Define Before
You Deploy
Why AI literacy, not vendor adoption, is the critical success factor for automotive dealers.
Section 01 // Executive Summary
The problem is sequencing.
Automotive dealers are spending on AI. Budgets are moving, vendor contracts are being signed, and there is genuine enthusiasm across operations of every size. The intent is real. The urgency is understandable. The problem is sequencing.
Most dealerships are making tool adoption decisions before their people have the foundational literacy to evaluate what they are buying, use it effectively, or protect their business data in the process. The result is a familiar pattern: money moves, adoption stalls, and the expected productivity gains do not materialize.
This paper argues that the critical success factor for AI adoption in automotive retail is not the tool, it is the operator. Specifically, it is the ability of dealership staff at every level to define a business problem clearly before they touch an AI tool. That skill, problem definition, determines whether AI generates real workflow value or produces a stream of mediocre outputs that get accepted without scrutiny or abandoned in frustration.
The evidence is not theoretical. Research from Deloitte, DataCamp, IBM, and BCG consistently shows that organizations pairing AI investment with structured literacy training are nearly twice as likely to see strong returns. The OECD has concluded that current training supply is insufficient to meet growing demand for general AI literacy skills across the workforce.
The recommendation for dealers is straightforward: define the business problem before you evaluate the tool. Invest in baseline literacy before you expand vendor spend. Let your team's capability become the filter through which purchasing decisions get made. The dealers who build that foundation now will have a compounding advantage over those who keep buying first and learning later.
INTERACTIVE / Sequence comparison
Two ways to sequence AI adoption.
One produces frustration. The other compounds.
Tap each card below to compare the two sequences.
Buy first, learn later
- 01Money moves before capability does
- 02Underused subscriptions stack up
- 03Staff frustration grows; tools get abandoned
- 04Budget cycles repeat the same pattern
INTERACTIVE / Pulse check
Where does your dealership actually sit?
Three quick questions. Your answer plots against the capability × adoption-speed matrix.
Answer each question to plot your dealership on the matrix.
How would you rate your team's current AI skill level on the floor?
How structured is your team's current AI training?
How much vendor pressure to adopt new AI tools are you feeling right now?
Answer all three to position your dealership.
Section 02 // The Floor
The state of AI in automotive retail.
The public-facing AI tools are accessible, affordable, and increasingly capable. And for dealers who get the fundamentals right, the productivity gains are real and demonstrable. Staff who learn to use AI with a clear framework are producing better customer communications, faster internal reports, and more consistent content, often in the same session where they first encountered the tool.
But enthusiasm and capability are not the same thing. What is visible across dealerships from coast to coast is a gap between the speed at which dealers want to move and the skill level that currently exists on the floor. That gap is not a criticism of the people in the room. It is an honest description of where the industry sits today, and it matters enormously for what happens next.
The skill level across most dealership operations remains novice. Staff are opening AI tools and typing vague requests. They are getting mediocre outputs. They are either accepting those outputs and putting them into the world, or concluding the tools are not useful. Neither outcome serves the business.
Meanwhile, the vendor landscape has expanded aggressively. Every major DMS, CRM, inventory platform, and agency now has an AI feature to sell. Some of these are genuinely useful. Some are rebranded automation. Some are early-stage products that will mature over time. The challenge is that dealers are being asked to evaluate and purchase these tools before they have the literacy to tell the difference.
That is the core tension this paper addresses. The opportunity in AI is real. The risk is not that dealers will adopt too slowly, it is that they will spend without the foundation to capture the value they are paying for.
Section 03 // Vendor-Led Adoption
The order of operations is the problem.
The loudest voices in the room about AI at most dealerships are not internal. They are vendors. And that is a sequencing problem.
Dealers are being presented with demos, feature announcements, and subscription upgrade offers at a pace that outstrips their team's ability to evaluate what they are actually seeing. A demo environment is controlled. Real value requires real workflow integration, and that requires a skilled operator who knows how to define what they need, structure a request, and evaluate whether the output is actually good.
Without that foundation, purchasing decisions get made on the basis of what looks impressive in a thirty-minute presentation, rather than what solves a specific, documented workflow problem. The result is a library of subscriptions that gets underused, a team that is frustrated by tools that never quite deliver what was promised, and a budget cycle that repeats the same pattern.
The enterprise data supports this pattern at scale. Research from IBM indicates that only one in four AI projects achieves its expected return on investment. BCG research shows that seventy-five percent of businesses see no tangible value from AI investments, with the root cause being tool adoption without adequate people preparation. These are not small-scale experiments, they represent billions of dollars in investment producing negligible returns because the human capability layer was not built before the tools were deployed.
The scale of the broader spending pattern reinforces the urgency. Companies are projected to spend $644 billion on generative AI in 2025. Yet only six percent have begun meaningful workforce upskilling, despite eighty-nine percent acknowledging critical skill gaps in their organizations. The gap between investment and preparation is not narrow. It is structural.
