The last time I was in Target - a few years ago – some blank faced assistant manager was trying to wave me into a self-checkout queue seventeen people deep and snaking back across six or seven of the 20 or so closed cashier lanes. He had one phrase to affect compliance and it was so bland I can’t remember what it was. I left and now shop local where I can. Not to support Memphis, but local stores rarely have the economies of scale required to treat customers like they are being processed out of a refugee camp.

On paper both roles are customer service and one terrified automation is cheaper than several cashiers. In reality, the cashier provides a service and is hopefully pleasant about it, the automaton is blandly giving orders without explanation. So naturally, I hate him.

The view from corporate HQ isn’t much better. Companies are famously scrapping layers of middle management for AI decision making - making a shrinking cadre of middle managers responsible for a more decisions over which they have no control and can’t explain. At best, this exponentially increases the likelihood of employee burnout; at worst, revenge. Either is hard to catch until it’s too late.

Chesterton’s Fence

Both cases are examples of tearing down a Chesterton’s Fence. Early 20^th century writer G.K. Chesterton offered one of the most enduring principles of good-decision making: political, cultural or corporate. The “happy reformer” happens on a fence across a road and, seeing no purpose for it, declares that it should be torn down. Chesterton’s response is wonderfully irritating: “If you don’t see the use of it, I certainly won’t let you clear it away. Go away and think. Then, when you can come back and tell me that you do see the use of it, I may allow you to destroy it.”

In short:

“Don’t remove a fence until you know why it was put there in the first place.”

This isn’t to cling to the past, but merely think it through before barreling into the future. Businesses are littered with invisible fences. Some need clearing, some don’t, some are technical and some regulatory. A fair share of the banking regulation the sector is trying to scrap was put in place in the wake of the 2008-09 financial crisis. Regulatory fences cleared away in the 1980s were put up after the crashes that triggered the Great Depression. Time will forget the memory while preserving the rule. Which leaves the next generation convinced that they’ve discovered an unnecessary complexity or friction.

Ask the Follow Up

Artificial intelligence is brilliant at improving observable workflows and identifying inefficiencies in duplicate steps, bottlenecks and manual processes.What it doesn’t do so well is put these functions in the context of why they were there in the first place. It can only see the quantifiable patterns that it can see. Evolution is hidden.

Complex systems - even business systems - operate more like eco-systems than machinery where feedback, connections and context matter. Every fence protects something: efficiency, reputation, compliance or even the sort of catastrophic error that happens once a decade… or generation. Which is why everyone forgets what it was for in the first place.

Most initiatives to soup up a system ask the first question: What happens if we remove this? But fail to ask the second: What disappears with it?

The Field Guide

Before removing any long-standing process, habit or policy, try running it through a simple test.

1. The Origin: Who built this and what problem was it trying to solve? It was the answer to a question that once mattered. Find the question.

2. The Stakeholders: Consult the people operating within the process. Granted, they may be have a self-interest in preserving the status quo, but they may also reveal something that doesn’t appear on a dashboard.

3. The Replacement: If the original problem is irrelevant, or if there’s now a better solution, then clear the fence. But replace it with something. Good reform is subtraction paired with design.

There’s a broader lesson that applies to markets and even personal routines. It is easier to eliminate a bad habit once you know what itch it is scratching: Stress relief? Predictability? Control? Until you understand the original function, you’re unlikely to replace it with something that lasts.

Know what’s happening. Know what it’s going to cost you. If your team is redesigning or positioning for a new market or product the biggest risks are often the ones hidden behind yesterday’s fences. Drop the 4717 a line for a decision framework to distinguish legacy bureaucracy from hidden strategic safeguards before change becomes an expensive lesson.

Let's Talk

So you close that sale. Now get off the phone as fast as possible and make another call. If you are on a trading floor a lot of noise will come over the line, but the only thing a customer needs to hear is success - that confidence in your voice. Humans, you see, are social little monkeys and that is the way we perceive competence. And competence is how we alleviate uncertainty. As a rule of thumb it’s one of our worst traits.

Let’s define our terms: competence is essentially your ability or expertise in something. Generally it comes from exposure, practice, intent and time. Natural talent and dazzling genius are helpful, but not necessary. Confidence is, more or less, the ability to project that competence. Like dazzling genius, some expertise is nice when projecting all that self-assurance, but at best it’s irrelevant. At worst it will actively work against you.

