Post 12: Maybe the Idea Isn’t the Discovery—Maybe It’s the Machine That Finds It

After all of this thinking, I expected to have a business idea. Instead, I ended up with something I find more interesting.

A method for looking.

We started with $100,000 and a goal of reaching $10 million. Then came scale.

Then excess.

Unused Wi-Fi. Tools. Boats. Warehouses. Solar electricity. Factories. Corporate reports. Walnut shells. Unexpected corporate combinations.

Every example seemed different. But gradually the same structure kept appearing.

The economy may contain hidden relationships  – One party has something another needs. But they do not recognize each other because the resource has the wrong name.

Or exists at the wrong time. Or is too fragmented. Or lacks permission. Or appears inside an unrelated industry. Or is recorded in a database nobody would think to combine with another database. Or is a capability hidden inside a company whose conventional business label obscures it.

The problem may not be scarcity. It may be representation.

The emerging discovery engine

Imagine representing companies, facilities, materials and infrastructure as:

properties, capabilities, inputs, output, unused capacity, recurring purchases, waste, streams, energy requirements, equipment, skills, certifications, data distribution, geography

Then ask the system to search for unusual relationships:

Who produces cheaply what someone else buys expensively? Who disposes of something another industry manufactures? Who owns capability another company is paying to build?

What byproduct can substitute for another company’s purchased input? What company has everything required for a valuable adjacent business except one missing capability? What two unrelated businesses become disproportionately valuable when combined? What should be happening in a corporation but is not?

Where is a resource already being used while its originator captures very little downstream value?

The moat cannot simply be the algorithm

A corporation may eventually reproduce the software. So the asset must accumulate. Historical observations. Entity relationships. Verified outcomes. Material-property mappings. Facility capabilities. Corrections. Failed hypotheses. Successful substitutions. Corporate behavior. Rights. Perhaps contracts. Perhaps proprietary measurements.

The idea is not to make the concept impossible to understand. The goal is to make the accumulated asset expensive and time-consuming to reproduce.

Now we need to test it

The next step is not to raise $25 million. It is not to build a giant platform. It is not to start contacting corporations.

The next step is a small discovery experiment.

Choose perhaps ten deliberately unrelated industries in one geographic area. Profile 50-100 businesses.

Identify what goes in, what comes out, what sits idle, what gets discarded, what is purchased repeatedly, and what capabilities exist. Then remove the industry labels.

Translate everything into functions and properties. Cross-match the results.

Most relationships should be useless.

Some will be obvious. A few may be interesting.

We are looking for one that makes us stop and say:

“There is no reason these two things should have anything to do with each other… but this actually makes sense.”

Then we investigate.

A grading system

Grade D: interesting coincidence.

Grade C: technically plausible relationship.

Grade B: meaningful economic value.

Grade A: large, repeatable national inefficiency.

Grade X: if true, this changes how an industry or corporation should think about a resource, capability, investment or competitor.

Grade X should create one reaction:

“If this is true, why isn’t somebody already doing this?”

The project is deliberately still pre-idea

That is not a failure.

It is actually the point.

I do not want to force an idea because I decided I needed one.

I want to build enough ways of seeing the economy differently that an opportunity eventually becomes difficult to ignore.

Maybe the eventual business is a material substitution.

Maybe it is corporate intelligence.

Maybe it is M&A.

Maybe it is a proprietary dataset.

Maybe it is something none of these posts has even mentioned.

For now, the question I am carrying into the next phase is:

What already exists in abundance, but is economically invisible because nobody has represented it in the right way—and what expensive problem becomes solvable once that hidden equivalence is recognized?

That is where Part 1 ends.

Part 2 begins when we start looking for the first real anomaly.

Post 11: What Happens When Two Ordinary Companies Create One Extraordinary Company?

At some point another possibility occurred to me.

What if Company A has X. Company B has Y. Neither company is particularly extraordinary.

But combining X and Y creates Z—a business or capability worth dramatically more than A and B separately. This is not the usual private-equity roll-up where one HVAC company buys another HVAC company to reduce overhead and gain scale. I am interested in combinations nobody would naturally put together.

