The Analyst Was Never the Market

Why the demand for analysis always dwarfed the demand for analysts — and what that means for where this category actually grows

The reflexive question about every analytics tool that can answer questions in plain language is what it means for the analyst. It is asked with real concern, and it is the wrong frame for anyone trying to understand the market underneath. “Will this replace the analyst?” treats the analyst seat as the thing being bought and sold — as if the size of the opportunity were the number of analysts employed, and the only question were whether that number goes up or down. But the analyst seat was never the market. It was the small, visible, affordable tip of a demand curve that has always been mostly underwater.

Start with the gap between two numbers that are almost never compared. One is the demand for analysts — the count of companies that looked at their need for answers, decided it was large and steady enough to justify a salary, and hired someone to own it. The other is the demand for analysis — the total volume of questions worth answering across every company that has data, whether or not anyone was ever paid to answer them. The first number is a hiring statistic. The second is the actual market. And the first is a rounding error against the second, because hiring an analyst was a high bar that most teams, and most companies, never cleared.

They never cleared it for an ordinary economic reason. An analyst is expensive, and expensive capabilities get rationed. A company weighs the steady stream of questions it would like answered against the cost of a person to answer them, and unless that stream is large and reliable, the math says don’t hire — make do, guess, decide on intuition. So the questions do not go away; they go unpriced. They sit in the heads of founders and operators and product leads at the millions of companies that have data and no analyst, and they never become demand that anyone counts, because demand only gets counted when it converts into a purchase. The analyst seat is what demand looks like after it has survived a brutal affordability filter. Almost all of it does not survive.

This is the inversion the replacement framing hides. The market was never the supply of analysts; it was the suppressed demand for what analysts do — a demand that was vast, real, and almost entirely unmonetized, because the only way to satisfy it was a hire most buyers could not justify. When the cost of answering a question collapses, the interesting thing is not what happens to the analysts who were hired. It is what happens to all the demand that was sitting below the affordability line, waiting for a price low enough to surface. That demand was always the market. It just never had a product cheap enough to convert it.

Notice how differently this reads from the incumbent’s view of the same category. Established analytics tools price against the company that already has an analyst: per seat, per event, per tracked user, sold into a team that has already decided analysis is worth a budget line. That is a sensible way to sell to the population above the affordability line — the companies that already cleared the bar. But it is, by construction, competition for demand that already exists. Everyone in that market is fighting over the same buyers who were always going to spend. The metric you price on goes up when the customer succeeds, which means the customer learns to ration the very visibility they are paying for — and the un-analyzed majority below the line stays exactly where it was, priced out, uncounted, unserved.

The larger opportunity is the opposite company. Question-rich and analyst-poor: plenty worth knowing, no one to ask, and no realistic path to a hire. There are vastly more companies shaped like this than there are companies with a staffed analytics function, and between them they hold the majority of the demand for analysis that has never once shown up in a vendor’s pipeline. Reaching them is not a discount play on the existing market. It is a different market — one that only exists at a price beneath the threshold where analysis was ever worth it before. A product that lands under that threshold is not selling a cheaper version of the analyst seat. It is converting demand that the seat could never have reached, because the seat was the thing pricing that demand out.

This is the point at which someone fairly asks whether all of this is just a polite way of saying the analyst’s job goes away — and it is worth answering directly, because the answer is no, and the reason is structural rather than reassuring. There are two populations, and the function behaves differently in each. In the companies that already employ analysts, the routine conversion of questions into answers stops needing a dedicated queue, and the analyst climbs — toward the genuinely hard investigations, the shared data model everyone now depends on, the questions that are difficult because the framing is contested rather than because the query is hard. That is augmentation, the same way spreadsheets moved controllers from arithmetic to finance. In the far larger population of companies that never had an analyst, there is no seat to displace. The function simply becomes available for the first time. Across both, the labor pool is not shrunk; it is freed in one place and newly served in the other. Undermining analysts is not the mechanism here — and it would be a strange strategy, given that analysts are among the most fluent and enthusiastic users of a tool that finally lets them stop being the queue.

Hold the two populations together and the shape of the market comes into focus. The category is not “BI seats” and its ceiling is not the analytics budget of the companies that already buy analytics. The category is unmet demand for the analyst’s function, and its ceiling is closer to the number of teams that have questions and no one to ask — a far larger figure, most of which has never been addressable at any price an incumbent could offer without unwinding its own model. That reframes what expansion looks like. You do not start in the enterprise and discount your way down to the rest; you start where the economics already work — the un-analyzed majority that converts the moment the price is low enough to clear the bar they could never clear — establish the base there, and climb into the analyst-equipped enterprise later, where you arrive selling augmentation to a function that is delighted to be augmented, not replacement to a function braced to resist it.

There is a deeper reason this market stayed invisible for so long, and it is the same reason it is surfacing now. For the entire history of analytics software, satisfying the demand required a person, because the systems holding the data could not reason about it — someone had to translate a question into instructions a machine could run, and that someone was expensive, and expense is what kept the demand underwater. The thing that changes is not that the questions got more numerous or more valuable. It is that the translation stopped requiring a salary. Once a system can take a question in plain language and reason its way to an answer the way a capable analyst would, the affordability filter that suppressed most of the demand simply lifts — and the demand that was always there, uncounted, begins to convert. The market was not created by the tool. It was uncovered by it.

Step back, and the pattern is the oldest one in software: a capability that was a department becomes a feature of the whole economy, and the market turns out to be far larger than the headcount that used to deliver it. The demand to reach someone was never the number of operators working the switchboard; it was everyone who ever wanted to place a call. The demand to keep a record of a moment was never the number of professional photographers. In each case the real demand was the vast, suppressed appetite of everyone who could not previously afford the capability — and it converted the moment the price fell through the floor that had been holding it back. The analyst was never the market either. The unmet demand for what the analyst did always was, and it is finally cheap enough to count.

Key Takeaways

• The demand for analysts is a hiring statistic; the demand for analysis is the actual market — and the first is a rounding error against the second.
• The analyst seat is demand that survived a brutal affordability filter. Almost all of it didn’t survive — it sits, unpriced, in companies that have data and no one to ask.
• Incumbents price against the company that already has an analyst. That’s competition for demand that already exists; the un-analyzed majority stays priced out.
• A price beneath the threshold where analysis was ever worth it doesn’t discount the old market — it converts a new one.
• This doesn’t undermine analysts: where they exist, the function climbs them upward; where they don’t, there was never a seat to lose. The market was always the unmet demand, not the headcount.