Good enough is now the most expensive zone in automotive
- Sean Cassy
- 5 hours ago
- 12 min read
TL;DR — For thirty years, "good enough" was the rational strategy in automotive. The math rewarded it. That math just broke. AI didn't have to become brilliant to disrupt this industry — it just had to become cheap, and it has. A $20/month consumer AI subscription now out-researches, out-compares, and out-thinks most of the touchpoints customers hit at average dealerships and average vendors. Good enough, once the safest zone in retail, is now the most exposed. This article walks through why, who gets squeezed first (vendors, not dealers), why optimization can't save you, and the actual path out — built on a three-layer diagnostic I've been refining for two decades. |
I was reading an article this morning by Ruben Hassid called You can't beat AI. Solid piece. Worth your 10 minutes if you have them.
But one sentence in the middle of it stopped me cold.
"The threat is not that AI becomes perfect. The threat is that AI becomes cheap enough to replace good enough."
I read it three times.
Because that's not the conversation we've been having in automotive. We've been having the AGI conversation. The "will AI take my job" conversation. The "wait until the models get better" conversation. All of it pointed at some future moment where AI becomes smart enough to matter.
Ruben just pointed out that we already passed the moment that actually matters. AI didn't have to become brilliant. It just had to become cheap. And it has been, for a while now.
Today is May 27, 2026. A $20/month ChatGPT or Claude subscription, in the hands of a customer who knows how to ask three good questions, will out-research, out-compare, and out-think the majority of the touchpoints that customer is about to hit on your dealer website, your CRM follow-up, your salesperson's product knowledge, and your vendor's "AI-powered" marketing tool.
Not the best dealerships. Not the best salespeople. Not the best vendors.
But the ones running on good enough.
And here's where I want to slow down, because this is the part that's going to land wrong if I don't say it carefully.
Good enough wasn't laziness. It was math.
I've been in this industry for 20 years. I've worked with over a thousand dealerships. I'm not standing on a hill throwing rocks at people who've been operating in the good enough zone, because for thirty years, good enough was the rational strategy.
Being excellent at every single touchpoint — excellent website, excellent SEO, excellent CRM follow-up, excellent merchandising, excellent salespeople, excellent video content, excellent social, excellent BDC scripts — would have burned capital you couldn't recoup. The cost of intelligence (people, time, training, expertise) was too high relative to the marginal return on excellence.
So the industry settled at an equilibrium. Everyone agreed to it. Customers tolerated it because they had no alternative. Vendors built businesses around delivering it. Dealers staffed around it. OEMs trained for it. Good enough wasn't a character flaw. It was the math working itself out.
That equilibrium is dead. And it's not dead because dealers got lazy or because vendors stopped trying. It's dead because the math flipped.
The cost of intelligence dropped roughly 6,000x in four years. The alternative to your salesperson, your website, your follow-up, and your vendor's tooling is now sitting in every customer's pocket for $20 a month. And it's getting smarter and cheaper every single quarter.
Which means good enough — the zone that used to be the safest, most capital-efficient place to operate — is now the most expensive zone you can be in. Because it's the only zone AI is already cheap enough to replace.
That's the inversion. Yesterday's smart move is today's most exposed position.
Why would the customer travel across town?
Put yourself in the shoes of a customer shopping for an SUV next weekend. Not five years from now. Next weekend.
She opens her phone. She has $20/month Claude or ChatGPT. She doesn't think of it as AI. She thinks of it as the thing she asks questions to. She types:
"I'm looking at a 2026 Equinox, a RAV4, and a CR-V for a family of four in [your city]. We do one road trip a year. Compare them on real-world reliability, total cost over five years, and which one holds value best on the local used market. Be honest about the weaknesses."
She has, in about 90 seconds, a comparison that is more thoughtful, more honest, and more thorough than what 80% of salespeople could deliver in a 45-minute walkaround. Not because the salesperson is bad. Because the salesperson is operating at good enough. And good enough just got compared against $20/month.
Now extend the same exercise to the rest of the customer journey.
Why would she sit through a good enough sales presentation when AI gives her a tailored one for free? Why would she rely on a good enough dealer website when her assistant can pull spec sheets, reviews, and forum threads across every brand in town in 10 seconds? Why would she trust good enough finance education when she can ask AI to explain the difference between subvented rates, residual leases, and balloon payments — in plain English, with her specific income — before she walks in? Why would she even call your BDC when she can get the answer faster from the thing in her hand?
The answer is: she wouldn't. And increasingly, she isn't. The shopping behavior has already shifted. Most of us just haven't seen the data catch up yet.
This is the part of the conversation that's going to be uncomfortable for some readers, because it implicates the entire customer-facing layer of the dealership at once. But the discomfort is the signal.
The customer journey was built for a world where intelligence was expensive and dealers had a near-monopoly on it. Both of those conditions just stopped being true. The journey hasn't caught up to the new reality yet — but it's going to, fast, because customers will keep choosing what works for them whether or not the industry's ready.