1 + 1 should equal more than 2

Imagine:

Company A is worth $8 million.

Company B is worth $7 million.

But A+B could plausibly be worth $40 million.

Why?

Perhaps A removes B’s largest operating expense. B solves A’s utilization problem. Combined purchasing crosses a pricing threshold. A’s distribution gives B national reach. B’s data makes A’s operation more valuable. Or together they create a capability neither business possessed independently.

That additional value did not exist inside either company. It emerged from the combination. Maybe we stop searching for companies and instead, search for capability equations.

I think the traditional way of thinking may be to ask:

“Which companies should merge?”

But what if we instead, ask:

What combination of capabilities creates disproportionately greater value?

Therefore, the equation could be:

Capability A + Distribution F + Data Q + Certification Z = Strategic Outcome X

Then, once the equation is known, we can search for the cheapest way to assemble it. That may involve buying two companies. But it might also mean acquiring one company and licensing something from another. Or a minority investment. Or an exclusive contract. Or buying a particular asset rather than either company.

This may fit my preferred role

I would rather not become CEO of a 400-person operating company. But I could potentially be the person who discovers the combination.

The role could be:

originator sponsor data/IP owner deal architect equity holder

Experienced operators can run the businesses.

The intellectual leverage comes first. Financial leverage comes later.

AI changes the search space

Suppose there are tens of thousands of potentially acquirable businesses. The number of possible pairs is enormous. Humans cannot seriously evaluate every combination. But a structured capability database could potentially ask:

Which two apparently unrelated companies create the greatest nonlinear increase in enterprise value when combined? That is a fascinating computational problem.

The $100,000 does not need to buy the companies. It might build the system that discovers the $25 million transaction. Outside capital can enter after the opportunity has been identified and validated.

That is a much more plausible path from modest capital to substantial ownership than trying to make the original $100,000 itself compound 100 times.

Post 10: The Most Valuable Information May Be What Didn’t Happen

Financial analysis normally studies what happened.

Revenue increased.

Margins fell.

Inventory rose.

A factory opened.

Headcount grew.

Capital spending increased.

But there is another category of information that may be just as important:

What should have happened—but didn’t?

Imagine a manufacturer historically operates ten factories and produces ten million units.

It builds two additional factories.

Production falls to eight million units.

Why?

Maybe demand collapsed.

Maybe old facilities were replaced.

Maybe a supplier failed.

Maybe a product changed.

Maybe labor disappeared.

Maybe production moved overseas.

Maybe a fire, water problem, regulatory issue or equipment failure reduced output.

The point is not to guess.

The point is that the discrepancy itself contains information.

Corporate reports as telemetry

What if annual reports and financial statements are not merely financial documents?

What if they are telemetry from physical systems?

Capital expenditure tells us something about physical investment.

Inventory tells us something about production and demand.

Headcount tells us something about labor requirements.

Facility disclosures tell us something about capacity.

Margins tell us something about input costs and pricing.

Individually, these are ordinary financial variables.

Together, they describe a physical organism.

Add unrelated datasets

Now combine the corporate report with other signals.

Satellite imagery shows parking activity changing.

Job postings disappear around certain facilities.

Local permits show equipment removal.

Import records show a supplier stopped shipping a component.

Electricity consumption changes.

A patent suggests a new manufacturing process.

A competitor suddenly increases inventory.

A local news article mentions a water restriction.

No single signal proves anything.

But together they may explain why expected output failed to appear.

Negative-space data

I call this negative-space analysis.

The system builds an expectation:

Given everything we know, what should happen next?

Then it watches reality.

When:

Expected ≠ Observed

the discrepancy becomes a research target.

This could potentially reveal hidden idle capacity, supply-chain problems, strategic changes, emerging demand, operational constraints or future shortages before conventional statistics make them obvious.

Excess can become a sensor

This also changes how I think about excess.

Perhaps we do not need to sell excess at all.

Maybe changes in excess tell us something valuable.

A factory’s idle capacity falls sharply.

Warehouse availability tightens.

Freight capacity disappears.

Electricity consumption rises.

Hiring increases.

Permits appear.