I once read a study that actually plotted the confidence/competence axis. You’d think (and hope) you’d get one of those graphs where both axes start at zero gently increase together. They don’t. You’d be wrong. Again. What you get a U-shaped graph with high levels of confidence at the beginning and the end. Like a horseshoe, but not nearly so lucky. A person with almost no competence gets very confident with enough self-belief, YouTube and AI doing all the heavy lifting. As we gain competence and real experience, however, we naturally start to lose our “irrational” confidence as we begin to understand a) the enormity of what you don’t know and b) that most people are just making it up too. Over in the far high end is someone with a rational confidence justified by competence. Should this person have much hair at all, it will be grey.

Discovery v. Confirmation

Confidence, in ourselves and others, is a mental and emotional labor saving device: It tamps down on the uncertainty washing over us and saves cognitive calories in a world of information overload. And for that it is rewarded handsomely. The executive who inspires confidence and certainty goes further than the one explaining the complex reality of a situation. This isn’t all BS, a certain amount of irrational confidence is often the entry fee for extraordinary achievement. The trouble is when success starts feeding the confidence machine. Success has a nasty habit of convincing us that our judgement is better than it is. Unlike our mistakes, we take full credit for our successes and discount the role of dumb luck and everyone else involved. The QED being that when we do make mistakes, we cling to them for longer.

Technology has made this problem worse because you can build a model to prove almost anything. The data become less a tool of discovery that one of confirmation to “prove” your bad idea. At which point you become Bernie Sanders.

Remember, if you torture the data long enough, it will confess anything. So will graduate programs.

The MBA Playbook Trap

Business schools do a valuable thing: they expose students to proven strategies, famous turnarounds, and legendary entrepreneurial victories. They instill confidence, but I suspect that many MBAs never fully recover from case studies. Once you’ve fallen in love with a particular framework, you start seeing it everywhere. Playbooks are addictive because, like anything addictive, they essentially short-circuit something tedious in the brain - like thinking it through. Not too long ago, every new start-up was billed as the next Uber, but the ride-hailing app stepped into a very singular space to solve a unique problem: an over-regulated, corrupt, centralized and expensive taxi concession. There aren’t that many of those around to replace. So your start-up wasn’t the next Uber - it was just an app.

Instead of adapting the playbook to the environment, executives adapt their perception of the environment to fit the playbook. It’s a little like studying tank warfare and then deciding the mountain range in front of you isn’t actually there. The map becomes more important than the terrain. And that is idiotic.

The Field Guide

How do you tell confidence from competence? And where do you place it on the confidence curve?

In dealing with yourself, a modicum of self-awareness makes this easy but uncomfortable. Write it down - the nature of the problem that need to be resolves, what a resolution would look like, and how it would work given the nature of the known constraints - physical, social and political. That will give you a good picture where you are and, as a bonus, increase your competence.

With a team, you have no self-awareness, only black boxes. I have a friend who used to make people on his team explain everything like they were talking to third graders. This delivered two positives: laid out the problem and knocked the irrationally confident off their high-horse. The frame:

1. Can you explain it?

2. Can you explain it in reverse?

3. And crucially can you explain the dynamics of the larger system you are trying to affect?

If you can sort out the dynamics first, it will save you heaps of time, frustration and money. To know when you need to hit the brakes and start asking these questions, you need an early warning signal. And here it is:

Against all of your monkey instincts, the faster and louder the response, the more you should question it.

If you and your team is looking for solutions that plot for confidence and competence

Let's Talk

That explosion in financial and logistical innovation during the eighties literally - not figuratively – changed the world. It enabled the widest and fastest global market expansion in human history. I knew it at the time and I was in high school. It was so obvious that even the great Communist powers in Moscow and Beijing decided that Karl Marx was, not the put too fine a point on it, full of it.

Perhaps it was capitalism’s golden age. In market history, however, it was a golden anomaly. Those innovations produced leaner inventories, fewer redundancies, tighter staffing, and hyper-globalized supply chains. A couple of technological revolutions later and we are still obsessed with moving faster, from product cycles to AI adoption to saying things like “decision velocity.” Markets were beautifully optimized for that golden age as it existed.

Now take a good look at the most operational efficient person in your life – chances are that she does not like surprises. And she is certainly no fan of: Pandemic whiplash, war in shipping lanes, cyber-risk, energy shocks, spiking interest rates or inflation, supply-chain nationalism and the truly ill-advised return of the holy war. All of which is producing a level of variety and noise in a system built on efficient stability that is cannot absorb.