For dealers, the version of this problem is proportional but identical. The budget is smaller, the stakes are more immediate, and the margin for wasted spend is tighter. A dealership that signs a $1,500-per-month AI platform contract without first establishing whether its team can use the tool effectively is not making a bold investment in the future. It is making an expensive guess.
The solution is not to stop evaluating vendor tools. It is to change the order in which things happen. Literacy first. Tool evaluation second. Purchasing decisions last.
INTERACTIVE / Industry signal
Scroll into view to watch each figure count up.
Of enterprise AI projects achieve their expected ROI.
SRC / IBM
Of businesses see no tangible value from AI investment.
SRC / BCG
Projected global generative-AI spend in 2025.
SRC / Industry analysis
Of companies have begun meaningful workforce upskilling.
SRC / 2025 data
INTERACTIVE / Calculator
Estimate your own cost of guessing.
Drag the sliders to model your annual waste.
“I'm not saying stop buying AI tools. I'm saying change the order of operations. Let your staff's literacy level become your filter for vendor conversations. If your team can't articulate the business problem the tool is supposed to solve, you're not ready to buy it yet.”
Section 04 // The Literacy Gap
What it actually looks like on the floor.
When the conversation turns to AI literacy, the instinct is often to frame it as a technology skill, something the IT department handles, or something that requires a certification or a formal course. That framing is wrong, and it is part of why the gap persists.
The core literacy gap in dealership operations is not about technology. It is about problem definition. The people working with AI tools every day are not failing because they cannot navigate an interface. They are failing because they have not been taught to answer four questions before they type:
- What is the specific task I am trying to complete?
- Who is the audience or recipient of the output?
- What format do I need the output in?
- What should this output never include?
Without that foundation, staff are left to figure it out on their own. There is no structured path, no shared framework, and no standard for what good looks like. The OECD has reached a similar conclusion at a policy level, finding that current training supply may not be sufficient to meet the growing need for general AI literacy skills across the workforce, and that most available AI training is targeted at specialists rather than the general working population.
In the dealership context, this plays out in specific and recognizable ways. A service advisor drafts a response to a difficult customer complaint. They open an AI tool, type something like 'write a response to an angry customer,' and get a generic paragraph that sounds nothing like the dealership, addresses none of the specifics, and would be embarrassing to send. They either send it anyway, rewrite it entirely, or stop using the tool.
A sales manager wants to create a follow-up email sequence for customers who came in but did not purchase. They ask the AI for 'a follow-up email' and get a single generic message with no awareness of the customer's situation, the vehicle they looked at, or the dealership's voice. The tool looks useless. The skill gap is invisible.
Data hygiene lives in this same space. Staff who have not been taught to define a task clearly also have not been taught to think about what information belongs in a prompt. Customer names, deal details, unreleased inventory specifics, and internal financial data routinely end up inside tools whose handling of that information is poorly understood.
INTERACTIVE / Prompt Lab
See the gap in one click.
Pick a scenario. Read the lazy prompt. Then apply the framework.
Switch scenarios and toggle between "The Guess" and the framework.
write a response to an angry customer
Dear Customer, we are sorry to hear about your experience. We take feedback seriously and will look into the matter. Please contact us at your earliest convenience.
INTERACTIVE / Data safety drill
Should this go in a prompt?
Eight snippets. Sort them. The habit is the point.
Sort each card into "Safe" or "Never", your score updates live.
Full customer name, address, and phone number from the deal jacket
Section 05 // Why Literacy Wins
The compounding judgment of a trained team.
The service advisor who has been taught to specify the role, the situation, and the format gets a response they can actually use. The sales manager who builds a proper prompt sequence gets a follow-up series that sounds like their dealership and addresses the specific purchase stage their customer is in. The GM who spends thirty minutes learning to structure a request produces an internal summary, a vendor briefing, or a staff communication in the time it used to take to write a rough draft.
These are not dramatic transformations. They are incremental, practical gains, and they compound. A team that knows how to use AI well does not just produce better individual outputs. It develops judgment about where the tools fit and where they do not. That judgment is what separates a team that captures real value from AI from one that keeps cycling through disappointing experiences.
The ROI data makes the case clearly. Research from DataCamp and YouGov, based on a survey of more than five hundred enterprise leaders, found that just over one in five organizations report significant positive return on AI investments overall. Among organizations with mature, organization-wide AI literacy upskilling programs, that figure roughly doubles.
Literacy also changes how dealers evaluate vendor tools. A staff member who has built a working understanding of how prompts shape outputs can ask sharper questions about any new product: how editable it is, what data it draws on, how its prompts are structured, and what happens when the input is ambiguous. Those questions expose the difference between a genuinely useful product and a well-packaged demo. A team without that foundation cannot ask them.
Literacy also changes how dealers protect their operations. A staff member who understands what an AI tool does with input data is far less likely to put customer information, deal details, or proprietary business data into a prompt carelessly. The habit of asking 'should this be in here?' develops naturally when people understand the relationship between input and output. It does not develop when people are simply handed a subscription and told to explore.