Those changes collectively may say:

Something economically important is happening here.

Excess has become a sensor for the state of the physical economy.

That is a much stranger and potentially more valuable idea than simply finding someone to use an empty warehouse.

Post 9: What If We Have Been Classifying the Economy Wrong?

Most business information begins with labels.

Retailer.

Manufacturer.

Hotel.

Hospital.

Warehouse.

Airline.

Restaurant.

Data center.

Those labels are useful for humans, but they may also prevent us from seeing opportunities.

A grocery store, for example, is not merely a grocery store.

It is a bundle of capabilities:

– climate-controlled space,  refrigeration & freezer capacity, electrical service, water and sewer, roof area, parking, loading access, security, internet connectivity, employees, storage, customer traffic, delivery routes, waste, heat, commercial zoning, known location –

None of those capabilities intrinsically requires the building to sell groceries.

“Grocery store” is simply the economic configuration currently sitting on top of them.

Factories have the same problem

Imagine a factory that makes drills.

Calling it a “drill factory” tells us what comes out the front door.

It may tell us very little about what the factory is capable of doing.

The facility may possess:

precision machining injection molding, motor assembly, battery assembly, electronics testing, powder coatin,g packaging, quality-control, systems distribution access, skilled labor, certifications, electrical capacity

Perhaps another industry desperately needs some combination of those capabilities.

The opportunity becomes invisible when the database says:

Industry: Power Tools

Build a capability graph instead

What if companies were represented differently?

Instead of:

Company → Industry → Products

we build:

Company → Facilities → Equipment → Processes → Materials → Skills → Certifications → Inputs → Outputs → Excess Capacity → Distribution → Data

Now companies that appear unrelated may suddenly sit next to each other.

The system can search for functional similarities rather than industry similarities.

False scarcity

This could expose something I call false scarcity. An industry may say, “There is not enough capacity to make Component Q.” Maybe that is true if we search only factories already categorized as Component Q manufacturers. But perhaps 73 unrelated factories possess 80-90% of the necessary capabilities and could be converted economically.

The capacity existed.

It was classified incorrectly.

False abundance

The inverse is also possible.

There may be millions of roofs, warehouses, vehicles or parcels.

They appear abundant.

But perhaps only 20,000 possess a very specific combination of properties required by an emerging technology.

Once that combination matters, those 20,000 become scarce.

The opportunity is not the roof.

It is recognizing the hidden characteristic before the market does.

A different map of the economy

This project is increasingly becoming an attempt to construct a different representation of economic reality. Not what things are called. What they can do.

Not which industry owns them. What properties they possess.

Not what a company currently sells. What its accumulated infrastructure could enable.

If that representation is better than the conventional one, unusual opportunities should begin appearing naturally.

Post 8: A Walnut Shell Isn’t a Walnut Shell

This may be the point where the entire project changed for me.

Being in the oil & gas business myself, I am familiar with how oil and gas operations have historically used materials such as ground up walnut shells in certain drilling applications. The industrial detail is less important here than the conceptual leap I am trying to make.

A walnut processor might see:

agricultural byproduct

An engineer might see:

a particulate material with a particular combination of hardness, density, particle size, compressibility and other physical properties

Those are completely different descriptions of the same thing.

One tells us where the material came from.

The other tells us what it can do.

Names can hide value

Suppose Industry A produces material X and considers it low-value.

Industry B buys material Y for hundreds of dollars per ton.

Nobody connects them because X and Y have different names, originate in unrelated industries, and appear in completely different databases.

But what if X and Y share the functional properties that actually matter?

Perhaps after modest processing, X can replace some portion of Y.

Now we have:

one industry’s excess can be equal to or greater than another industry’s expensive input

That is far more interesting than just ordinary recycling.

So we can stop cataloging objects and Catalog properties instead.

This very well may be one of the central ideas of the project.

So rather than tell the system:

“This is a walnut shell.”

We could tell it:

particle size hardness density porosity compressibility absorption thermal characteristics chemical stability biodegradability abrasiveness cost per ton annual availability geographic concentration

Then do the same thing on the demand side.

And we would not tell the system:

“This is an oilfield product.”