Everything but Comprehension

It is always dangerous to mistake movement for progress, but the knee-jerk reaction to all of this is to accelerate activity rather than improve understanding into what we might call “performative adaption.” AI pilots and innovation programs without strategy, transformation initiatives and dashboards that measure everything but comprehension.

Speed alone simply isn’t the stand-alone advantage that executives think that it is. The make or break advantage is getting to market at the right time – it’s no good dawdling – but the ability to adapt intelligently under changing conditions is the ability know the right time. Or at least have the best guess, and that will generally do. As the dictum on marksmanship goes: Slow is Smooth, Smooth is Fast. To put speed first is to sacrifice accuracy.

So what does a company look like if it is designed to benefit from chaos, not just survive it? As luck would have it, that’s the subject of or latest 4717 Insight – Five layers of the Anti-Fragile Company – exploring how firms become structurally stronger under volatility by reframing:

1. redundancy,

2. Keeping options open,

3. decentralized decision-making,

4. narrative intelligence and

5. “skin in the game” incentives

In short, build a company that doesn’t panic every time the map changes. AI won’t help you - as we’ve written before - its brilliant at a lot of things, but it is predictive ability is terrible if the model changes. And well, just look around you… the model is drastically changing.

The New Competitive Advantage

“Maximum velocity” models sound like loud, growing cars, but they are easily derailed. The crucial strategic shift that we’re seeing is from efficiencies to stability; ability to absorb shocks without triggering organizational hysteria or paralysis. That requires a degree of adaptable market innovation in the face of the unknown which, admittedly, is a hard sell to any body that involves a committee.

Uncertainty is uncomfortable for most people, but for people who rode to the top on a spreadsheet, the discomfort is neurotic. Corporations have almost certainly smothered more innovation than they’ve spawned avoiding uncertainty. The sort of semi-random trial and error market feedback needed for break-through innovation also requires being wrong in little ways during development. If managers were comfortable with that they’d be in commodities or VC.

Historically, the companies that tend to gain from technological shock are those that possess: spare capacity, distributed intelligence, and cultures capable of learning faster than competitors. Nature figured this out millions of years ago.

The Mental Model

This may get uncomfortable, so pour a drink. The awkward question leadership should probably ask more often: “What assumptions can our business not afford to change?” Get a decent whisky, because that question tends to produce an fiddly silence. It’s also where the interesting analysis starts.

Only stop when you think that you’ve saved the world – or want to share any of it with your dog – then drink a glass of water and go to bed.

Head to 4717 Insights to download Five Layers of the Anti-Fragile Company, exploring how organizations can position themselves to gain from uncertainty instead of merely endure it.

4717 Insight

The sad fact is that most companies prize the guy with the quick, decisive answer. He’s sharp, he’s loud and sweats confidence. This Master of the Universe can remove the pesky pall of uncertainty by solving the most obvious and immediate problem: The next visible number. He’s also probably creating a funnel packed with new problems.

Cut the price. Outsource the factory. Automate the department. Replace the analyst with AI. Reduce labor friction. Move fast and push growth and the metrics bloom for a glorious quarter or maybe six. Investors react! So do markets. They push back… hard.

Psychologist Daniel Kahneman spent years documenting the human tendency toward cognitive shortcuts — our habit of overweighting immediate visible outcomes while underestimating delayed systemic consequences. We do this especially under pressure or during technological shock or when compensation packages are tied to this quarter instead of this decade. And this is why second-order thinking increasingly matters more than raw intelligence.

This is not technical shock, some of these potholes are tried and true: Take price cuts. First-order thinking says lower prices increase demand. Which they briefly do, until customers begin associating the brand with cheapness. Competitors slash prices too. Margins collapse across the category. Suddenly the premium positioning — that delicate little psychological structure built over fifteen years of advertising and customer signaling — vaporizes in eighteen months.

Then there is the outsourcing efficiency we are all currently regretting. Yes, it lowered labor costs and delivered higher margins. But now our critical manufacturing base or components suppliers now exists beyond a geopolitical fault lines half a planet away. Political instability becomes operational instability. Suddenly a diplomatic dispute in the South China Sea is capable of wrecking your inventory forecast in Memphis by Tuesday afternoon.

The spreadsheet captured the savings, not the exposure.

