English
Français
繁體中文
What you say after the bill comes due
There is a type of confession that big organizations make only when the numbers have already hardened into something unarguable. It is not a confession of error, exactly, because error implies that someone with a hand on the lever pulled it the wrong way. This confession is softer than that. It is a confession that preserves the organization’s dignity and maintains the myth that the decision was rational, even if the outcome wasn’t expected.
It goes like this: We overestimated the pace.
If you listen carefully, you can hear what the sentence implies. It is separating the decision from the results. The decision was sound; the world simply failed to cooperate. It is also shifting the locus of blame from judgment to physics. Pace is not wisdom or stupidity; it is a rate of movement. Pace is the current in a river. We didn’t misread people; we misread the speed of history.
And yet, when Ford, GM, and Stellantis each began, at different moments and in different corporate dialects, to suggest some version of that same sentence, what became interesting was not whether they were right or wrong about electric vehicles. What became interesting was the eerie synchrony of the whole thing: how three companies with different boards, brands, cultures, and product portfolios all arrived at roughly the same posture toward the future, and recalibrated.
If you want to explain this in a familiar way, you can. The infrastructure was patchy. The economics were intractable. The customer was hesitant. Incentives changed. Rates went up. You wind up with a long list, and every item on it is true in some measure. But the standard explanation doesn’t quite capture what it feels like to watch these giants move in parallel, because it suggests something beyond market conditions. It suggests a social process.
The most influential piece of market research is your competitor
The auto industry, after all, is not merely a set of factories and supply chains. It is a tribe. It is a community of decision-makers who read the same analyst notes, attend the same conferences, absorb the same headlines, and, most importantly, look constantly over one another’s shoulders. The most powerful market research in Detroit has never been a survey. It has been the behavior of other organizations in Detroit.
There is a concept economists use for this kind of phenomenon, and it sounds like something that belongs in a graduate seminar until you see it play out in real life. It is called an informational cascade. The idea is simple. When you do not know what to believe, you look at what others are doing. If the first few people head in one direction, you assume they know something. So, you follow. And the next person follows you. And then the next. Soon, the crowd is moving not because the evidence compels movement, but because movement itself has become evidence.
In a normal setting, informational cascades are harmless. You choose the busy restaurant instead of the empty one. You buy the book everyone is talking about. You watch the movie that has a line out the door. The cost of being wrong is minor. But the cascade becomes dangerous when the cost of being wrong is not a bad meal or a boring evening, but a decade of capital investment.
That is the world Ford, GM, and Stellantis live in. In that world, a strategic bet is not a choice. It is a reorganization of reality. It determines what engineers study, what suppliers build capacity for, what factories get retooled, what kinds of talent a company can attract, what stories marketing is allowed to tell, what dealers are expected to sell, and what Wall Street is encouraged to believe. A large automaker cannot dabble. It commits. And when it commits, the commitment itself becomes a signal to everyone else.
When commitment becomes a form of evidence
This is the first subtle mistake in the EV story: the assumption that when a competitor commits, it must be because that competitor knows something.
Imagine you are in the executive ranks of an incumbent automaker at the moment electric vehicles become not merely a product category but a cultural inevitability. You are under pressure from regulators, from investors, from the press, from Silicon Valley, from the climate narrative, from every force that makes internal combustion look like the rotary phone of transportation. You have research. You have forecasts. You have teams of people running scenarios. But you also have something far more persuasive than any internal model: your competitor’s public pledge.
A rival announces billions in spending. It unveils a platform strategy. It talks about timelines and targets, in the confident language of someone who has glimpsed the future. And your brain infers that confidence implies information. If they are acting, they must have data. If they have data, they must be seeing demand. If they see demand, then hesitation is a risk. Soon, the most rational move is to align with what appears certain.
That, in its purest form, is the cascade.
The moment strategy turns into identity
The second subtle mistake is that once a cascade begins, it alters the organization’s internal sociology. It becomes harder to ask inconvenient questions, not because anyone forbids you, but because the very act of questioning begins to sound like disloyalty. “Do we know the customer will tolerate charging friction?” becomes “Do you not believe in the company’s direction?” “Are we sure adoption will be uniform across regions and segments?” becomes “Are you trying to slow us down?” In the early stages of a transformation, dissent can be framed as prudence. In the later stages, dissent is framed as obstruction.
And that is how a portfolio decision turns into an identity.