Instead, we say:

“We need a material possessing properties A, B, C, D and E within these cost and performance limits.”

Now, with the power of classification and machine learning the computer can ask a question humans rarely ask:

What things already exist in enormous excess that possess A+B+C+D+E?

The answer might be walnut shells.

But the important possibility is that tomorrow the answer could be something nobody has ever tried.

Cross-domain equivalence

I call this cross-domain equivalence.

Two things that markets consider unrelated possess sufficiently similar functional properties to substitute for one another.

And the great thing is that this can extend far beyond materials.

A parked electric vehicle and a stationary battery both contain electrical storage. In fact, there have been stories of people charging their Teslas at off peak times, then using that stored energy during peak times to charge other devices. Essentially using their vehicle as a large battery backup. A big huge portable power pack.

Another example that I recently encountered involved a ski resort during summer hiking activities. I had taken a ski lift to hike at the top of a mountain. Upon returning to take the lift back down to the base of the mountain, I was notified that there was lightning in the area and the lift was not going to be operating, however, they were sending a shuttle to pick myself and several other “passengers” up to take us back down the mountain. While I was initially expecting an ordinary “shuttle”, I was surprised to see a security guard show up with his security truck, filled with tools and equipment only for us to load up on our return journey. This was no ordinary shuttle but yet a security vehicle, excess capacity for security operations during the day, repurposed into a makeshift passenger shuttle. While this seems obvious now, the added expense of a separate shuttle operations for the rare times lightning strikes occur is impractical. These types of situations happen every day. It is in these unconventional areas that I am more interested in.

A delivery truck and another fleet traveling the same geography both contain transportation capability.

A warehouse and an underused commercial building both contain protected volume.

A phone and a dedicated sensor both contain sensing capabilities.

The labels are different.

The functions overlap.

The real invention may be the finder

If we discover one surprising substitution, that may be commercially valuable.

But something bigger is possible.

What if we can create a system that repeatedly discovers these relationships?

That would not be a marketplace for agricultural waste.

It would be a discovery engine for hidden economic equivalencies.

The walnut shell would merely be proof that the world contains relationships conventional categories prevent us from seeing.

Post 7: When Your Excess Is Valuable—Just Not to You

Not too long ago, I purchased a home that was equipped with Solar panels. It was a Tesla 12.8kW system that had an annual production of around 20,000 kWh at the time it was inspected at purchase.

When I consumed electricity, I paid normal rates. When the house produced excess electricity and exported it to the grid, the compensation could be tiny by comparison. The exact economics vary by utility, location and rate structure, but the observation was what interested me.

The excess was not idle. Someone was using it.

Yet the excess credited was much different than the excess produced. In both cases, there was excess capacity and excess demand simultaneously. Unfortunately, the power companies have found a way to exploit this. They simple charge more for excess demand than they credit for excess capacity. This is a highly regulated industry and shows the market driven solution to obvious excess on either side.

That is different from an unused screwdriver or empty warehouse. I call it value-stranded excess.

Used inefficiently may be more interesting than unused

Many excess opportunities require creating demand.

Value-stranded excess is different because demand may already exist. Supply exists. Infrastructure exists. The transaction exists.

The interesting question is why the value is allocated the way it is.

An originator may lack scale, storage, distribution, aggregation, information, timing flexibility, contractual rights, standardization, predictability or negotiating power.

Map the value, not just the waste

This suggests value leakage mapping.

Follow a resource through the economy. Who creates it? Who handles it? Who transforms it? Who ultimately consumes it? Where does its economic value increase dramatically? Who captures that increase?

A manufacturer may pay to dispose of a byproduct another company purchases as an input. A truck returns empty while another shipper pays for transportation. A business generates data incidentally while someone downstream aggregates similar data and sells expensive intelligence. A facility pays to remove heat while another process pays to create heat.

The opportunity may not be physically buying and reselling the resource.

It may be discovering where enormous amounts of value are already being created but captured somewhere other than where the underlying capability originates.