AI, though, is a technical shock: One employee now does the work of five and operating costs fall. And that blowhard with all the answers starts banging on about “scalability.” The second-order effects are harder to visualize: The organization quietly stops developing human capability. Junior employees never learn foundational judgment because the machine performs the cognitive apprenticeship layer for them. Five years later the company discovers it possesses thousands of prompt operators and order takers. With almost nobody capable of actual synthesis, strategic reasoning, or independent thought. The organization became faster and much dumber.

Most organizations think in straight lines while operating inside complex systems. First-order thinking is becoming automated: AI can optimize, algorithms can accelerate and software streamlines. Second-order thinking – and that is where you are going to avoid that funnel of future problems – still requires judgment, pattern recognition and some instinct into human behavioral psychology. In short the ability to imagine not merely an outcome, but the chain reaction after the outcome.

The Field Guide:

The IT revolutions didn’t create the problem, it just sped it up as systems become faster, denser, and more interconnected. So ask five questions before major decisions:

1. Action: What exactly are we doing?

2. Immediate Gain: The why – What obvious reward are we pursuing?

3. Likely Reaction: How will customers, competitors and the rest respond?

4. System Adjustment: How does the market adapt? Does the why still hold?

5. Long-Term Exposure: What hidden vulnerability have we accidentally created?

It sounds easy, but second-order thinking is uncomfortable. It slows that sweet dopamine rush of immediate optimization. It forces you to confront tradeoffs your team would rather ignore, but the markets won’t. The companies that survive the next decade will not necessarily be the fastest. They will be the ones capable of seeing around corners while everyone else is staring at dashboards.

If your firm is entering a new market, launching a product, or expanding geographically this year, we help leadership teams pressure-test assumptions before reality does it at full speed.

Let's Talk

If David Ogilvy didn’t actually say “The trouble with market research is that people don’t think what they feel, they don’t say what they think and they don’t do what they say.” he should have. People aren’t lying, entirely, but most of us don’t have introspective access to our own motivations. From my own experience in advertising, I can tell you that focus groups won’t tell you much more than 1) the opinion of the one or two strongest opinions in the room, and 2) that humans are very social creatures.

E-commerce, that all seeing snitch to our when no one is watching id, does a great job on recording what we do, but not why. Feeling drives most of what we do even if, unlike the scientific process, they don’t come with explanations. I suspect that this because we are better off not knowing why we do things. We’d absolutely hate ourselves.

Ultimately, you don’t really care why the customer is buying that car, only that she does. Until she doesn’t. Then you’ve got a problem. Counter-intuitively, the more data we have, the less room that there is for factors that are hard to quantify or compute. The real why, however, matters a great deal when it comes to product innovation or just trying to gain market share. Subconscious motivations is crucial if you want to change behavior. And that is hard to know when the customer can’t articulate it herself.

I’ve written before that when you hit upon a “black box” – and there is no blacker box that human motivation – it is better to manage it’s output rather than control it.As social creatures, a great of those black box feelings get justified under a widely held consensus. It will make superficial sense, and so remove uncertainty on a logical level. On the social level many consensuses are widely held simple because they are widely held. And that, if you are trying to introduce your sensible innovation into the market, is your barrier.

Typical is Better than Average

At the risk of causing your corporate betters to pull what left of their hair out, I’d suggest a low-cost, out-of-the-box approach with a higher probability for success: Ignore the middle of the market and focus on a few outliers. It’s hard to understanding this looking at a bell curve, but if you stick to the data, the average person isn’t a typical person. In fact, the mathematically average person is very rare, if non-existent. On the other hand, you can’t swing a cat without hitting three typical people… and getting filmed by a dozen more.

The typical person is, as their mothers would gently put it, “special”: quirky in their own way, pleasantly eccentric or “a bit of an asshole.” Like the average person, an outlier that is a completely unsocialized dumpster-fire of a human is also mercifully rare and won’t affect your sales. Focusing on a quirky outlier will deliver underserved markets, where you will find your early adaptors less bound by the widely held consensus for the simple reason that they’ve made peace with being somewhat eccentric.

The QED is that you face lower entry barriers and quicker uptake at first. If you succeed, the second order effect is that these quirky souls will provide the social proof against a widely held consensus standing in the way of mainstream adoption. If a launch fails, costs and wider reputation is saved - because it was always a niche offering.

