AI is about to make that painfully obvious.

It won’t turn everyone into a great builder.
It will separate those who know what to build from everyone else.

This isn’t just happening in software. Software is just where you can see it first.

For a long time, generalists won for a simple reason. Building was expensive.

If you were going to invest in software, services, or anything complex, you had to spread that cost across as many customers as possible. One product. Many use cases.

Everyone compromised a little. That was the trade.

I learned this early, in the first few years building my first company.

I sat down with two great brand strategists and walked them through everything we were going to do. Full service. All industries. Big vision.

They stopped me.

“No customer cares about that.”

I thought breadth was impressive. Clients wanted relevance.

So we cut. We focused only on ecommerce, then narrowed further to one platform, and eventually to marketing, not development.

At the time, it felt like we were shrinking the business. In reality, we were making it the best option for a specific audience.

Then something interesting happened. The more specialized we became, the more partners we had wanting to refer us business. We stopped competing with everyone and started becoming useful to them.

Once we focused, we started to see things others didn’t. The problems weren’t abstract anymore. They were specific to that platform, that client profile, that workflow.

So we built software around it. Not broad platforms, and not custom tools for each client, but systems behind the scenes that helped us serve that audience better.

We could move faster, make better decisions, and avoid predictable mistakes. Everything was shaped around what that niche actually cared about.

The value wasn’t in the software itself. It was in how precisely it helped us deliver for that user.

We bootstrapped from $0 to $25M in ARR this way, without outside capital, compounding at roughly 25% a year, 11 of 12 years.

It kept proving itself.

Back then, we had to specialize because building was expensive. We couldn’t afford to be unfocused.

Now AI removes that constraint.

Anyone can build.

That’s not the advantage anymore.

The advantage is knowing what to build, for whom, and just as importantly, what not to build.

The market won’t converge on one tool that does everything.

It will fragment into smaller, sharper products, each built for a narrow audience with a specific need.

We’re already seeing this.

The winners will be the ones who go deeper, not broader. That could be small upstarts, or incumbents that rethink how they build.

Relevance is what people love.

And once they feel that, everything else starts to feel generic.

Share

Subscribe now

This looks like an ecommerce story. It is. But the implications extend well beyond ecommerce.

As agents take on more discovery and execution, markets will slowly shift away from persuading people in the moment and toward being easy for machines to evaluate and trust. Brand still matters a lot, but agents are increasingly becoming an important customer alongside humans.

In ecommerce, the implications are fairly obvious:

• Product data matters more than ever

• Price, availability, and fulfillment reliability drive selection

• If an agent can’t understand or trust you, you won’t consistently show up

This won’t change everything overnight. Humans are still very much in the loop. But directionally, the companies that do well may start to look boring in a good way. Clean data. Strong operations. A consistently good experience.

This won’t stop at retail

Anywhere an AI agent can evaluate options and take action, the same logic applies. SaaS procurement. Travel booking. Insurance quoting. Hiring marketplaces. Even parts of professional services.

In all of these, the “buyer” increasingly looks like:

• A system optimizing for objective certainty over subjective persuasion

• Inputs that can be compared cleanly

• Outcomes that can be verified

A little less storytelling. A lot more proof. Less friction.

This won’t replace human judgment. Agents will make the first cut, aggressively filtering what humans see. The importance of that filter will grow over time.

Former Exclusive Concepts team members know we were preparing for this shift years ago. Exclusive now goes to market as Go Fish Digital, led by a team that has helped more ecommerce businesses scale than just about anyone.

Your new biggest customer may not be human. The companies that adapt early will benefit.

What does it mean to adapt?

It starts with getting your data house in order.

• For ecommerce companies, it means clean, structured, rich product data. Accurate inventory. Clear pricing rules. Price surveillance tied to share of voice, like what GrowByData supports, aligned to a well-defined profit optimization strategy. Reliable shipping and returns. What the agent sees has to match what happens.

• For SaaS and B2B vendors, it means packaging and pricing are explicit. Features, limits, integrations, security posture, and contract terms are easy to evaluate. No “talk to sales” ambiguity when a machine is trying to compare options.

