The Great Clearing · Article 5 · Pick a Side of the Chasm
Automotive has spent a lot of years explaining why buying a car is different.
It is more complicated than buying a television. Financing is regulated. Trades have to be appraised. Inventory is physical. Titles matter. Taxes vary. Incentives change. Credit matters. The customer may be upside down. A vehicle can cost more than some people’s first house did.
All true.
Unfortunately, none of it requires the customer to enjoy unnecessary friction.
That distinction is becoming strategically important because the companies applying pressure to automotive retail are not necessarily trying to prove the dealership is obsolete. They are identifying places where the traditional experience normalized inconvenience, uncertainty, repetition, or opacity and asking whether those gaps have economic value.
Increasingly, the answer is yes.
Carvana is monetizing convenience and continuity. Amazon is bringing familiar commerce expectations into automotive while still relying on participating dealers for inventory and the physical relationship. Google is building infrastructure designed to allow intelligent systems to move customers from discovery toward increasingly consequential commercial actions. AI is giving consumers new ways to research, compare, understand, and eventually act.
None of these developments guarantees the death of the dealership.
They do something more useful.
They expose where the customer experience has been vulnerable all along.
A customer-experience chasm is opening. On one side are businesses using intelligence to redesign the experience they intend to deliver. On the other are businesses using AI to automate the experience they inherited.
Those two strategies may use similar technology.
They will not produce the same future.
One of the most dangerous phrases in any established industry is that’s just how it works.
Sometimes it is true. Certain constraints are real. A lender has requirements. A state has regulations. Physical assets need inspection. Risk has to be managed. Complex transactions create legitimate complexity.
But mature industries also become very good at losing track of the difference between complexity that is inherent to the transaction and friction that accumulated around the way the industry chose to manage it.
The customer rarely makes that distinction for us.
They simply experience the result.
They experience whether the information online survives the arrival at the store. Whether the vehicle shown as available is actually available. Whether a price can be understood without decoding six disclaimers. Whether the work completed before the visit still counts once they walk through the door. Whether the person greeting them knows what has already happened. Whether the service department remembers the vehicle. Whether the next channel feels like a continuation of the relationship or an entirely new company.
Cox Automotive’s 2026 Car Buyer Journey research points directly at this expectation. Twenty-five percent of new-vehicle buyers in the study used AI during the shopping process, while 59% of respondents who used AI-powered assistance reported high satisfaction with that assistance. Among mostly digital buyers using AI assistants, 84% reported high satisfaction with the overall buying process.
The interesting signal is not simply that customers are using AI.
It is why increasingly digital experiences work when they work.
Cox’s broader analysis emphasizes efficiency, confidence, continuity, and flow. Buyers completing more steps digitally tend to spend less time in-store and report higher satisfaction. When progress carries forward, the experience feels coherent. When information has to be re-entered, numbers need to be reconstructed, or completed steps are discarded during a handoff, the customer immediately feels the gap.
The customer is not grading your technology stack. They are grading whether dealing with your business feels unnecessarily difficult.
This is the opportunity outsiders keep seeing.
Not the absence of dealerships.
The presence of friction.
Automotive has spent years debating Carvana from almost every possible angle.
The business model was impossible. Then the valuation was impossible. Then the debt was impossible. Then profitability was impossible. Then somebody discovered another vending machine and the conversation started over.
There are legitimate questions to ask about any large retailer’s economics, operating model, valuation, and long-term competitive position. But there is one increasingly difficult fact to debate: a meaningful number of customers are willing to buy vehicles through the experience Carvana has created.
In the second quarter of 2026, Carvana reported 197,325 retail units sold, up 38% year over year, and $7.376 billion in quarterly revenue, up 52%.
Those numbers do not prove that every element of Carvana’s experience caused its growth. They do make it harder to dismiss the operating model as an irrelevant edge case.
What Carvana understood early was that there is economic value in reducing uncertainty around a process customers historically experienced as fragmented.
One environment.
One representation of the inventory.
A clearer sense of progression.
Fewer moments when the customer is expected to understand the internal organizational structure of the retailer simply to continue buying a car.