Section 06 // The Framework
Define Before You Deploy - three steps.
The shift from vendor-led to literacy-led AI adoption does not require a significant structural change. It requires a different order of operations. Three sequential steps change the outcome.
Step 1 - Define the business problem first
Before evaluating any AI tool, vendor-built or general purpose, the question to answer is: what specific workflow problem is this supposed to solve?
That question has to be answerable in concrete terms. Not 'we want to be more efficient' but 'our service advisors spend forty-five minutes a day writing follow-up summaries that customers don't read.' Specific problems lead to evaluable solutions. Vague aspirations lead to a stack of vendor logos and no measurable outcome.
Step 2 - Build baseline literacy before expanding spend
Baseline literacy does not require a week-long certification or a specialized course. It requires a few hours of structured instruction focused on one skill: how to construct a clear, effective request.
A useful framework for this, one that has proven effective across dealership roles from receptionist to general manager, organizes every AI interaction around four elements: the role you are asking the tool to play, the specific instruction you are giving it, the parameters that shape the output (format, length, audience, tone), and the exceptions that define what it should never do.
That structure is not complicated. It is learnable in a single session. And it produces an immediate, visible difference in output quality that gives staff a reason to keep using the tools rather than abandoning them after a few disappointing results.
Data safety belongs in this same session, framed practically rather than as a compliance lecture. Staff need clear answers to two questions: what types of information should never go into an AI prompt, and what does that mean for the specific tasks they are trying to accomplish.
Step 3 - Let capability filter the tool decisions that follow
This is not about being skeptical of vendors. There are genuinely useful AI tools built specifically for automotive retail, and dealers who evaluate them from a position of literacy will get more value from the ones they choose. The goal is informed purchasing, not reluctant purchasing.
INTERACTIVE / R-I-P-E Builder
Build a prompt the right way.
Role · Instruction · Parameters · Exceptions. Edit each field, copy the result.
Tap a chip preset for each field, then copy your prompt.
Role: Service Advisor at a franchise dealership Instruction: Draft a follow-up email to a customer who declined an extended warranty Parameters: Tone: warm, plain, accountable. Length: under 120 words. Format: email. Exceptions: No corporate apology language. No legal disclaimers. No discount offers.
INTERACTIVE / Readiness Assessment
Three questions. One honest assessment.
Answer Yes or No to see where your organization stands today.
Answer all three to reveal your readiness verdict.
01Have you identified one staff member per department to develop working AI literacy as part of their existing role?
02Before approving new AI vendor spend, do you require an internal champion to demonstrate the specific workflow problem it solves in your environment?
03Have you run at least one structured AI literacy session for staff before expanding tool access?
Answer all three questions to see your readiness assessment.
Section 07 // For Leadership
What this means for dealer principals and general managers.
The three steps below are operational. But they require a decision at the leadership level to implement. Dealer principals and general managers are the ones who set the conditions under which AI adoption either produces real value or produces expensive frustration.
The practical leadership ask is narrow and achievable:
- Identify one staff member per department who will develop working literacy with AI tools, not a dedicated AI role, but a responsibility assigned to someone already doing the job well.
- Before approving any new AI-related vendor spend, require the internal champion to demonstrate what specific workflow problem it solves, in your workflow, not in a demo environment.
- Run at least one structured literacy session before expanding tool access. A few hours of foundational instruction changes what your team does with every tool they touch afterward.
The competitive framing is simple: AI tools are a commodity. Literacy is a differentiator. The window to build that advantage before it becomes a catch-up exercise is open now, but it will not stay open indefinitely.
Section 08 // Conclusion
Define before you deploy.
Dealers do not have an access problem. AI tools are available, affordable, and increasingly integrated into the platforms the industry already uses. The access question has largely been answered.
The readiness question has not.
The gap between AI investment and AI return in automotive retail, as in every other industry, comes down to whether the people using the tools have been given a framework for using them well. That framework starts with one skill: define the business problem before you touch the tool.
That skill is teachable. It does not require technical background, extended training programs, or significant budget. It requires a decision to invest in people before platforms, to build the human foundation that determines whether every tool purchase that follows produces value or produces frustration.
Define before you deploy. The tools will still be there when your team is ready to use them properly.
Take the full paper with you.
Download the complete PDF white paper to share with your leadership team, or revisit any section above.
Section 09 // About the Author
Brent Wees
Workshop Designer
Brent Wees is a Certified AI Trainer and workshop designer who has spent the past year working directly with automotive dealerships on the practical realities of AI literacy, how staff actually use these tools on the floor, where the skill gaps live, and what it takes to build a foundation that makes vendor evaluation, data safety, and productivity gains actually attainable.
His work focuses on the operator, not the technology, the person sitting at the keyboard who decides whether AI becomes a workflow asset or an expensive subscription.