• For travel, insurance, and marketplaces, it means availability, constraints, exclusions, and outcomes are clearly defined. What’s covered. What isn’t. When something applies and when it doesn’t.

• For professional services, it means scopes, typical timelines, success criteria, and past outcomes are clear. Less storytelling. More evidence.

Subscribe now

Across all industries the principle is the same.

Anything a human currently figures out through back and forth needs to be upfront and obvious to a machine.

That’s how agents decide who gets considered and who fails the first cut.

This morning Jensen Huang explained why that last 10% is about to matter a lot more. He was on CNBC at the Synopsys AI Engineering Conference talking about NVIDIA’s new $2B stake in Synopsys, and the conversation quickly shifted to a bigger point. Consumer AI was built on “close enough.” Enterprise AI cannot be.

Here is what Jensen said

“The important concept here is accelerated computing and AI is revolutionizing every single industry. Of course it started in the consumer segment, because getting every answer 90% right is a pretty magical thing.

But in the industrial segment, that 10% you don’t get right becomes mission critical. And so it takes longer for us to create the tools, create the technology necessary, for the industrial sectors, the enterprise segments, to be able to adopt this technology. But the time is now here.

[…] During 2016 the world’s scientific supercomputers was 90% CPUs and 10% GPUs. This year it’s 90% GPUs and 10% CPUs. The platform shift has happened.

What Synopsys does in engineering is founded on principled physics. We are now making this shift in this industry. It is one of the most computing intensive industries in the world, and we are elevating it with AI.”

This matters for a few groups

1. Consumers. Your AI tools will get more reliable as enterprise standards force accuracy higher. Better grounding. Better context. Less drift.

2. Operators and founders. The last 10% is where real enterprise value is. This is the shift from fun demos to workflows that actually run the business.

3. Enterprise buyers. Once the hardest scientific workloads flip to GPUs, the rest of the economy follows. You are stepping into a new compute baseline whether you planned for it or not.

4. Investors. The consumer phase created demand. The enterprise phase creates margin. Precision AI is where durable economics show up.

5. Engineers and product teams. Everything from chip design to supply chain planning is about to compress cycles and rebuild itself around accelerated computing.

How I’m looking at this

The remaining 10% is where the risk and opportunity sits. It is where regulation lives. It is where trust either forms or breaks. It is also the part “There Is No Magic Bullet” calls out, the part that never gets solved by one breakthrough.

It also ties back to what I wrote in “Billions of Small Minds” (here). The real advantage goes to leaders who know where their business sits on the adoption curve. Not the hype curve. The adoption curve. Some companies can live with “close enough” for a while. Others can’t. Some markets move fast. Others punish mistakes. Timing matters more than enthusiasm.

Seen this way, the 10% stops looking small. It becomes the real frontier. Planes, hospitals, factories, and banks operate there. Almost right is still wrong in those environments.

If 90% is good enough for your business, move fast and experiment (my Speed to Value Matrix can help).

For most companies it is not. Not without guardrails. Not without training. Not without people who know how to use these tools with discipline. Knowing where you sit on that curve is becoming a leadership skill.

Scott

Share

He meant that the real world is never binary. A sensor can tell you if something is hot or cold. But intelligence, human or artificial, needs to understand context. The weather, the surface, the motion, the pattern. What is happening around the thing, not just to it.

That simple point hit me as profound. Because it is the same trap many leaders fall into when deciding what to build or where to focus. They think in if/then statements. If the market changes, then we pivot. If the technology works, then we scale.

But the world does not cooperate with conditional logic. The variables are too many, the timing too messy.

What this founder understood, maybe without saying it outright, is that clarity comes from seeing what is truly happening now. Focusing where the need is real and measurable. That is how you earn the right to build what comes next.

It reminded me of lessons I learned years ago about timing, complexity, and when to go all in.

In 2007 my company began running conversion tests for eCommerce sites. Early A/B and multivariate testing. I was convinced it would change everything.