Post 6: The Warehouse That Was Secretly Another Business

Years ago I worked with a friend who had a successful online business. He was in the e-commerce industry and sold products online. He shipped thousands of packages every day, yet his warehouse was using only roughly 25-30% of its available space.

I suggested offering fulfillment services to smaller, non-competing online retailers.

His employees were already receiving inventory, they were walking aisles, picking orders, packing boxes and preparing shipments 5-6 days a week. Third-party inventory could occupy unused space and flow through much of the same infrastructure with very little impact to his operation. In fact, his existing operation would have greatly benefitted.

New revenue could reduce old costs

While he did a significant shipping of his own products, addition shipping volume with UPS, FED-EX, USPS could have likely improved carrier rates. He could increase his bulk purchasing of packing material, boxes and tape to reduce supply costs. Essentially, greater warehouse utilization could lower effective cost per shipment on each and every item he sold. Labor could become more productive by redirecting staffing and eliminating any “down time”. With enough scale, he may have even been able to justify integrating producing packaging, inserts and other materials internally for additional revenue.

New customers would have likely reduced operating expenses while providing an additional and diversified income source. This became evident one day when the servers were down and no orders were being processed. Staff in the warehouse was idle while had there been fulfillments from partnering companies, the losses, if any, may have been mitigated a bit.

But, the warehouse was not the important discovery

The owner thought he had an online retail business.

But he had accidentally assembled receiving, storage, inventory systems, employees, picking, packing, carrier relationships, shipping volume, purchasing power and loading infrastructure.

He had accidentally built much of another business.

That produced one of the most useful questions in this project:

What companies have accidentally built businesses they do not know they are in?

A business builds infrastructure for purpose A. But A requires capabilities B, C, D, E and F. Perhaps C + E + F can perform an entirely different economic function.

This is emergent capability.

The warehouse story matters because it demonstrates the difference between asking:

What does this company do?

and

What could this company do with what it already has?

Post 5: The Strange Economics of Excess Capacity

Excess capacity is different from excess inventory.

If a factory has 1,000 unused bolts, the bolts can sit on a shelf. If a machine could have operated one additional hour yesterday but did not, that hour is gone forever.

Capacity is often perishable.

Capacity is purchased for peaks

Imagine 1,000 businesses each needing up to 100 units of some capability.

If every business independently prepares for maximum demand, society builds 100,000 units of capacity. But suppose their combined demand has never exceeded 30,000 units because their peaks occur at different times.

Seventy thousand units of theoretical capacity exist because everyone individually needed certainty.

No participant behaved irrationally. The inefficiency appears only when the system is viewed collectively.

Unreliable excess can be economically invisible

Suppose a corporation needs 100 units of reliable capacity. Six businesses have 8, 17, 4, 21, 12 and 38 spare units.

Mathematically, that equals 100 (I actually just double checked the math to be sure). Economically, it may equal zero.

The corporation does not want six contracts, integrations, compliance reviews and uncertain suppliers.

This suggests a transformation:

Excess → Identified → Classified → Permissioned → Standardized → Aggregated → Predictable → Reliable → Tradable

Information and coordination can transform the resource without changing it physically.

Permission may be scarcer than capacity

A resource can exist but remain unusable because of contracts, privacy, security, regulation, liability or technical incompatibility.

Sometimes the scarce commodity may be permissioned capacity: excess that has been identified, standardized and made legally and operationally usable.

Aggregation can also manufacture reliability. One source may be unpredictable; thousands of partially independent sources can become statistically predictable.

The opportunity may eventually be turning fragmented excess into institutional-grade supply—or identifying where that transformation is possible before anyone else does.

Post 4: Why Do We Own So Much Stuff We Almost Never Use?

I have more than 30 screwdrivers, multiple hammers, drills, saws and other power tools. Some may go two or three years without being used. Some tools I bought for one project and one intended purpose yet it is still in the garage, unused and occupying its place on a shelf.

Millions of homeowners are in the same situation.

Why don’t we sell them? Because keeping them can be perfectly rational!

The main reason is the replacement value versus the liquidation value.

