Hi Reader

Every so often, a run of ideas lands that makes you rethink things you thought you understood. This week delivered a handful of those: about how brands grow, why acquisitions falter, what keeps users loyal and how quickly AI might reshape whole sectors. Let’s dive in.

Enjoy!

The surprising truth about how tiny brands grow

The Ehrenberg-Bass Institute just published a five-year study on “tiny brands” (those under 1 per cent share), and it punctures one of marketing’s most cherished myths: that small brands grow through niche loyalty. They don’t. Most have less loyalty than expected, and even the rare “excess loyalty” brands don’t outperform. What matters is penetration. The tiny brands that scaled did so by finding far more new buyers, with loyalty rising only as a by-product of that broader reach. It’s a useful correction for anyone tempted to pour limited energy into superfans, because even the tiniest brands win by reaching more category buyers, not by deepening loyalty with a narrow base. Read the report here.

Why most acquisitions fail

A useful piece from Professor Harry Kraemer (Kellogg School of Management) makes a point that most deal models ignore: the biggest destroyer of acquisition value is talented people leaving. Studies put failure rates at 70–75 per cent, and the pattern is consistent – teams freeze, uncertainty festers, the best employees leave and the acquirer suddenly owns half the capability they thought they were buying. Kraemer’s advice is to communicate early, over-explain everything and treat talent retention as part of the deal price. Worth reading for anyone planning to buy or sell in 2026 – check it out here.

The real battleground in AI is its UX

Google launched Nano Banana Pro recently with the usual fanfare about sharper images and faster inference, yet the more interesting story is how little these capability jumps matter once users feel at home with their current AI offering. Behavioural research shows products need to be nine times better to overcome the comfort of the familiar, which is why people stay loyal to tools like Claude or Netflix despite credible alternatives. Familiarity creates switching costs that raw performance cannot. AI companies still talk as if benchmarks win markets, but the real moat is the interface that becomes second nature. The products that win are the ones people do not want to leave because leaving means re-learning how to use them. There’s a fascinating short piece on it here.

The outdoor brand trolling Flat Earthers (brilliantly)

Outdoor apparel brand Columbia has launched one of my favourite stunts of the year: an open challenge to Flat Earthers to locate the “edge of the Earth” and win… the entire company. CEO Tim Boyle delivers the message himself in a short film wandering through his HQ, casually offering up office furniture, gondola cabins and mounted deer heads to whoever proves the planet has an edge – provided they wear Columbia gear to do it. It’s a perfect example of how irreverence, when backed by genuine product confidence, can cut through a crowded category. Creative Review has the story.​

Zuckerberg dials down the metaverse dream

Mark Zuckerberg is rewiring Meta’s future, shifting billions away from the metaverse just four years after rebranding the entire company around it. Reality Labs has burned more than $70bn since 2021 and now faces budget cuts of up to 30 per cent, with hardware, Horizon Worlds and whole teams reportedly on the line. Investors cheered the pivot, reading it as a long-overdue acknowledgement that VR never found its must-have use case. The bet now is AI: Meta spent roughly $72bn on it this year alone, putting almost every non-AI initiative under scrutiny. And on the last earnings call, executives didn’t say the word “metaverse” once – a small but telling sign of a company moving on. Find out more here.

How to design meetings people don’t dread

Most meeting pain isn’t caused by talkative colleagues or bad coffee – it starts earlier, when nobody is clear on why the meeting exists. A new guide from FM Magazine walks through the fundamentals: define the purpose, cap the invite list, appoint a chair, keep the agenda tight and use AI tools for scheduling and notes. None of it is radical but applied consistently it turns meetings from energy drains into fast, focused working sessions. Get the lowdown here.

Why weirdness is disappearing (and why we should worry about it)

A Substack story by Adam Mastroianni suggests that society is experiencing a “deviance recession”: fewer misfits, fewer risks, fewer new ideas. From teen behaviour to architecture to the internet itself, everything is converging on the safe, the expected and the optimised. His argument is that as life gets safer and more valuable, people become less willing to risk anything – including doing something original. It’s a revealing lens on why creativity feels harder today, and why businesses may need to deliberately engineer weirdness into their teams. Get the story here.

The workforce simulation that shocked MIT

MIT has built a digital twin of all 151 million American workers – mapping 32,000 skills to every new AI capability. Their conclusion is that today’s AI could technically automate work equal to 11.7 per cent of US jobs right now, concentrated not in Silicon Valley but in the cognitive back-office roles surrounding manufacturing, healthcare and finance. The report is fascinating because it separates hype from exposure: the systems aren’t deployed yet, but the capability exists. Essentially, it’s a clear early-warning system for what’s at stake over the next five years if you’re a knowledge worker. Check out the MIT study here and – for those who want to ensure they’re ahead of the curve – here’s a good Reddit thread outlining the free AI learning courses run by Google.

AI prompt of the week: design a retail sales promotion that works

If you want an AI assistant to help you build a proper commercial plan – not just a list of discounts – you need to give it a structured brief. This version forces the model to think strategically, ask clarifying questions and tie activity to measurable outcomes.

Act as a senior retail sales strategist. Your task is to design a six-month sales promotion plan for my business. Before you begin, ask me 5–7 targeted questions about our product, customer base, pricing, seasonality and current challenges. Once you have my answers, produce a detailed plan that covers: messaging, pricing tactics (e.g. limited-time offers, bundles, loyalty rewards), promotional channels, a timeline and clear KPIs. The plan should focus on driving engagement, improving conversion and sustaining sales lift beyond the promotion period.

The LOI: small document, big consequences

Think of this as the first draft of your exit – whoever defines it controls the negotiation that follows. These seven points are where buyers signal their intentions and shape the path to closing.

Drop me a line

If you’ve come across anything surprising, counterintuitive or just plain useful lately, I’d love to see it. Some of the best items in this newsletter have come via subscribers. Thanks for reading and see you next Sunday.

Cheers!
Adam

Join the Exit Mode Masterclass!

​Sign up for free and learn how to get exit ready in just six months

​
​Unsubscribe · LinkedIn​