Traditional dealerships can point to plenty of advantages Carvana does not possess in the same way. Local presence. Immediate human expertise. OEM relationships. Physical inventory. Service infrastructure. Community reputation. Long-term ownership relationships. Parts. Technicians. Brand expertise. The ability to look a customer in the eye when the situation becomes complicated.
Those are enormous assets.
But the customer only receives value from an advantage they can experience.
A structural advantage trapped behind operational friction is still an advantage on the balance sheet. It may not be an advantage in the customer’s journey.
That is the warning Carvana represents.
Not that every dealership should become Carvana.
That every dealership should understand why a customer might prefer an experience designed from a blank sheet of paper over one assembled through forty years of accumulated handoffs.
Amazon Autos creates a different kind of pressure because Amazon is not trying to persuade customers to learn an unfamiliar shopping behavior.
It is bringing an extremely familiar shopping behavior into an unfamiliar category.
Through Amazon Autos, customers can browse participating dealership inventory, examine vehicle information, understand transparent pricing, explore financing, purchase eligible vehicles online, and ultimately interact with the participating local dealer. Amazon has expanded the program beyond new vehicles into used and certified pre-owned inventory, and its dealer network now spans more than 130 U.S. cities.
The interesting strategic decision is what Amazon is not trying to eliminate.
Its own description of the program emphasizes that streamlining browsing and purchasing allows participating dealers to concentrate on the physical moment when customers arrive, pick up the vehicle, and begin what Amazon characterizes as the foundation of a longer-term relationship.
Think about that architecture.
Amazon is not arguing that humans no longer matter.
It is asking whether humans should spend their most valuable customer-facing time compensating for preventable transaction friction.
That is a much more serious competitive idea.
A customer accustomed to seeing an item, understanding the price, reviewing relevant information, maintaining context, completing appropriate steps, and then moving naturally into fulfillment does not forget that expectation because the product now has four wheels.
The transaction can legitimately become more complex.
The experience does not receive permission to become incoherent.
Amazon’s real competitive export into automotive is not e-commerce. It is the expectation that progress should survive the journey.
This should not frighten dealers who are willing to rethink the architecture around their customer.
In many ways, Amazon’s approach reinforces the continued importance of the local dealer. Inventory is local. The vehicle is physical. Delivery or pickup still creates a human moment. Service begins after the transaction rather than ending with it.
The opportunity is enormous.
But so is the expectation.
The first article in this series argued that the automotive conversation around AEO, GEO, and whatever acronym survives the week is too small because the major technology platforms are already building beyond the answer.
Google’s 2026 agentic-commerce work makes the point increasingly difficult to ignore.
With Universal Commerce Protocol, Google is creating a common language intended to let agents, businesses, and payment systems interact across discovery, buying, and post-purchase support. With Universal Cart, Google is connecting shopping activity across Search and Gemini, with YouTube and Gmail designed to join that environment as the rollout expands.
Again, automotive transactions have unique requirements. Nobody needs to pretend buying a vehicle is identical to buying socks.
The important part is the pattern.
The interface is becoming increasingly capable of preserving intent and context while helping the consumer move toward an action. Discovery no longer has to terminate at a list of links. Recommendation no longer has to terminate at an answer. The system can increasingly understand what the customer is trying to accomplish and help connect that intent with a commercial capability.
For automotive, this means the journey may become more distributed at exactly the same time customers expect it to become more coherent.
A shopper could begin with an AI assistant, move into video, evaluate dealership reputation, explore inventory, interact with a salesperson, calculate a trade, investigate financing, return to an AI system, schedule an appointment, and eventually enter the showroom.
From inside the dealership, those look like channels.
From the customer’s perspective, it was Tuesday.
The customer journey is becoming more distributed. The customer’s tolerance for experiencing that distribution is going down.
That is why intelligence architecture, interoperability, knowledge continuity, and human enablement matter more as AI becomes more capable.
The industry’s response cannot be another independent bot at every stage.
The customer does not need seven intelligent agents introducing themselves during one purchase.
The business needs enough shared intelligence that whichever human or digital surface appears next can continue helping.
If all of this sounds like an argument that technology companies inevitably win and traditional retailers inevitably lose, it is the opposite.
The dealership’s fundamental strategic position remains extraordinary.
Dealers possess something technology platforms spend fortunes trying to approximate: a dense collection of real-world relationships surrounding an enormously important customer asset.