Suppose a saw costs $150 to replace but might sell used for $40. To get the $40 I need to take a picture of it, making sure I get the best lighting possible (my wife has become a master of lighting and insists that if you are going to take a photo to sell something, it should be the best it can be), list it on Craigslist, Facebook Marketplace, OfferUp etc, deal with multiple “Is this still available?” responses from “buyers”, arrange pickup at my residence or plan to meet someone at Home Depot at a particular time only for them to show up 20-30 minutes later after they have already messaged me that they are 5 minutes away. Of course there is shipping, packaging and all that comes with that side of the transaction as it meeting in person is too much trouble. Only to accept that in 6 months, I may spend $150 replacing it because my wife decided she wants a few inches taken off of her office ottoman legs.

So I keep it.

The saw is useful, valuable and almost completely idle. The economics of transferring it are worse than letting it sit.

That is a friction-stranded asset.

I am not really buying a saw

When I buy a saw, I am buying the ability to cut wood whenever I need to.

The saw is simply the mechanism guaranteeing that capability.

The same applies to ladders, generators, trailers, pressure washers, boats, RVs and aircraft. Owners tolerate low utilization because immediate availability has value.

Businesses do the same thing with excess equipment, warehouse space, backup generators, safety inventory, servers, suppliers and machinery sized for peak demand.

This raises a larger question: How much excess exists because ownership is currently the easiest mechanism for guaranteeing future access?

This is not a tool-rental idea

As previously mentioned, a neighborhood tool marketplace is exactly the kind of operational business I do not want to build.

The screwdriver is a clue, not the product.

It reveals a structural pattern:

valuable asset + low utilization + high replacement value + low liquidation value + transaction friction + demand for guaranteed availability

The interesting question is where this structure exists when the assets are worth $100,000 or $10 million.

I am looking for the industrial equivalent of the screwdriver problem.

Post 3: There Is Too Much of Almost Everything

For post 3, I am continuing from the closing of Post 1 in this series where I am further expanding on abundance, surplus and opportunities. These posts are meant to be short, concise and will provide the initial foundation and layout for the concepts we will explore later on in the series. It may be helpful to understand the basis by dipping your toes in the water before diving head first into the deep end.

Look around your house. How many cables, clothes, dishes, tools and electronics do you own?

Now zoom out to every house, apartment, hotel, restaurant, warehouse, store, office and factory.

We live in an astonishing world of abundance.

Wi-Fi exists almost everywhere. Computers sit idle. Phones sit on nightstands. Cars are parked for hours. Tools may not be touched for years. Clothing goes unworn. Food is discarded. Warehouses are partially empty. Factories could often produce more. Electricity can be valuable at one time and nearly worthless at another.

The first instinct is to call this waste. I think that is too simple.

Excess has different forms

Some excess can be stored. – Lumber unused today can be used next year.

Some cannot. Unused bandwidth at 2:13 a.m. cannot be put in a warehouse and used next Thursday. That makes it perishable capacity.

The same applies to an empty airline seat, machine hour, compute cycle, freight capacity, restaurant table or electricity at a particular moment.

Other excess is intentional. Companies maintain backup equipment they hope never to use. A homeowner may own a generator used once every five years. A factory keeps spare parts because shortage could shut production down.

That excess purchases certainty.

Redundant abundance

Imagine an apartment building where hundreds of households independently purchase internet plans capable of far more bandwidth than they normally consume. Hundreds of routers may sit only feet apart and while yes, they are all connected to the internet, they are disconnected from each other in the sense that each user is essentially paying for excess.

Each household rationally purchased enough capacity for its own peak. Collectively, however, the building contains enormous redundant capacity.

Everyone prepares for their own worst case, but everyone’s worst case does not happen simultaneously.

The resource may not be the excess

It is tempting to turn every observation into a marketplace: unused tools, empty rooms, spare bandwidth.

Those may in fact turn into businesses in the future such as Uber or VRBO, but that is not my goal with this project.

The more interesting questions are why the excess exists, why it persists, what information is missing, what permission is missing, what standard is missing, and what aggregation threshold is missing.

Maybe excess itself is not the asset. Maybe it is evidence pointing toward the real asset.

In Post 4, we will take a look into the question as to why we own so much stuff and why we almost never use it.