Layoffs Are Being Rebranded as a Turnaround. KPMG Is a Different Story.

The Netherlands' unemployment rate rose for a third consecutive month this week, to 4%, according to Statistics Netherlands. In the same week, economists told the Financial Times they'd finally found genuine evidence of a UK productivity boost, the first real sign of one in years. Both of those numbers are accurate, published by people whose job is to be right about this, and neither one explains why Diageo, Starbucks, Sanofi and KPMG all cut jobs in the same seven days, three of them reaching for the same manageable word to describe it.

Employer branding works better when people share what they know. EBClub brings practitioners together to exchange ideas, compare experiences, and learn from the people doing the work.

Explore EBClub

This week's Paper Cut is about what "turnaround" is doing for employers that the actual layoff number can't, and what happens once people notice the gap between the word and the number.

This Week’s Paper Cut's

Watch or listen here ↯

We keep tracking to a minimum. The video is hosted on YouTube and the podcast is hosted on PodBean, which may place cookies that measure viewing and personalize content. We only load the player after you consent to these cookies. You can update your cookie settings any time.

"Turnaround" is the word. Redundancy is the sentence.

Diageo's headcount has fallen by 6.4% under new chief executive Dave Lewis, the Telegraph reported, with the cuts landing hardest in its North American business unit, according to a separate BevNET report, as the Guinness, Johnnie Walker and Baileys owner works through a profit slump. Starbucks confirmed in a WARN Act filing this week that it's cutting more than 200 corporate roles, 120 of them tied to Seattle-based staff who declined to relocate to the company's new $100 million Nashville office, as CEO Brian Niccol's two-year turnaround plan reaches its final stage. Sanofi filed its own notice with the Massachusetts Executive Office of Labor and Workforce Development confirming 229 job cuts at Blueprint Medicines, the Cambridge biotech it acquired for $9.1 billion last year; a spokesperson described the cuts as "thoughtful organisational decisions to align our structure with our long-term business priorities."

KPMG doesn't get to use that kind of language. The firm's Australian arm has begun cutting roughly 500 positions, including dozens of partners, after a scandal in which partners allegedly misused confidential client information to chase business from rivals of their own clients, a mess that's already cost it a major client in Lendlease and put a $70 million audit contract under review at Macquarie Group. There's no turnaround plan attached to that one, just a governance failure working its way through the org chart.

The irony arrived from an unrelated direction. Vishal Garg, who became one of the most recognisable executives in tech after firing more than 900 staff over a three-minute Zoom call in 2021, has been removed as chief executive of Better Home & Finance, PeopleMatters reported, in what the company's board described as a matter of "judgment, temperament and credibility." Garg disputes the circumstances and is fighting to get the job back. It isn't a direct consequence of the 2021 layoffs, but the fact that his removal is being reported through that frame at all says something about how long a reputation like that one actually follows you.

Employer Brand POV

Separate your redundancy language from your turnaround language before either one goes out. Diageo, Starbucks and Sanofi all reached for a word that implies strategy and momentum; KPMG's cuts trace back to a scandal it can't spin that way, and treating all four the same in your own comms would be dishonest by association. If your organisation is cutting people this quarter, decide first whether you're actually turning something around or just cutting costs, and use the word that's true, not the one that sounds better in the town hall.

Related reading on EBN: What IKEA's AI Reskilling Story Looks Like Now It Has Made Layoffs Too

The productivity number finally moved. The unemployment number didn't get the memo.

The Financial Times reported this week that UK productivity is showing its first genuine signs of improvement in years, a welcome number for a government under pressure over years of weak output. A day before official labour market data was due, a separate survey covered by the Economic Times found UK employers stuck in what researchers are calling "low-hire, low-fire" mode, neither adding roles nor cutting them at any real pace. Across the North Sea, Statistics Netherlands reported the Dutch unemployment rate rose to 4% in July, up from 3.8% in June and the third consecutive monthly rise, with 404,000 people now unemployed and the increase concentrated among men under 45.

None of these numbers contradict each other technically. They're just measuring different things, in different places, at different points in a labour market that can't seem to move in one direction for more than a month at a time.

Employer Brand POV

Publish the hiring number your own function is actually seeing, not the one from the trade press. UK productivity is improving in aggregate and Dutch unemployment is rising in the same fortnight, and both of those are true at a national level while telling you almost nothing about whether your open roles are filling faster or slower than they were in June. If your business is quoting the good number in a town hall, publish your own hiring and attrition figures alongside it, because your employees already know which one actually describes their week.

Related reading on EBN: But AI Will Make Us All More Productive. Won't It?

Hybrid was sold as the happy medium. A 7,704-person study says otherwise.

Hybrid work has been treated for years as the sensible compromise, all the flexibility of remote work with the connection of an office. A peer-reviewed study out of the University of Texas MD Anderson Cancer Center, covered by HCAMag, tested that assumption against 7,704 employees split across remote, hybrid and onsite arrangements, and found fully remote staff reporting the highest wellbeing, at 4.22 out of 5, ahead of hybrid workers at 4.12 and onsite workers at 3.89. Turnover followed the same pattern: 7.8% for remote staff within a year of the survey, against 8.8% for onsite.