The dealership has inventory a customer can touch today. It has employees with firsthand product knowledge. It has technicians capable of maintaining the vehicle for years after the initial transaction. It has local reputation, OEM relationships, operational history, customer relationships, financing capability, physical facilities, community knowledge, and countless human interactions that produce information no generic model could independently invent.
A salesperson understands why a family is nervous about moving from an SUV into a truck. A technician recognizes a problem from the sound the customer makes while trying to describe it. A service advisor knows when the technically correct explanation is not yet the useful explanation. A used-car manager can see what makes two seemingly identical vehicles meaningfully different. A dealer principal can authorize an exception when the right answer does not fit perfectly inside the workflow.
These are not embarrassing remnants of a pre-digital business.
They are differentiated intelligence.
The mistake would be allowing that intelligence to remain invisible to the rest of the journey.
The dealership’s greatest defense against commoditization may be the depth of human knowledge already inside it. Its greatest risk is failing to make that knowledge available when and where the customer needs it.
This is why the future should not be framed as humans versus AI.
The much more interesting question is what happens when dealership employees are surrounded by an intelligence layer capable of making their expertise accessible, giving them better context, removing administrative friction, and amplifying what they know across the customer’s journey.
That is a competitive experience very few pure technology companies can reproduce.
There is a paradox sitting underneath the AI transition.
As machines become better at performing routine cognitive work, the moments in which human judgment actually matters become easier to see.
Consider how much of the traditional dealership experience has required a person simply because the technology was inadequate. Someone had to retrieve the information. Someone had to move data between systems. Someone had to remember the conversation. Someone had to create the follow-up. Someone had to recreate the quote. Someone had to tell another department what the customer had already said.
Those were human tasks.
They were not necessarily human value.
AI and interoperable systems can remove enormous amounts of that work. If leadership is thoughtful, what remains is not a hollow dealership staffed by customers talking to armies of agents. What remains is a much better environment for the people capable of creating trust, judgment, empathy, expertise, creativity, accountability, and the occasional exception that saves a relationship.
The service advisor gets to advise.
The salesperson gets to understand the decision instead of reconstructing the funnel.
The technician gets better access to useful context and more time to solve the actual problem.
The marketer gets to interpret customer signal rather than spend the afternoon formatting output for five disconnected platforms.
The manager gets to manage exceptions rather than monitor software-generated activity.
That is the experience opportunity AI creates.
The purpose of an intelligence layer should not be to remove humans from the customer journey. It should be to remove enough meaningless work that the humans who remain can create disproportionate value.
That is a fundamentally different vision from replacing every human interaction with an autonomous system because autonomous systems became available.
One treats people as an operating expense waiting to be optimized.
The other treats human capability as an asset worth amplifying.
Customers will feel the difference.
The central argument of The Great Clearing has never really been about technology.
It is about leadership under amplification.
AI makes more possible. More content. More software. More automation. More customer interaction. More analysis. More personalization. More decisions made faster. More actions happening without waiting for another person to click the button.
That abundance makes discretion more important.
Article 1 argued that executives should stop mistaking another optimization acronym for transformation. Search is moving toward recommendation and action, which means businesses need a coherent knowledge environment rather than another collection of tactics.
Article 2 asked what happens once intelligence can act. The answer cannot simply be automating every inherited process. Leaders have to decide which work should disappear, which should become autonomous, which people should become more capable, and which human moments deserve intentional protection.
Article 3 confronted the inevitable flood of software that follows when creation becomes cheap. Builders should build. Executives should experiment. But the quality bar has to rise with the blast radius, and executive discretion becomes critical when a working demo can arrive months or years before the architecture underneath it deserves trust.
Article 4 cleared the desk. The next strategic layer is not another point solution. It is an intelligence architecture capable of understanding the organization across systems of record, inventory, customers, employees, policies, content, provenance, permissions, and performance—and then activating that understanding responsibly.
Now comes the part leadership cannot delegate.
What kind of experience are we using all of that capability to create?
AI can amplify almost any operating philosophy. Leadership decides which philosophy deserves the amplification.
A dealership can use AI to create more outbound pressure, more manufactured urgency, more generic content, more activity, more opaque automation, and more efficient versions of processes customers already dislike.