The squeeze hits vendors before it hits dealers.
Here's the part that nobody in our industry is saying out loud, so I'm going to.
Dealerships have physical moats. Inventory. Service bays. Financing relationships. The test drive. The trade-in appraisal. The F&I office. The handshake. AI doesn't replace those overnight, even when it gets dramatically cheaper. There's a real-world layer to selling cars that doesn't fully evaporate.
But vendors? Vendors whose entire product is "good enough website, good enough SEO, good enough social automation, good enough email cadences, good enough chat widget, good enough video tool, good enough reputation management"?
Their product is intelligence. There is no physical moat. And the people doing the replacing are the dealers' own internal teams, sitting at their desks with a $20/month Claude subscription, slowly figuring out that they can produce better output in 15 minutes than what they're paying $4,000/month for.
This is happening right now. I'm watching it happen in real time with the dealerships I work with. Internal teams are starting to ask the question — quietly, then less quietly — "Why are we still paying for this?"
The 80% of vendors who built businesses on delivering good enough are in a worse position than the dealers who buy from them. The dealers at least have inventory and service. The vendors don't have anything except the work, and the work just got commoditized.
If you're a vendor reading this — I'm not saying your run is over. I'm saying the good enough version of what you do is. The vendors who survive this are going to be the ones who move above the line where AI is cheap. The ones who rebuild around taste, judgment, accountability, and strategy that no $20/month tool can replicate.
The ones who don't will spend the next 24 months watching their renewals quietly stop coming back, and they'll blame the economy, or rates, or tariffs, or anything other than the math.
You can't optimize your way out of this.
Here's the trap I see most dealers and vendors falling into right now: they're trying to optimize their way out of a problem that isn't an optimization problem.
They're A/B testing harder. They're tweaking the funnel. They're squeezing 3% more conversion. They're sending more emails, making more calls, posting more on social. They're working harder inside the same architecture that got them to good enough in the first place.
That's the end of the hustle culture, and it's already over — most people just haven't gotten the memo.
You can't out-hustle the math. The math says intelligence is cheap and getting cheaper. No amount of effort inside the old system fixes that. Optimization is what you do inside an equilibrium. When the equilibrium itself collapses, optimization is rearranging deck chairs.
What works now is architecture. Building a different operating system. Not "use AI tools." Not "add ChatGPT to our workflow." Actual infrastructure: extracted expertise, structured intelligence, repeatable systems, an intelligence layer your team owns rather than rents.
That's a fundamentally different game than the one most of the industry is still playing. It's not harder work — it's different work. And the people who figure out the difference between those two things in the next 12 months are going to look up in 2027 and find themselves operating in a completely different competitive position than the rest of the industry.
So what do you actually do.
If you're reading this and feeling the squeeze — if you're looking at your vendor stack and thinking "none of this is good enough anymore," if you're looking at your team and wondering how to move them out of the good enough zone, if you're a vendor wondering how to move yourself out — you're not alone, and you're probably right.
But there is a path. You don't have to guess your way out of this.
Across 35 years in business and 20 specifically in automotive retail, I developed a diagnostic lens for separating any new capability into three layers: Fundamentals, Frameworks, and Velocity. I call it the Mastery Engine. It's how I ran marketing operations from the outside for dealer clients who generated over $2 billion in tracked sales.
The same three-layer pattern shows up everywhere — in how a car is built (frame and structural rigidity first, then steering and brakes and suspension, then the engine), in how the Wright Brothers beat their competitors to flight (years on aerodynamics before they ever strapped on an engine), in why the industry's transition to the internet broke in the late 1990s and still hasn't been fixed (we uploaded inventory and built lead forms, then never installed the conversion psychology underneath).
Applied to AI, the layers go in this order, and the order is not optional.
Fundamentals — the extraction work
This is where almost every dealer AI initiative breaks before it begins, because almost every dealer AI initiative skips this entirely. Fundamentals is the work of pulling what's buried in your head and your team's heads — your customer avatars, your dealership voice, your market knowledge, your operational instincts, the institutional wisdom nobody's ever written down — and turning it into something AI can actually work with.
Your best salesperson knows things about your customers that aren't written anywhere.
Your GM knows which leads are worth calling at 7 pm and which ones aren't, and can't tell you why — they just know. Your service manager carries a mental model of which warranty issues escalate and which ones don't. Your finance manager has pattern recognition built over thousands of deals that lives entirely as instinct. That's what AI needs in order to produce output that sounds like your dealership instead of output that sounds like every other dealership using the same chatbot vendor.
Extraction is unglamorous work. It looks like conversations, documents, voice memos turned into structured files. It's slower than dumping prompts into ChatGPT. It's also the only thing that separates AI output that converts from AI output that reads generic. The vendors selling you "AI-powered" anything skipped this step entirely. That's why their output reads the way it reads.