The study's co-author, Dr Stefanie Johnson of the University of Colorado Boulder's Leeds School of Business, pointed to autonomy rather than proximity as the likely driver, and the research found no support for the standard argument that remote staff feel less connected to their organisation's culture. Only around 13% of employees currently work fully remote, according to the Stanford-affiliated WFH Research, which makes this a genuinely uncomfortable data point for the return-to-office mandates that have defined the last year of workplace policy.

Employer Brand POV

Trust the wellbeing data over the return-to-office instinct, at least enough to ask why the gap exists before closing it with a mandate. A 7,704-person peer-reviewed study found remote staff reporting higher wellbeing and lower turnover than onsite colleagues, with autonomy as the likely driver, not proximity to a manager. If your RTO policy is justified by a connection argument this study found no evidence for, it's worth checking whether the policy is solving a real problem or just a comfortable one.

AI is quietly taking something back, and workers are quietly protecting themselves against it

The Conversation reported on new research into what happens when people hand thinking tasks to AI rather than just execution tasks. A 2025 study by the University of Melbourne and KPMG, surveying 48,000 people across 47 countries, found two in three people use AI regularly at work, while 61% have received no training in it at all; a July 2026 study by the Centre for AI Safety found even top-performing AI agents fail roughly 85% of projects to a standard acceptable for commissioned work. Researchers are calling the resulting output "work slop," polished-looking but structurally unreliable, the same failure mode that forced Australia's government to accept a partial refund from Deloitte last year over a commissioned report containing fabricated citations.

Fast Company reported Microsoft and LinkedIn's own research pointing in a similar direction from a different angle: as AI automates execution, employee stress is at record highs and loneliness is doing much of the damage, according to Gallup data cited in the piece, because time once spent talking to colleagues is increasingly spent prompting a model instead.

Workers already seem to be responding to a version of this even before anyone's named it for them. "Career cushioning" is booming, Fast Company reported separately, driven partly by a World Economic Forum estimate that puts the average skill's half-life at around four years, closer to two in AI-adjacent fields; the behaviour looks less like job-hunting and more like a handful of hours a month spent quietly proving you've kept pace. Business Insider carried a blunter version of the same anxiety in Gregory Brooks, a 57-year-old in Rochester, New York, describing the "humiliation and panic" of job hunting after a voluntary resignation, uncertain whether his age reads to employers as experience or expiry. And the Associated Press, in a piece picked up by Yahoo Finance, reported that the monitoring tools many employers adopted during the pandemic are now generating detailed "dossiers" on staff, according to Wilneida Negrón of the worker-advocacy nonprofit Coworker, often shared with outside data brokers without clear disclosure, according to a joint investigation by Vanderbilt, Northeastern and UC Berkeley researchers.

Employer Brand POV

Protect the thinking time your AI rollout is quietly taxing, and say so before your best people start career-cushioning around you instead of trusting you with it. Two in three people in a 48,000-person study use AI regularly, and 61% have had no training in it at all, which is exactly the gap that produces the "work slop" this week's research warns about. If your employer brand still promises growth and stability while your own staff are quietly building a certification on the side in case the ground moves, that gap is the actual story, not the AI adoption number in the board deck.

Related reading on EBN: Have We Reached Peak Human Intelligence?

The Cut List

None of this week's numbers disagree with each other by accident. UK productivity is up, say the economists. Dutch unemployment is up too, for the third month running, say the statisticians. Diageo, Starbucks and Sanofi all reached for the same manageable word, turnaround, to describe cutting people, and KPMG didn't get the luxury of that euphemism, because its 500 cuts trace back to a scandal it started itself. Meanwhile, 7,704 healthcare workers told researchers, unprompted, that being trusted mattered more to their wellbeing than being watched. Put those findings on the same table and a pattern shows up that no single headline was built to reveal: the organisations doing the most cutting, watching and rebranding this week are also the ones with the least evidence that any of it is actually working.

Takeaways

• Diageo's headcount has fallen 6.4%, with cuts landing hardest in its North American business unit, as CEO Dave Lewis works through a profit slump.
• Starbucks confirmed more than 200 corporate layoffs in a WARN Act filing, 120 tied to staff who declined relocating to its new $100 million Nashville office.
• Sanofi will cut 229 workers at Blueprint Medicines, the Cambridge biotech it acquired for $9.1 billion in 2025, between October 2026 and June 2027.
• KPMG is cutting roughly 500 positions, including dozens of partners, following a scandal over the misuse of confidential client information.
• UK productivity is showing its first genuine signs of improvement in years, according to economists cited by the Financial Times.
• The Netherlands' unemployment rate rose to 4% in July, the third consecutive monthly rise, according to Statistics Netherlands.
• A peer-reviewed study of 7,704 healthcare employees found fully remote staff reporting the highest wellbeing (4.22/5) and lowest turnover (7.8%), ahead of hybrid and onsite colleagues.
• A 48,000-person study found two in three people use AI regularly at work, while 61% have received no training in it at all.
• "Career cushioning" is booming, partly driven by a World Economic Forum estimate that the average skill's half-life is now around four years.

Trending on EBN

SOURCES