Or it can use intelligence to create continuity. To make the organization easier to understand. To preserve customer progress. To surface expertise. To give employees better context. To remove repetition. To make promises more reliable. To know when automation is appropriate and when a capable human should appear.
The technology may be similar.
The leadership is not.
The chasm opening in automotive is not between traditional dealers and technology companies.
It is not between digital retail and physical retail.
It is not even between AI-native companies and legacy companies.
Those are useful categories for conference panels.
The more important divide is between organizations that treat intelligence as a way to preserve the operating model and organizations willing to use intelligence to transform it.
One side asks how AI can make the existing funnel move faster.
The other asks why the customer is still being forced through the funnel in the same way.
One side automates the handoff.
The other asks why the handoff exists.
One side produces more content.
The other builds organizational knowledge worth amplifying.
One side deploys more agents.
The other defines what should be autonomous, what should remain human, and what should disappear entirely.
One side buys another system.
The other builds an intelligence architecture capable of making every appropriate system more useful.
That is the choice.
And it is becoming urgent because the market does not need to wait for the automotive industry to reach consensus.
Carvana can keep refining its experience. Amazon can keep importing familiar commerce expectations. Google can keep connecting discovery with increasingly agentic commercial infrastructure. AI systems can keep becoming more capable. New builders can keep attacking gaps incumbents ignored. Customers can keep carrying expectations formed everywhere else in their lives into the dealership.
The dealership has every asset necessary to compete.
It may even have the better hand.
But it has to play it.
There will not be durable winners on both sides of the customer-experience chasm. Eventually, the side that makes the customer’s life easier becomes the side customers teach the market to expect.
So clear the desk.
Find the signal beneath the noise. Establish which knowledge deserves trust. Eliminate the processes that no longer deserve to exist. Automate where automation removes friction. Amplify the people whose expertise makes the business genuinely useful. Open the infrastructure so knowledge can move securely. Measure what customers actually experience rather than celebrating machine activity for its own sake.
Deploy with purpose.
Execute ethically.
Lead by example.
Then use AI aggressively to amplify the version of the dealership worth preserving.
The industry does not need another declaration that everything is dead.
It needs leadership willing to decide what should live.
Previous: Clear the Desk: The Next Automotive Stack Is an Intelligence Layer. →
Return to the series hub: The Great Clearing: AI Is About to Amplify Everything. Choose Carefully.
Where Car Buyers Actually Search in 2026 →
See how discovery is spreading across traditional search, AI, social, video, dealer properties, and physical retail—and why continuity across those environments matters.
The Human Signal Surge →
Why identifiable expertise and firsthand experience become more valuable as generic machine-generated information becomes abundant.
Fewer Clicks, More Sales →
A practical look at what changes when AI influences discovery and decision-making before the customer ever reaches the website.
Cox Automotive: 2026 Car Buyer Journey Study →
Current evidence on AI adoption, digital efficiency, shopper satisfaction, and the growing importance of continuity throughout the buying process.
The experience described in this series requires dealership knowledge and customer context to move securely across the systems the dealer chooses. It requires authorized employees, agencies, OEM initiatives, builders, and AI systems to participate in the journey without forcing the customer—or another human—to become middleware between them.
That means documented APIs, webhooks, modern authentication, authorized third-party access, appropriate agent standards, portability, provenance, auditability, revocation, and dealer control.
Open infrastructure is not the absence of governance. It is the foundation for governed continuity.
Read and sign the Open Interoperability Demand →
Close the experience gaps. Open the infrastructure. Let dealers plug in.
Hrizn v6 helps dealership marketing teams move from fragmented activity toward a connected operating advantage—bringing intelligence, creation, human participation, distribution, proof and improvement into a more coherent system.
And the August rollout is not finished. What remains in v6 continues pushing toward a larger operating model in which trustworthy dealership knowledge, creators, inventory, intelligence, distribution, interoperability, and activation can work together to make the business more useful across every surface the customer touches.
The unlock ahead is not more AI for the sake of AI.
It is giving OEMs, dealers, agencies, vendors, employees, and ultimately customers a more coherent way to participate in the same intelligent experience.
We Rise Together.
