Everyone Bought The Tools, Almost Nobody Changed The Work

McKinsey finds workflow redesign to be among the strongest predictors of value, and most organisations have bolted AI onto processes they never revisited.

Sign up for my Substack daily AI newsletter here.

See my AI Training course portfolio for corporate Business Leaders here.

Article link: https://johanosteyn.substack.com/p/everyone-bought-the-tools-almost?r=73gqa&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true

Follow me on LinkedIn: https://www.linkedin.com/in/johanosteyn/

There is a comfortable way to read the adoption statistics and an uncomfortable one. The comfortable reading is that almost everyone is now using artificial intelligence, so your organisation is keeping pace. The uncomfortable reading is that almost everyone is using it and almost nobody can show what it earned them.

CONTEXT AND BACKGROUND

McKinsey’s latest global survey drew responses from 1,993 participants across 105 nations between late June and July 2025, weighted by each country’s contribution to global GDP. Eighty-eight percent report regular AI use in at least one business function, up from 78 percent a year earlier, while approximately one-third say their organisations have begun scaling. Thirty-nine percent attribute any level of EBIT impact to AI, and most of those put it below 5 percent. Six percent qualify as high performers, meaning they attribute 5 percent or more of EBIT to AI and report significant value. Fifty-one percent have seen at least one negative consequence from AI use, with nearly a third reporting consequences from inaccuracy. McKinsey’s own summary is that meaningful enterprise-wide bottom-line impact continues to be rare.

A separate study reaches the same place by another route. PwC surveyed 1,217 senior executives across 25 sectors and six regions, finding that 20 percent of companies capture 74 percent of AI-driven returns, with the most capable delivering financial performance 7.2 times as high as everyone else.

INSIGHT AND ANALYSIS

The interesting part is what separates the leaders, because it is not the technology. They buy from the same suppliers as everyone else. McKinsey tested 31 variables and found that fundamentally redesigning workflows made one of the strongest contributions to meaningful business impact, with high performers nearly three times as likely to have done it, three times as likely to strongly agree that senior leaders demonstrate ownership, and more than a third committing over 20 percent of digital budgets to AI. PwC describes the same behaviour more concretely. Its leaders do not add AI on top of workflows, they integrate it into standard operating processes, redesigning customer support so the system runs inside the case management tool rather than bolting on a separate chatbot that agents must consult and copy from manually. They are also 80 percent more likely to systematically track business impact. Note what that list describes. Every item is a decision taken by people in a boardroom rather than a capability delivered by a vendor. The common failure is not choosing badly, it is layering a new tool onto a process designed around its absence and expecting the process to yield something it was never built to produce.

IMPLICATIONS

South African organisations sit inside this distribution with less room for error than most. Constrained capital and pressure on margins mean an underperforming AI programme cannot be quietly carried for three years while everyone waits for maturity. Local enterprises also face fluctuating operational costs on hyperscaler infrastructure because of rand volatility against the dollar, so the spend side is harder here even where the value side matches global norms. Three consequences follow. Adoption is no longer a reportable achievement, since 88 percent of your peers can claim it and it distinguishes nothing. The reporting question for a board is not how many functions use AI but which workflow was redesigned around it, by whom, and what changed as a result. PwC’s practical suggestion is worth adopting directly, being a monthly review in which only projects showing measured movement on a defined business metric receive further funding.

CLOSING TAKEAWAY

Three years into this, the honest summary is that the tools arrived and the work stayed the same. That is not a criticism of the technology, which does what it does. It is an observation about organisations, which find buying easier than changing. Two independent studies covering more than three thousand companies between them point to the same conclusion, and it is not a flattering one for the majority. For a South African board, the useful question at the next meeting is not what AI can do. It is which process in this business was rebuilt around it, and if the answer is none, the returns discussion is premature.

Johan Steyn is a prominent AI thought leader, speaker, and author with a deep understanding of artificial intelligence’s impact on business and society. He is passionate about ethical AI development and its role in shaping a better future. Find out more about Johan’s work at https://www.aiforbusiness.net