The Field Guide

If you’ve ever had the terrible idea that you’d like to be a novelist, here is your chance. Create a character profile of an early adaptor. Don’t write anything as silly as “who wouldn’t want this?” That will have you innovating for average, you want to shoot for typical. And people, typically, are a little weird. If they weren’t they wouldn’t be early adopters.As an outline:

1. Basic Information: name, age, sex, job and income

2. Problems: How is your product or innovation going to help them

3. Positives: What are their hopes and ambitions that put your product in a good light?

4. Negatives: Or put them off it?

If noting else, the human that fits this profile will be grateful that someone knows that they exist. Then you have a maven who will spread the word.

The first step to a market is an audience. If your team needs help an audience strategy.

Let's Talk!

It was in the “Murff of Arabia” phase of my career when I first heard the dictum about marksmanship under pressure: Slow is Smooth, Smooth is Fast. It’s counter-intuitive at first glance, but clear to anyone familiar with firearms. When you focus on speed, it costs you in accuracy that won’t be recouped. However, if you focus on accuracy, there is a small short-term cost in speed that is recovered quickly as speed increases with competency. It struck me, Sig Sauer P226 in hand, that the same principle applies to strategic thinking and creative problems solving.

In a world where data comes at us in a fire-hose and quick, reactive declarations are put on a pedestal, Slow is Smooth, Smooth is Fast may help you outwit a cyber-verse of instant answers and everyone using them. Like a gun, you do need to have an existing mental model of strategy thinking in place and know how to use it. After that, however, what you really needs is a clear sight picture: know what you’re aiming at and the context in which the target appears.

Slow is Smooth

In his brilliant and mercifully short book Hare Brian Tortoise Mind: How Intelligence Increases When You Thin Less, Guy Claxton writes about the importance of the deliberative mode - d-mode - in problem solving, and how it works with that much faster subconscious pattern detecting engine. When facing a puzzle the best bet is to learn what you can about it, put that picture in context, then shut the noise down. The pothole most people step into is believing that more data is always better. In my experience, context will turn relevant data into information, while allowing you to tease it out from the noise. Then slip into d-mode where you think about the problem without further input, for about ten minutes.

Honestly, this may be the hardest part for the spazzed-out modern employee. Perhaps impossible if you don’t have a door to shut. Until you get used to it, that ten minutes will last forever. You can’t wander off and expect the subconscious to do it for you because it shut down when the fire hose of data was still roaring and is still operating on auto-pilot. You have to give the puzzle some deliberate thought. Go for a walk if that helps, but stay focused. Repeated studies show that these ten minutes of thought will drastically change the way you see the puzzle. Don’t worry about writing in down yet. It is only by processing information in d-mode that you discover - and understand – what you already know about the problem. That is the smooth created by the slow.

Smooth is Fast

Now that you’ve smoothed your thinking, put pen to paper and write out the problem, the constraints, any second order effects to create that all-important “clear sight picture” of what you need to happen. At that point you are about 75% to a solution that has been turned over several times while being applied to the vast databank of models and patterns of the subconscious. The cross-referencing between the two modes is key here. Your thinking and approach will lose that cork tossed in an angry sea feel. It will become a smooth thought pattern.

The Field Guide

What looks like wasting 10 or so minutes, and another five or ten writing it out, is likely to save you hours of work later on. And probably a bucket of money in the bargain. Unfortunately, if you work in an office, the hardest part of the above isn’t applying the mental model, but being allowed to execute it. I once had a manager say to me, and I quote: “Murff, quit being right and get back to your desk.” Most managers have a dangerous love of certainty preferring quick certain decisions to more creative approaches with higher aggregate payoffs that are hard to calculate.

The short answer is try not to look like this is what you’re doing, and for Pete’s sake don’t announce what you’re about to do.They will stop you. As your strategic thinking smooths out toward a definite target the frenetic noise produced by information overload is replaced by clear signals of cause and effect as it applies to the puzzle at hand. Effective strategic thinking and problem solving is in the balance between short-term gains and a sense of medium -to -long term second order effects.

The good news is that if you hit a couple of bull’s eyes with creative and innovative problem solving, then not only will you have a door, you can do any damned thing you want.

Looking for solutions that aren’t obvious to competitors and their chatbots?

Let's Talk

That was the way that CEO Brian Moynihan explained what Bank of American was up to after posting $8.6bn in quarterly profit cut 1,000 jobs. Together, the six biggest US banks made a combined $47bn last quarter – up 18% year on year – while slashing some 15,000 jobs. Wall Street, it appears, has pivoted from its line that AI will do all the boring work and leave the high-value work to high-value cats in their deal slides. In a particularly ironic cut, Citi is slashing jobs in its “AI Champions and Accelerators” program. The efficiencies are showing up in the quarterly numbers, investors are pleased… so what could possibly go wrong?

