The published forecast is not wrong because it lies. It is wrong because it measures intention instead of budget.
A capacity expansion decision worth $47M does not get built on a hunch. It gets built on a forecast — a published timeline showing when a market will be ready to absorb a new product, a new plant line, a new segment of supply. The executive sponsoring that decision reads the forecast, sees "24 months to mainstream adoption," and treats it as a fact about the world. It is not a fact about the world. It is a fact about what respondents said they intended to do, aggregated into a curve that looks precise enough to bet a board decision on.
That is the assumption this argument challenges: that a stated-intent forecast and the behavior of the people who actually control budget are the same thing, just expressed at different levels of granularity. They are not the same thing. They diverge, and the divergence is not random noise around a correct number — it runs in one direction, and it runs long.
What "24 months" actually meant
One company came within six weeks of committing $47M to expand production capacity for a new market. Their published forecast put mainstream adoption 24 months out — comfortably inside the window that would justify the capital outlay. Before final sign-off, the board commissioned a 14-day primary research mandate: not another survey of stated intent, but direct contact with the people who would actually have to release budget for the product to be adopted.
The mandate reached 34 verified procurement decision-makers inside the target application segment. The number who had allocated current-cycle budget for the technology was zero.
This is the point that gets lost when a forecast is treated as a single number rather than as a claim about a specific population's behavior: the adoption itself was not in dispute. The technology was real, the market interest was real, the eventual demand was almost certainly going to materialize close to where the forecast said it would. What was wrong was the clock. The timeline was off by 28 months — not because the forecast's authors were careless, but because a forecast built from stated intent has no mechanism for detecting whether the budget cycle behind that intent has actually opened. Zero of 34 is not "adoption is slower than expected." It is "the purchasing infrastructure required for adoption does not yet exist in this segment," which is a different fact, arrived at through a different method, and invisible to a survey instrument that asks people what they plan to do rather than what they have already funded.
The company did not commit the $47M. The research that changed the decision cost $42,000 and took two weeks.
The variable executives are not measuring
Executive strategy processes are generally well built for one kind of question — is this market real — and poorly built for another: is this market's buying committee currently positioned to act. Published forecasts, industry reports, and stated-intent surveys are optimized to answer the first question. They aggregate opinion, sentiment, and forward-looking statements from a wide population, and they do it well. But a board authorizing $47M in capital is not really asking "will this market exist." It is asking "will the specific decision-makers who control budget in this segment release that budget inside the window my capital plan assumes." Those are different questions, and the second one cannot be answered by widening the survey sample — it can only be answered by asking the narrower, verified population that actually holds the checkbook whether the budget line exists yet.
This is the variable that determines outcome and is rarely the variable being measured: not sentiment, not stated intent, not aggregate market size, but current-cycle budget allocation among the specific decision-makers who would have to say yes. A market can be unambiguously real and unambiguously not-yet-fundable in the window a capital plan requires, and no amount of additional stated-intent data resolves that gap, because stated intent and allocated budget are not points on the same curve. They are two different signals that happen to get reported using the same language of "adoption."
Why the gap runs long, not short
It is worth naming a pattern rather than treating this as an isolated case: when a stated-intent forecast diverges from verified decision-maker behavior, the divergence tends to run in the direction of overestimating readiness, not underestimating it. Respondents answering a general market survey have every incentive to describe their organization as forward-leaning and no mechanism forcing them to disclose where their own budget cycle actually stands. A procurement decision-maker asked directly and confidentially whether current-cycle budget exists for a specific technology has no such incentive to overstate readiness — the question is narrow enough that a false "yes" is easy to catch and costly to have made. That asymmetry is why the correction, when it happens, tends to push timelines out rather than in. The 28-month gap in this case is not an anomaly to be explained away; it is what happens when a wide, low-friction survey instrument is compared against a narrow, verified, budget-specific one.
The decision this changes
None of this argues against forecasting, and none of it argues that stated-intent research is worthless — it answers the question it is built to answer. What it argues is that the size of the capital commitment should determine the specificity of the evidence behind it. A decision that can absorb being wrong by a few months can reasonably run on a published forecast. A decision that cannot survive being wrong by more than two years cannot. The $47M in this case was not protected by a better forecast. It was protected by a narrower, faster, verified check against the specific population whose behavior — not whose stated intent — actually determines when the money would have been safe to spend.
The question worth asking before the next capital commitment is not "what does the forecast say." It is "has anyone verified, this cycle, whether the people who would have to fund this have actually funded it yet." Fourteen days and a fraction of a percent of the capital at risk is what it costs to find out before the board vote instead of after.