We could measure what made customers buy, refine design choices in real time, and build a flywheel of growth that competitors could not copy. The logic was perfect. The results were tangible and incredible.

And still, almost nobody cared.

Some did not trust it. Others found it too complex. Many preferred intuition. Most preferred to simply copy what others were doing. We were right, but early.

A few years later we built price optimization systems that used real-time data to find the profit sweet spot. Again, the math worked. But organizations were not ready. Pricing touched supply chains, partners, brand perception, and too much friction followed.

Those experiences taught me that technology rarely fails on capability. It fails on timing.

Markets move when systems are ready to move. Technical readiness, organizational readiness, emotional readiness.

Today everyone is talking about AI

Markets have priced the S&P 500 as if productivity transformation is already here. But the numbers tell a different story.

Only 3.8 percent of U.S. firms currently use AI to produce goods or services, according to the U.S. Census Bureau. A 2024 BCG study found that 74 percent of companies have yet to show tangible value from their AI investments.

Even among leaders, most of the thinking still happens in pilot programs and slide decks, not in the real economy.

Meanwhile, infrastructure is still being built. McKinsey estimates that 70 percent of global data-center demand by 2030 will support advanced AI workloads, a number that shows how much plumbing remains unfinished.

Every technology curve looks flat for a long time. Then, suddenly, it bends upward.

Most leaders misread this curve. They either rush in too early, burning resources before the system is ready, or wait too long and miss the compounding when it starts.

The best operators know the difference. They look for alignment between readiness and return. They invest when the signal is clear enough to move the system, not just their curiosity.

Meta’s “mobile first” move came at exactly the right moment. Amazon’s patience with AWS was the opposite side of the same skill, knowing when the world is about to catch up to your conviction.

AI’s curve will bend, but it will not bend the same way for everyone.

In some industries the use cases are already obvious. Being able to see into buried pipes with context, for example, where basic cameras fail because the footage once required human analysis. That capability now exists in the field, creating real operational value.

In many other areas, the return on AI still depends on experimentation, infrastructure, and organizational change. The signal is weaker and the cost of chasing it too early is high.

Leaders cannot rely on a single market curve to guide them. They have to localize it to their own business and environment. The question is not whether AI transforms the economy, but whether the conditions around their specific use case are ready for it.

That decision is a down payment on validated learning. Go slow enough to learn. Move fast when the learning is clear.

In a world that rewards fast learning, the real constraint is not effort. It is speed of impact.

A simple way to use this in planning is to rank initiatives by three variables…

• Impact measures the potential business result, such as revenue, margin, or risk reduction.

• Speed measures how quickly the impact becomes real.

• Scope measures how complex or disruptive it will be.

Score each initiative on a 1–10 scale, debate the rankings with your team, and decide which opportunities are ready for meaningful investment and which should remain in exploration mode.

When you do this, you stop chasing the broad AI curve and start shaping your own pace of impact.

If you want a simple tool to put this into practice, I built the Speed-to-Value Matrix. It’s a clear way to see which ideas create value fast and which can wait.

View the Speed-to-Value Matrix

Scott

PS: Notes from the research

• 74 percent of companies still struggle to achieve measurable value from AI (Boston Consulting Group, AI Adoption in 2024).

• Only 3.8 percent of U.S. firms currently use AI to produce goods or services (U.S. Census Bureau, Business Trends & Outlook Survey).

• 70 percent of global data-center demand by 2030 will support advanced AI workloads (McKinsey, AI Power and Infrastructure Report, 2024).

• 75 percent of enterprise data is expected to be created and processed outside traditional data centers by 2025 (Gartner, via Otava, 2024).

• The Edge AI market is projected to exceed $160 billion by 2030, growing about 24 percent annually (NGP Capital, Edge AI Market Outlook, 2024).

Most teams try to do too much.
Not because they lack focus, but because everything sounds important when you say it out loud.

The truth is simple.
Some ideas create value fast.
Some take a long time.
Some never pay off.

The hard part is not generating ideas.
It is the alignment.
Getting everyone to agree on what to do first is where things break down.