Frameworks — the structures that organize fundamentals into operation
Once the fundamentals are extracted, frameworks are what turn them into repeatable, executable systems. Custom GPTs and Claude Projects built on your extracted expertise. Content engines that produce sales copy in your voice. Video engines that produce VEO-format scripts using your actual customer avatars. Lead-handling workflows that route differently based on the patterns your top performers carry in their heads. The tools and templates that organize the underlying knowledge into something a team can execute against — in your voice, for your market, against your real customers.
Frameworks are where most AI consulting starts. That's why most AI consulting fails.
Without fundamentals, frameworks are scaffolding around nothing — they look impressive in a demo and produce generic output in production, because there's no extracted expertise inside them to draw from.
Velocity — speed, scale, and eventually agents
This is where the leverage everyone's been pitched for two years finally becomes real. Sustained content output. Cross-channel campaign production at a cadence no traditional agency can match. Agentic workflows that handle complex tasks autonomously. The full AI OS that some dealers are starting to dream about — and that the entire industry is genuinely heading toward.
But only once the foundation is built to support it. Agents installed on a foggy foundation rot in six months. AI content produced without brand voice reads generic. Automation deployed on un-extracted processes just accelerates the wrong work, faster, at scale, with more confidence — which is worse than not automating at all.
Velocity without fundamentals is expensive failure. Every AI program in this space is selling velocity. Almost none of them install the foundation first. That's why most of the AI initiatives launched in dealerships over the past 24 months have produced a pile of subscriptions and a vague sense of falling further behind, instead of compounding operational leverage.
The activation layer underneath all of it
Here's what changes everything once you understand it: you don't install AI into a dealership. You activate the operator, and the dealership transforms as a downstream consequence. The unit of installation is the person, not the building.
Every other AI consultant, platform, and vendor pitches dealership-level transformation. The dealer principal or GM is treated as the buyer of a technology adoption. The frame is dealership-shaped: better conversion, more leads, reduced cost per sale, operational efficiency.
That frame has a problem. AI doesn't actually install into buildings. It installs into people. A Custom GPT built on your dealership's extracted intelligence is a tool. The capability to build it, refine it, deploy it, and iterate on it lives inside the operator who learned to do it.
If the operator moves on, leaves, sells, gets promoted — the capability moves with them. If the dealership tries to operate AI without that activated operator inside the building, the infrastructure rots fast.
Which means whatever you build, you want to build through a human who's been activated as an AI-native operator, not around one. The capability stays with that person. It travels with them. Into your next dealership, into your group, into whatever you eventually point your attention at next. The dealership gets transformed. The capability is yours for life.
This is what AI Apex is built to do.
AI Apex is a 90-day, one-on-one program that activates automotive operational leaders — dealer principals, GMs, VPs of Operations, COOs, marketing directors with real strategic authority — as AI-native operators. Fundamentals first. Frameworks second. Velocity last. The Mastery Engine as the diagnostic lens that keeps the order right, so what gets installed actually compounds instead of rotting.
It's not a course. It's not a tool tour. It's not a software subscription. It's one-on-one work between you and me over 12 weeks, where we extract what's already in your head and your team's heads, build the custom tools and engines that operationalize it, and activate you as the kind of operator who can keep building, refining, and extending the capability long after the engagement ends.
The dealership gets transformed. You keep the capability for the rest of your career — inside this dealership, inside your group, inside your investment portfolio, inside anything else you eventually decide to build.
If you're feeling the squeeze this article is describing, the next step is a conversation. Not a sales pitch. A diagnostic conversation where I'll ask you about your current operating model, the layers you've built or skipped, and the specific places where good enough is starting to cost you more than it used to. By the end of the call, you'll know whether AI Apex is right for your situation — and if it isn't, you'll have a much clearer picture of what is.
[Book a diagnostic call] — limited slots each month, deliberately. Activation is one-on-one work, and the program isn't scalable in the way most consulting offers are. That's the point.
The choice nobody escapes.
Move above good enough. Or watch the math close the gap on you.
The 5% who refuse to be TKO'd out of this industry are already building. The other 95% are still optimizing.
Which one are you?
About the author:
Sean Cassy is a seasoned marketing professional with a passion for transforming businesses through powerful marketing strategies. With over 35 years immersed in the world of marketing, and as the co-founder and partner of Turbo Marketing Solutions for the past 20 years, Sean has a rich history in delivering results. He has personally crafted over 2,500 marketing funnels, edited 5,000 videos, and generated leads that have culminated in over $2 billion in sales for clients.
Sean’s deep involvement with AI marketing tools from companies worldwide, coupled with his vast experience in the automotive marketing industry, has uniquely positioned him as a thought-leader in the AI marketing space. He is now committed to leveraging his expertise to help businesses across all verticals seize the AI opportunity early, and gain a competitive edge.
Sean’s wealth of experience, continuous learning, and proven track record in delivering results underscore his Expertise, Authoritativeness, and Trustworthiness in the field of AI marketing.
Comments