Superficially, replacing the quants – the number crunchers – makes sense because that is the sort of thing that AI does really well without people like me adding a lot of caveats. Since the early 1980s, when Jim Simons pioneered quantitative trading, much of whats done these days is already fully automated anyway. AI will produce the same output, and be vulnerable to the same blind spots as before. It will just achieve this faster and cheaper.

The forward looking grenade will show up in how those new black boxes will affect the entire system. As I’ve written before greater efficiencies very often lead to instabilities within a complex system. Predictive AI and quantitative trading are essentially the same thing and are very capable provided nothing changes. All the trading data is backward looking. It was Jim Simons’s good luck to pioneer quantitative trading in the dying gasps of the Cold War where the Soviet bloc and China began making moves to open up their economies to a stable global market. Since the 1980s the only change to the system has been, more or less, gentle but constant growth. I won’t describe the 2008 Financial Crisis as a hiccup – it was more full-body dry heave – but slowed growth over a decade, it didn’t reverse it.

In finance, timing is everything. It is the bad luck of AI to find its legs on Wall Street at precisely the moment that all that backward looking market data the quants trained their models on is changing. The universe of gently expanding growth that has prevailed for the last 30 or so years is collapsing, and the patterns it provided aren’t going to be of much use going forward.

And even that isn’t really the problem.

Second Order Effects

That the AI investing boom has distorted markets and we are teeing ourselves up for another financial correction is tiresome, but these things happen. The snake in the grass is that Wall Street, and the system that it underpins, is erasing its own talent pipeline in the name “eliminating work & applying technology.” A vaguely sentient human can see that the global economies are being rewired as alliances shift and commerce is weaponized. The old trading patterns won’t map as we increasingly rely on technologies that rely – entirely – on prediction rooted in old patterns. Telescope the current trend a few years and the industry is virtually assured not to have enough human managers to absorb the “noise” and chaos generated by all the AI black boxes. Money is a very, very visceral issue for humans. The knee-jerk market reaction will not fit into any predicable pattern which will amplify the amount of noise coming out of the models.

And when that happens the system will collapse.

The Field Guide:

Honestly, just stop and use your imagination. You want to use the power of AI to improve your position here, try prompting a report on whatever it is your chewing on with a conclusion and recommendations. Then generate a second report arguing the opposite conclusion. Read both with a pen in hand. Then put the pen down and – this is the hard part – sit for ten or fifteen minutes thinking about the matter with no interruption, no input. Just stay focus for ten minutes. You can’t out think AI, but that’s how you outwit it.

Preparing for a market expansion or product launch and want to test second-order risks? Let’s Talk!

Let’s unpack that start-up mantra “move fast and break things.” You’ll hear it when corporate behemoths - generally at the wrong end of an earnings report - start banging on about getting back to the explosive growth start-up mode. It’s as romantic as people who live by quarterly reports get, but beware: If you are going to move fast, you better know what needs breaking.

That’s not always obvious because as companies scale, decision-making becomes abstracted from market realities, despite bringing greater resources and more data to the problem. Responses become sluggish and consequently, one step behind the faster and smaller rivals.

In my experience, larger companies break the wrong thing, creating a disconnect between a decision and its real-world consequences. As “decision distance” grows, systems become unstable because they are insulated from reality in a way that start-ups aren’t. The variable isn’t generally size, or even the added complexity of the larger system, but in the way various components of a complex system interact with each other.

It might help to visualize a start-up as a single, market-facing “operations” system. As a business grows, it will add more market-facing “operations” systems that must be managed within a larger system. Without getting too into the weeds, think about Benoit Mandelbrot’s concept of fractals: Nearly all complex systems – broccoli, river networks, humans, snowflakes, and lightning bolts – are made up self-repeating smaller systems. It is a mathematical language to describe the organized chaos of the natural world. In the case of living systems, those fractals must have some way to coordinate to move forward – a “management” system.

Why Start-ups Feel Smarter:

The short answer is a tighter feedback loop between decisions and a market-facing “operations” system. Market inputs produce a lot of information and “noise” which is handled on a tight feedback loop. The real-world reacts quickly – often painful, but incredibly efficient. Add another “operations” system and amount of information increases exponentially. You will need an additional “management” system to make company wide decision if you want avoid absolute chaos.