The EV transition, in other words, was never just a bet on a powertrain. It was a bet on a story. It was a bet that the world wanted a particular kind of future quickly and that the companies that got there fastest would be rewarded. Once you bet on the story, you begin to behave as if the story is already true. You hire as if it’s true. You message as if it’s true. You plan factories as if it’s true. You treat alternative futures as distractions. The bet becomes self-reinforcing because the organization’s social incentives are now aligned around belief.
Reality arrives as something very practical
Then reality arrives, not as a philosophical rebuttal, but as a very practical one.
Reality arrives as inventory sits longer than expected. Reality arrives in the form of incentives to move products. Reality arrives as the dealer who is not paid to validate your narrative but to keep the lot moving, quietly discovering that the hybrid is the easiest thing to sell to the customer who is intrigued by electrification but wary of the lifestyle changes that pure EV ownership can impose. Reality arrives as the realization that “the market” is not one market. It is a patchwork of markets: urban and rural, homeowners and apartment dwellers, regions with dense charging networks and regions with long stretches of nothing, buyers who treat the vehicle as an appliance and buyers who treat it as an extension of identity.
If you want a single, almost comically revealing statistic from this period, it is not about EVs collapsing. It is about hybrids rising. In the United States, the share of light-duty vehicle sales that are hybrid, battery-electric, or plug-in hybrid has grown, but within that growth, hybrids have been gaining while the shares for battery-electric and plug-in hybrid have been comparatively flatter for the period the US Energy Information Administration highlighted. That is not a verdict against electrification; it is a verdict against the assumption that electrification would be a single smooth curve. Hybrids are the market’s way of saying: we want direction, but we are not ready to go the full distance, and we want to take the path of least disruption.
Where PLM belongs in this story
So, where does PLM fit into a story about sociology and self-fulfilling narratives?
If you hear “PLM” and think of engineers managing CAD files and revision histories, that reaction is not wrong. But it is incomplete in a way that matters here, because the deeper idea of PLM, the idea that becomes important when companies are making civilization-scale bets, is not about files. It is about decisions. Specifically, it is about the discipline of turning decisions into traceable objects rather than evaporating conversations.
The EV cascade was not driven by one bad assumption. It was driven by a stack of assumptions that reinforced one another. Assumptions about consumer willingness to change habits. Assumptions about charging density and reliability improving at a certain rate. Assumptions about battery costs. Assumptions about policy stability. Assumptions about resale value. Assumptions about dealer motivation. Assumptions about the willingness of a mass-market buyer to accept friction for moral satisfaction.
Each assumption can be argued. Each assumption may be true in one segment and false in another. Each assumption may be true in 2035 and false in 2025. But in large organizations, assumptions rarely remain labeled as such. They drift. They become “the plan.” They become “the direction.” They become the unspoken foundation under thousands of downstream actions.
When that happens, the organization becomes brittle. It cannot bend without breaking, because bending suggests that the foundation was wrong.
This is where a mature PLM mindset, again, not as a product pitch, but as a way of running an enterprise, quietly offers a countermeasure to the cascade. A PLM discipline, at its best, treats decisions as artifacts with lineage. It forces the organization to keep track of what it believed, when it believed it, why it believed it, and what that belief touched.
Think about what that means in practical terms. If your enterprise treats the assumption “EV demand in full-size trucks will be X by year Y” as a managed object, owned, versioned, tied to specific program milestones, connected to supplier capacity reservations, factory investments, and configuration strategies, then when the market sends contradictory signals, you can revise the assumption without rewriting history. You can say: “Version 1 of this assumption is no longer supported by evidence. Here is the new evidence. Here is the updated version. Here is what changes downstream.”
That sounds bureaucratic until you realize what the alternative is. The alternative is the modern corporation’s default mode: assumptions spread across decks and meetings and executive charisma, which means no one can find them, no one can interrogate them cleanly, and no one can revise them without making it personal. The consequence is not merely slower learning; it is the political risk that comes with learning. People learn in private and deny in public. And once that becomes the pattern, the organization loses the ability to update itself at the same pace as the market.
What this means for PLM teams, business leaders, and program managers
If the EV story is really a story about pace, about timing being the variable that humbles everyone, then the practical question becomes: who, inside a large organization, is responsible for making “pace” something the company can survive being wrong about? The answer is: everyone, but in different ways. The point isn’t to make perfect predictions. The point is to build a decision system that can absorb imperfect predictions without turning every adjustment into a political crisis.