Most planning tools compare impact to effort.
They miss the more important question: how fast the impact will arrive.

Effort is not the real constraint.
Speed of impact is.

Why speed is the real constraint

Speed matters because the world does not wait.

Markets shift faster than planning cycles.
AI lowers barriers to entry.
Customers change expectations quickly.

In that kind of environment, the constraint is not how hard something is.
The constraint is how long it takes to learn whether it works.

Fast learning creates momentum. Slow learning kills it.

Effort matters, but speed matters more.
Speed of impact is what keeps you aligned with reality.
It determines whether you compound learning or fall behind.

A small idea that proves value fast beats a big idea that reveals nothing for six months.
You can always scale something that works.
You cannot fix something that teaches you nothing.

The core idea

Judge every idea by three things.
How much it matters.
How fast it creates value.
How hard it is to do.

Impact.
Speed.
Scope.

These are not equal.
Speed is the governing constraint.
Impact is the reason to move.
Scope shapes how you roll it out.

The matrix

Below is the Speed-to-Value Matrix.

Impact on the vertical axis.
Speed of impact on the horizontal axis.

Deploy Now

High Impact.
Fast Speed.

Invest and Sequence

High Impact.
Slow Speed.

Experiment

Low Impact.
Fast Speed.

Defer

Low Impact.
Slow Speed.

Scope, the third dimension

Scope is how many people and systems need to change.
That is all you need to know.

Scope does not move ideas between quadrants.
Impact and Speed already do that.
Scope tells you how to handle an idea once it is in the right place.

Scope helps in four practical ways.

1. It prioritizes ideas inside a quadrant

Low Scope moves early.
High Scope needs more preparation.

2. It simplifies big ideas

If an idea is High Impact and Fast Speed but High Scope, shrink it.
Find the version that creates fast value with less disruption.

3. It prevents overload

Too many High Scope initiatives will stall a team.
Scope makes the hidden load visible.

4. It breaks ties

When two ideas have similar Impact and Speed, Scope decides priority.

Simple guidance:
Low Scope moves fast
Medium Scope needs sequencing
High Scope needs simplification and senior ownership

How to use the framework

1. List the ideas.
Ten to thirty is plenty.

2. Score each idea.
Impact. Speed. Scope.
Use a simple one to ten scale.

3. Plot by Impact and Speed.
Place ideas roughly. Precision is not the goal.

4. Use Scope to set intensity.
Low Scope moves fast.
Medium Scope needs a plan.
High Scope moves only after simplification and sponsorship.

5. Turn the map into a plan.
Deploy Now becomes near term priorities.
Invest and Sequence becomes mid term bets.
Experiment becomes time boxed tests.
Defer becomes not now.

An example

Below is a practical example using AI and operations initiatives.
Each numbered dot marks one idea and is color coded by Scope.

Green is Low Scope.
Yellow is Medium Scope.
Red is High Scope.

High Impact and Fast Speed goes first.
High Impact and Slow Speed gets a plan.
Low Impact and Fast Speed gets a small test.
Low Impact and Slow Speed gets dropped.
This is why the matrix works. It makes the sequence obvious.

Why this works

It forces focus.
It cuts through wishful thinking.
It stops teams from treating everything as a priority.
It turns noise into a clear plan.
It aligns the team around learning speed.
It helps you move quickly without burning people out.
The Speed-to-Value Framework focuses teams on work that creates value fast.
It is a simple system for a world that moves faster every day.

Scott

When you look closely, the cause is almost always the same.
No shared framework for what matters most.
No clarity on what the company values when things get hard.

Without that clarity, every question floats up the chain.
The executive becomes the bottleneck.
Action stalls.
Culture erodes.

It looks like a business problem, but it’s really an integrity problem.
The system lacks internal consistency.

When values are clear and lived, people know how to act without waiting for permission.
Speed happens naturally.
Leaders stop managing by exception and start leading by principle.

In 2012, I learned this lesson the hard way.

A large part of our business relied on understanding Google’s search algorithm. Overnight, Google changed it. What had been a strong, growing company suddenly lost its footing.