This necessarily creates looser feedback loops for companies as they scale by creating distance between market signals and a decision. Not only is the information delayed, but it’s likely been averaged so that while the facts remain, actionable signals are lost. In my experience, “operations” are also much better at dealing with rouge market outliers – which will really mess with your data quality.

Some of this is unavoidable. Some of it – like endless rounds of management sign offs – is not. Those loose feedback loops don’t just slow you down, they give off a false security from having more data and more people in on the sign-off. You think that you’re making better decisions, you’re often just making better defended decisions. And that’s not the same thing.

The Feild Guide:

As much as you can, allow individual systems to address most of this noise locally, as it arises, rather than through central planning. This will decrease the amount of “noise” coming out of each, increasing the available bandwidth of management systems to deal with problems being passed up the chain. Every fractal in the greater system also needs to communicate with each other in a way that the critical signal is understood. And those signals need to be sent fast enough to be the basis of action without too much lagging debate from within management. The second order effect here is decreased employee turnover, which increases the likelihood of this communication being handled informally.

A simple way to spot trouble is to ask two questions of any major decision:

1. How far is the decision-maker from the consequences?

2. How quickly will we know if we’re wrong?

Now place it mentally on this grid:

• Low distance + fast feedback = safe to move quickly

• Low distance + slow feedback = add early warning signals

• High distance + fast feedback = bring decision-makers closer

• High distance + slow feedback = decentralize to “operations” system

In the real world is that once market contact is made, speed of learning beats any planning. Growth doesn’t make companies dumber. It just pushes them further away from the consequences of their own decisions and makes market intelligence theoretical. The companies that scale well aren’t the ones with the best strategy decks. They’re the ones that systematically engineer their way back to painful market clarity.

If you’re scaling and want to avoid cognitive drift, let’s talk!

We’re all focused on a single, critical chokepoint, but it’s only one of many: From manufacturing and minerals processing bottlenecks in China and, for much of the world outsourcing national security to the United States is starting to look ill-advised. Chokepoints are a hazard of our obsession with system optimization, forming when easy-to-calculate metrics bury long-tail risks that are less definitive. Yet, just because something is hard to quantify or price, doesn’t mean that it doesn’t exist. We’re seeing the dangers in supply chains, but less obvious is in the decision process.Chances are you’re missing the most crucial choke - the one behind your headlights.

One of the major break-throughs is in allowing smaller firms to expand into new markets and punch above their weight. Product teams use it tech to synthesize customer feedback, marketing uses it for positioning and messaging, operations deploy it for forecasting and leadership uses it strategic framing Decisions are made with efficiency and certainty Speed increases and costs decline. And someone at least looks like an unassailable genius.

Remember that these efficiencies aren’t the same as stability and, like other chokepoints, the danger is hard to see at first. Nothing goes catastrophically wrong, but it’s isn’t quite right either. Where decisions were once formed through discussion - perhaps not rigorous but from different sets of eyes all with skin in the game - are generated by synthesis rather that construction and stress-testing. If you want to know if you’re there - just ask someone to explain and break-down a decision or tactical operation without notes. If no one can, chances are that the thought process was outsourced. Double check that work.

Practically, the solution to sidestepping these mental chokepoints is shockingly simple: Write process notes: informal jottings on how you came to a decision or position. This may be terrifying to the average human, but only for a moment and it’s worth it. Stress-testing both the thinking and the conclusion enables you to explain both without notes, backwards, sideways and on the far end of happy hour. It is a hell of a way to overcome prospect objections by creating a lateral thought process that includes the second-order thinking absent in so many “optimized” decisions.

The Feild Guide

So hang your process on this frame:

1. Immediate effect: What is the direct and intended outcome?

2. Behavioral response: How will individuals and organizations respond?

3. System response: How will the market or environment adjust?

Remember that systems are not static, they respond to competition and optimization with adaption. Break into a new market and their will be second order affects. You can use AI to create product positioning and launch plans, but everyone using it is drawing from the same echo-chamber. This is less innovation than convergence on a “sameness” that is seductively comforting but it will have an effect on the market as well. And if your strategy is built on the same logic, it isn’t going to do you any favors.

The remedy isn’t rejection of technology any more than blind acceptance. Use AI to crunch the numbers and set the table, as it were. A little structural discipline applied to your thought process will make it resilient and accessible as markets change and react.

The history of business is, in many ways the history of mistaking temporary advantage for enduring control - only to have the plan fly out the window once markets adjust. AI isn’t going to mitigate this, but accelerate it. Firms that survive the next hairy decade or so will be the ones that adopt the most advanced tools, but retain the ability to think independently of them and the mental chokepoints that they create.