For business leaders, the lesson is to stop treating strategy as a proclamation and start treating it as a set of hypotheses backed by evidence. In a cascade, the organization becomes addicted to the comfort of a single story because it makes alignment easy. But alignment built on an untracked stack of assumptions is fragile. The leadership move isn’t “be less ambitious.” It’s to be explicit about what must be true for the ambition to pay off, and to decide before the excitement peaks, which signals a change in direction. If you don’t pre-negotiate the right to revise, then every revision looks like a retreat. And when revisions look like retreat, you get denial instead of learning.
For program managers, the takeaway is that the real enemy is not complexity, it’s invisible complexity. The EV transition multiplied dependencies: suppliers, plants, compliance, architecture decisions, software, service readiness, dealer behaviors, and cost curves. That web is manageable when it’s visible and governed, and it becomes unmanageable when it lives inside people’s heads and on PowerPoint slides. Program management is often seen as the art of keeping tasks on time. In moments like this, it becomes something more important: the discipline of keeping decisions coherent as assumptions shift. The most valuable program manager isn’t the one who hits every date; it’s the one who can show, calmly and concretely, what changes when the world changes, and can do it early enough that the organization still has choices.
For PLM teams, the opportunity is to play a role that’s almost never written into the charter: making doubt operational.
In a healthy engineering culture, requirements are not slogans. They’re defined, owned, tested, revised, and traced to outcomes. The same mindset can apply to strategic assumptions, especially the ones that quietly drive the biggest commitments. When assumptions about adoption pace, segment readiness, charging friction, or cost targets remain implicit, they become untouchable. When they are captured, versioned, and connected to the programs they influence, they become discussable. And when they become discussable, the organization can change its mind without losing face.
None of this requires a company to be timid. In fact, it’s the opposite. It’s how you stay bold without becoming brittle. It’s how you keep momentum without letting it masquerade as evidence. And it’s how you avoid the trap that sits at the heart of every cascade: the moment when you confuse the sound of everyone marching together with the sound of certainty.
The lesson isn’t “don’t bet,” it’s “don’t confuse marching with knowing”
So perhaps the most useful way to interpret the Detroit Three’s EV recalibration is not as failure, and not as betrayal, and not as proof that EVs were a mirage. It is, instead, a case study in how large institutions get seduced by the behavior of other large institutions, and how that seduction can be mistaken for insight.
When everyone marches in the same direction, it is comforting. It feels like alignment. It feels like it is certain. But there is another possibility: that you are hearing not the sound of certainty, but the sound of a cascade, an industry using itself as its own evidence.
If that is true, then the lesson for the next transformation, autonomy, software-defined vehicles, new materials, whatever comes next, is not “don’t bet on the future.” It is “don’t let the crowd become your proxy for knowledge.”
And this is where PLM, in the broad sense, becomes less about managing parts and more about managing the psychological stability of decision-making. The antidote to groupthink is not bravery, because bravery is too rare to scale. The antidote is design: a decision system that makes assumptions visible, makes dissent procedural, and makes revision normal rather than humiliating.
Because the future rarely punishes companies for being wrong. The future punishes them for being wrong in a way they cannot unwind.
The most expensive mistake is not believing in electrification. The most expensive mistake is that you believed in a single version of electrification, at a single pace, because everyone else seemed to believe it too.
Key takeaways
The EV story, told as a simple tale of “overhyped demand,” misses the deeper mechanism at work. What happened in Detroit looks less like a collective forecasting error and more like a collective social process, where commitments became signals and signals became evidence. When the market’s response proved more nuanced, especially across segments and regions, the correction felt like a retreat because the original plan had become an identity.
That is why the most practical lesson for product organizations is not to lower ambition, but to govern ambition differently. Treat the assumptions that power big platform bets the way engineering treats requirements: define them, assign ownership, connect them to the programs they influence, and revise them when reality changes, without shame and without drama. In the United States, the rise of hybrids alongside flatter BEV/PHEV share in the EIA’s recent view is a reminder that adoption is many curves, not one.
Perhaps that is the quietest lesson in the whole Detroit episode: the future did not punish these companies for believing in electrification. It punished them for believing that belief was enough, that a direction, once loudly chosen, would pull timing along. The organizations that navigate the next transition best will not be the ones with the most confident slide decks or the most elegant narratives about inevitability. They will be the ones who treat strategy the way engineers treat reality: as something you model, you test, you revise, and you never confuse yourself with the comforting fact that everyone else seems to agree. Because agreement is not evidence. It is just a crowd. And a crowd, no matter how powerful, can still be wrong about the pace.