Revenue pressure grew. Clients were anxious. Payroll was looming. We just bought our first house. My first kid was on its way. And for the first time, I didn’t know what to do.

I took a day off and went to the Boston Public Library. I sat in the courtyard with a blank sheet of paper and drew a line down the middle.

On the left, I listed what I actually valued as a business leader.
On the right, I wrote what my actions suggested I valued.

The lists didn’t match.

What I wanted: to help people and businesses truly grow, to build something evergreen, to work with data and truth, to avoid easy answers.

What I was doing: chasing short-term results, optimizing around the algorithm, building a business that could be upended by a single change.

That afternoon changed everything.

I committed that every action in the company had to align with the values on that left-hand column. It wasn’t about words. It was about design.

That decision became the foundation for how we rebuilt. Instead of reporting tools, we built a data platform that could plug into any application, surface insights, and guide decisions across channels. It took longer (especially because we scrapped v1), but it fueled a decade of fast growth.

I didn’t want to raise outside capital and risk losing control of that vision, so we built an operation overseas. Not as a cost center, but as a modern workplace where people could thrive and give back to their community. A true win / win (one of our core values).

The company not only survived Google’s shift. It grew to become one of the largest independent agencies specialized in ecommerce.

All of that came from one afternoon of getting clear on what I valued and aligning my actions to it.

When leaders act out of alignment, it looks like they’re moving fast, always busy, always in motion, but they’re not building leverage. Every decision is ad hoc. Every debate repeats itself.

The company becomes a group of people playing different games.
Some value growth.
Some value cash in hand.
Some value prestige.
Some value control.

They can’t move together because they aren’t anchored to the same framework.

When values are real, speed follows.
People trust each other.
They understand the trade-offs.
They don’t need constant supervision or reassurance.

Years later, when COVID first hit, those same principles guided how I led.

Our eCommerce clients started pausing services. I’d watch a news alert about a state shutting down, and then see the cancellations come in. It was unsettling.

My instinct wasn’t to hide it. It was to tell the truth.

Every morning, I posted the prior day’s numbers to the entire company... how many clients paused, how much revenue we lost. It was uncomfortable. It made me nervous.

But it was consistent with one of our core values: being honest.

A client and friend once told me, “Action conquers fear.”

Sharing those numbers was my action.

People didn’t panic. They rallied.
They trusted the information because they saw both the good and the bad.
And as the numbers improved, the trust deepened.

Subscribe now

The groundwork of defining values and aligning behavior to them may feel academic. It’s the work people want to skip when they’re busy. But it’s the job of the architect.

A business without that foundation might still grow for a while. But it will always work harder than it should, and it will crack under stress.

Early founders often think they’re too small to bother.

Established companies think they’re too busy to reset.

They’re both wrong.

The time to define your values, and align your actions to them, is always now.

Scott

Share

PS:

If you’re looking to go deeper, read Robert Glazer’s new book, The Compass Within, an easy and important read about leadership, integrity, and personal alignment.

And learn more about Verne Harnish’s Scaling Up, integrating values into operational systems that sustain growth.

It’s a framework for better decision-making, built from the overlap of economics, behavior, and experience. Munger called it a “latticework of mental models.”

He laid out 24 causes of human misjudgment that explain why even capable people make poor decisions. Such as...

1. Psychological denial: Painful facts get edited.

2. Consistency and commitment bias: Once stated, a view hardens.

3. Social proof: The crowd feels right even when it isn’t.

4. Deprival super-reaction: Loss hits harder than gain.

5. Authority bias: Titles can steer teams into bad choices.

6. Liking and disliking biases: We trust those we like and tune out those we don’t.

His most important point was how they combine. When several act together, judgment collapses, and success becomes increasingly unlikely.

To guard against that, he used tools like “inversion,” asking not “How do I win?” but “What would guarantee failure?”

He also forced himself to revisit mistakes, no matter how uncomfortable, and looked at problems from multiple disciplines to break out of narrow thinking.

Munger’s goal was to see reality more clearly and act with fewer blind spots.

You can read it here

Leave a comment