If your team is preparing for a market expansion or product launch and wants to pressure-test second-order risks, let’s talk.

Half a dozen tabs across two monitors are staring at you: dashboards, Slack channels and that ChatGPT that’s really getting to know you well. With all that pre-processed information on demand, you’d think that you’d know what to do next. The answers are everywhere. So is strategic confusion.

Back in 1971, economist and cognitive scientist Herbert Simon wrote: “A wealth of information created a poverty of attention.” The man called it. And with that, welcome to the debut of the 4717’s Outwit newsletter.

The working memory of most humans can process about three to five “elements” at a time. So, of course, we developed a technology to deliver a near infinite stream of information against that biologically restrained capacity to process it. That fire-hose of information isn’t going anywhere, but the cognitive overload it creates will degrade your ability to make decisions on everything marketing strategies to where to eat lunch.

You’ve probably noticed the effects on your lunch plans, and are ignoring the other. The trick then, is to reframe the dilemma as one that you can control: a misallocation of resources. In this case, the resource your attention: It is a strategic asset - but a finite one. The second order effects sloppy allocation will get expensive.

Studies show that cognitive overload reduces decision speed as we subconsciously wait on that one more bit of data to deliver us from a crippling mental paralysis. The default work around is the quick, intuitive decision, which looks efficient but relies on mental short-cuts, called heuristics, to find a familiar pattern. It isn’t a bad way to handle a situation in a pinch, but it increases error rates and degrades judgement quality. It’s superficially rational, but a lot of mistakes are. The psychological attraction lies in that a decision, even a bad one, temporarily cuts off the overwhelming torrent of data for the feeling that something - anything - is getting done. This isn’t intelligence failure, but a normal system response to information overload and uncertainty.

So we upgrade to a disruptive technology that can solve the overload created by the last disruptive technology. But there are blind spots. You can address painful uncertainty with the AI model you’ve been monkeying around with but the certainty with which a conclusion is delivered may be more helpful in ending a debate than gaining clarity. AI is brilliant in many applications, but reducing the sheer volume of noise isn’t one of them - another 15 page report is a one-line prompt away.

Even with the AI assist, you may end up exasperating false correlations: AI is very good at quantifiable in-model relevance, but less so in wider-context, less quantifiable relevance. This brings us to what is called a signal and noise collapse. Herbert Simon’s work shows that humans aren’t very good at distinguishing high-value “signals” from low-value “noise.” A successful con-man instinctively knows that to persuade someone to make a bad decision, it helps to overwhelm a mark with low-value noise to bury critical signals. This makes false correlations appear more convincing as we search for a convincing pattern.

As for you and your team, the QED is that you will spend more time processing information than achieving a strategic insight.

So What?

Attention isn’t just scarce, it’s fragile – as anyone who has ever been interrupted in a state of “flow” will furiously attest. This can include anything from giving a presentation, writing a book or lining up a golf shot. Studies indicate that reviving an interrupted flow can take up to twenty minutes. If that focused thought needs to be abandoned, it takes about as much time to get fully immersed in the second task. The result, either way, is shallow, fragmented thinking instead of deeper analysis.

Understand that you aren’t managing information, but fragile, expensive attention. So how to structure focus?

The 4717 Field Guide:

AI does a superhuman job of crunching data, but if your ambitions extend beyond being a larger information system’s spare part, you need to develop the skill of information selection and interpretation. That skill starts with treating disciplined focus and attention as a strategic asset. To do that simply take the pressing question of the post-IT revolution market: “Do we have enough information?” And reframe it as “ What is the minimum amount of information needed to make a high-quality decision?”

It will be less than you’ve been conditioned to think, but more that the quick, intuitive decision maker would have you believe. Reframe the decision system by:

• Reduce inputs: Fewer dashboards, tighter briefs. Take that information and…

• Organize information for a clear “signal” visibility. Then…

• Allocating attention and focus with fewer interruptions and structured decision windows.

There is a reason it feels so good to shut down a fire-hose of information, it allows the brain to shift gears to process the situation in a way that cross-references the intuitive and hard to quantify with the purely rational. You might want to check your internal monologue the first couple of times you do it, it isn’t an easy sell. And yet it is the function, cost and reward for disciplined focus.

The resulting insight is clarity, and clarity is power.

Need to filter the noise from the signals? Let’s talk. Drop us a line at the 4717