to the mix. Useful for direction, shaky as a precise rate.
Third, the denominator. A 70 percent that includes “created some value but missed the target” is a very different statistic from a 70 percent of total write-offs. BCG’s own breakdown shows most of its 70 produced something.
How to use this number safely
For maintenance and reliability teams, this stat shows up in two pitches, and both are worth resisting.
The first is the scare pitch. A vendor cites “70 percent fail” to frame the consulting, the platform, or the managed service as the thing that beats the odds. Ask what definition of failure they’re using and where it was measured. The figure rarely survives the question.
The second is the excuse. A stalled CMMS rollout or a condition monitoring program that never got past the pilot gets written off with “well, most of these fail anyway.” That lets a project that failed for specific, fixable reasons (no data discipline, no clear owner, no tie to the work order flow) off the hook.
Skip the industry rate and measure your own project: adoption, data quality, and whether the thing changed a decision someone makes every week. That tells you more than any round number.
Where teams go wrong
The common mistake is treating 70 percent as a base rate and planning around it. The figure has a long and shaky pedigree, and none of the sources reviewed establishes a universal 70 percent digital failure rate under a consistent definition.
The second mistake is citing McKinsey for it. McKinsey’s digital-survey number is a success rate, and its 70 percent figure lives in separate change-management writing that it footnotes to a Forbes column, its own 2015 article, and Kotter. Quote McKinsey’s digital success figure accurately, or quote BCG’s “fall short” wording, but don’t merge them into a clean 70 percent digital fail with a logo attached.
Methodology
We rated each row on whether the figure traces to a named primary source with a stated method and definition, and whether it survives reading that source directly.
We read McKinsey’s 2018 digital survey, its 2015 Changing change management article, its 2020 tech-enabled-transformations article, and its 2021 Losing from day one report, plus BCG’s 2020 Flipping the Odds report and the 2019 HBR article, checking the definitions, sample sizes, and what each one cites. McKinsey’s cited versions point variously to a Forbes column, its 2015 article, and John Kotter.
The HBR article’s 70 percent and $900 billion trace to a 2018 Forbes column by a software executive, not to a study. For the origin, we relied on Mark Hughes’s 2011 peer-reviewed review in the Journal of Change Management, which traced the 70 percent across five published sources and found no empirical basis, a finding he returned to in 2022. Self-reported perception data caps at Medium. A real published figure that gets distorted in the retelling earns the original a Medium and the distortion a Low.
Common questions
Do 70% of digital transformations really fail?
There’s no clean study showing a 70 percent failure rate for digital transformations under a consistent definition. McKinsey’s 2018 Global Survey found only 16 percent both improved performance and sustained it. BCG’s 2020 report found 70 percent fall short of objectives, though most of those still created some value.
Did McKinsey say 70% of digital transformations fail?
McKinsey uses a 70 percent failure figure in its change-management writing, including a 2015 article that states it with no source shown and a 2020 article that footnotes it to a Forbes column and that same 2015 piece. Its 2018 digital survey is separate and measured success: only 16 percent both improved and sustained performance. McKinsey did not report a measured 70 percent digital failure rate.
Where does the “70% of transformations fail” statistic come from?
The broad “change fails” version goes back at least to Hammer and Champy’s 1993 book Reengineering the Corporation, which called a 50-to-70 percent figure an “unscientific estimate.” A 2011 peer-reviewed review by Mark Hughes in the Journal of Change Management found no empirical evidence behind the 70 percent claim across five published sources. The digital version echoes that older number rather than resting on a separate study.
What did BCG actually find about digital transformation success?
BCG’s 2020 report Flipping the Odds of Digital Transformation Success found 30 percent of transformations fully succeeded, 44 percent created some value but missed targets, and 26 percent created little or nothing. The “70 percent fall short” headline combines the second and third groups.
Is the $900 billion wasted on digital transformation figure accurate?
The $900 billion of $1.3 trillion figure comes from a 2019 Harvard Business Review article, which drew it from a 2018 Forbes column by Steven ZoBell, then CTO of the software firm Workfront. ZoBell applied the 70 percent failure figure to the year’s projected $1.3 trillion in digital transformation spending to reach $900 billion. No primary study sits behind the 70 percent.
Should maintenance teams worry their CMMS or predictive maintenance rollout will fail?
The 70 percent figure is too soft to plan around. A better read on an industrial digital project is your own data: user adoption, data quality, and whether the system changed a decision your team makes regularly. McKinsey and BCG both tie success to leadership and capability, not the technology alone.
The Short Version
Nobody has cleanly measured a 70 percent digital transformation failure rate. McKinsey found that only 16 percent of digital transformations both improved and sustained performance, against a strict bar. BCG found 70 percent fall short of objectives, most of them still creating some value. The “McKinsey says 70 percent fail” line conflates McKinsey’s broad change-management framing with its separate digital survey. The broad version of the round number goes back to a 1993 estimate the authors called unscientific, and a peer-reviewed review found no evidence for it. Don’t plan around the 70 percent. Measure your own project instead.
Sources
- McKinsey & Company, “Unlocking success in digital transformations” (Global Survey, October 2018). https://www.mckinsey.com/capabilities/people-and-organizational-performance/our-insights/unlocking-success-in-digital-transformations
- McKinsey & Company, “Changing change management” (2015). https://www.mckinsey.com/featured-insights/leadership/changing-change-management
- McKinsey & Company, “How transformation offices help tech-enabled transformations succeed” (February 2020); its 70% figure is footnoted to a 2017 Forbes column and the 2015 “Changing change management” article. https://www.mckinsey.com/industries/industrials/our-insights/how-transformation-offices-help-tech-enabled-transformations-succeed
- McKinsey & Company, “Losing from day one: Why even successful transformations fall short” (December 2021), which references Kotter for the 70% figure. https://www.mckinsey.com/capabilities/people-and-organizational-performance/our-insights/successful-transformations
- Brent Gleeson, “1 Reason Why Most Change Management Efforts Fail,” Forbes (July 25, 2017); the Forbes column McKinsey’s 2020 article cites for the 70%.
- Boston Consulting Group, Flipping the Odds of Digital Transformation Success (October 2020). https://www.bcg.com/publications/2020/increasing-odds-of-success-in-digital-transformation
- Behnam Tabrizi, Ed Lam, Kirk Girard, and Vernon Irvin, “Digital Transformation Is Not About Technology,” Harvard Business Review (March 13, 2019). https://hbr.org/2019/03/digital-transformation-is-not-about-technology
- Steven ZoBell, “Why Digital Transformations Fail: Closing the $900 Billion Hole in Enterprise Strategy,” Forbes (Forbes Technology Council, March 13, 2018); the source of the $1.3 trillion / $900 billion / 70% figures the HBR article repeats. [Confirm live URL before publishing: citation verified, exact forbes.com slug not yet confirmed.]
- Mark Hughes, “Do 70 per cent of all organizational change initiatives really fail?” Journal of Change Management, 11(4), 451-464 (2011). DOI: 10.1080/14697017.2011.630506. https://www.tandfonline.com/doi/abs/10.1080/14697017.2011.630506
- Mark Hughes, “Reflections: How Studying Organizational Change Lost Its Way,” Journal of Change Management, 22(1), 8-25 (2022). DOI: 10.1080/14697017.2022.2030980.
- Michael Hammer and James Champy, Reengineering the Corporation (HarperBusiness, 1993); Michael Hammer and Steven Stanton, The Reengineering Revolution (HarperBusiness, 1995), p. 14 (the “no inherent failure rate” line, as cited in Hughes 2011).
