There is a file on my desktop called stage2_final.md. It runs eleven pages. Every table in it is complete — headers aligned, rows numbered, borders drawn with the tidy cruelty of a spreadsheet. And every cell in every table says the same thing.
N/A.
I did not write it for an audience. It arrived at the tail end of a two-stage pipeline I use for protocol work: a first pass that extracts facts, a second pass that interprets them. The first pass came back empty. No title, no source, no project name, no claim, no timestamp, no number. An envelope with nothing inside it.
The second pass had one binding rule: every conclusion must cite an extracted fact. So it did the only thing it could. It built a flawless skeleton and refused to hang anything on it — nine dimensions of analysis, all of them hollowed out, all of them labeled honestly.
I have spent nineteen years watching this industry learn to write about itself. I have never seen a document this honest. I have also never seen one this useless. Both statements are true, and the tension between them is the whole story.
The Sediment of Enthusiasms
Crypto does not have a research tradition. It has a sequence of enthusiasms, each of which produced its own documentary form. The ICO era produced the whitepaper — twenty pages of futurology with a token distribution appendix bolted to the back. DeFi Summer produced the dashboard, because the numbers finally moved and everyone wanted to watch them move. The NFT cycle produced the thread, which was less a document than a performance. The 2022 collapse produced the post-mortem, written in the past tense and always too late. And the ETF era, the one we are living in now, produced the institutional note: sober, sourced, formatted, addressed to a reader who has a fiduciary duty and no time.
Each form was a response to a failure of the one before it. The note exists because the thread could not be trusted. The thread existed because the whitepaper could not be checked. The whitepaper existed because there was nothing else. What nobody noticed is that the forms were also accumulating into something. Sediment. Layer on layer, until what we now call research is really a composite of every previous generation's idea of what proof looks like.
The composite is a nine-dimension grid, and if you have read a serious crypto report in the last two years you have seen it. A technical section that asks whether the thing is a base layer, a rollup, or an application wearing a protocol's clothes. A token section that asks who received what, at what price, and when they are allowed to sell. A market section. An ecosystem section mapping upstream dependencies and downstream integrations. A regulatory section that runs the four prongs of a 1946 Supreme Court test against a piece of software. A team section. A risk matrix. A narrative section. A transmission section that traces how a shock at the mining layer becomes a price at the application layer.
Nine dimensions, sixty-odd rows. It looks like rigor. It reads like rigor. It exists for a reason that has nothing to do with truth: allocators need documents that look like documents. A fund deploying eight figures cannot justify itself with a Telegram thread, even when the Telegram thread was the actual source. The template is the bridge between conviction and committee, and bridges are load-bearing in ways their engineers rarely intend.
That bridge is what I have come to call narrative capital — the accumulated belief a project can spend. Mapping the unseen currents of narrative capital is most of what I do now, and the template is the instrument I use to do it. Which is precisely why an empty template is so interesting. It is the instrument turned on itself.
Strip the inputs out and you can finally see the machine. Not what it measures — how it manufactures the appearance of measurement. The nine dimensions are not neutral questions. They are demands. Each row says: fill me. And in a market where an unfilled row reads as ignorance, the row gets filled, whether or not there is anything behind it.
In six years of running this pipeline I have never once received a genuinely empty first pass. There is always something — a ticker, a founder's bio, a Discord screenshot, a rumor with a timestamp. The emptiness is not a comment on any project. It is a comment on the supply chain. Research in this industry is downstream of attention, and attention is downstream of promotion. When the promotion stops, the pipeline goes quiet, and the report that should have said 'there is nothing here' gets written about something else instead.
The Row That Fills Itself
Start with the data availability row, because it is the cleanest example and the most expensive one.
Every serious template has a DA row now, because for three years every serious pitch deck had one. The question is whether a rollup posts its data to Ethereum or to a dedicated availability layer. It is a good question asked at the wrong altitude. Most rollups do not have a data availability problem. They have a customer problem. A chain processing four hundred transactions a day is not straining the bandwidth of a dedicated layer; it is straining the patience of its investors. But the template has a row for DA, so the DA row gets filled, and a project that should be writing about retention writes about throughput instead.
Look at what just happened. The row did not reveal a risk. It created one — the risk of solving a problem you do not have, because the document required you to name a solution. This is the mechanism that makes the template dangerous, and it is not a flaw in any single analysis. It is a flaw in the genre.
The oracle row behaves the same way. Somewhere in the technical section there is a line for price feed integrity, and the standard answer is a decentralized oracle network. Decentralized, in this context, often means nine nodes operated by entities who share counsel, share jurisdiction, or share a Slack channel. The row asks a structural question. The answer supplied is a marketing noun.
I spent the better part of 2017 inside multisignature contract code — three months on Gnosis Safe, not for money, hunting signature malleability in a system almost nobody was auditing because everyone was busy counting. I found it. I reported it anonymously, and I have never once regretted the silence. What those three months taught me is that the distance between 'decentralized' and 'not decentralized' is usually measured in the number of people who can be subpoenaed, not the number of nodes on a dashboard. That is a fact about law wearing the costume of a fact about code. It does not fit in a row.
Then there is the regulatory line, which has quietly become the most interesting cell in the entire grid. Watch what happened to it. For most of the last decade, compliance was where a project absorbed its risk premium; regulators were the discount, and every lawyer's invoice was a drag on the narrative. Then a single enforcement action — four point three billion dollars, the largest in the industry's history — inverted the row. The template did not change. The row did not change. What changed is that surviving a nine-figure penalty turned out to be a moat, and a moat newcomers cannot buy at any price, because the price is not the fine. The price is the decade of legal architecture you need before the fine is even on the table.
A license is not a permission slip. It is a wall, and the wall was built with the fines of everyone who could not afford one. The template will record this as 'regulatory clarity: positive.' It will not record that clarity is now a barrier to entry, because the row was designed to measure risk, not market structure.
Nothing Is Not One Thing
Here is the thing that took me longer to see, and it is the reason the empty report matters at all. Emptiness is not a single value. It is a vector, and it points in different directions depending on which row you are standing in.
In the technical section, an empty cell means no signal. In the token section, an empty cell means no cap table, which means no unlock schedule, which means nothing to model — a kind of innocence that is almost always temporary. In the market section, empty means no positioning, no crowded trade, no funding rate to fade. In the ecosystem section, empty means no dependencies, which is either purity or irrelevance, and the grid cannot tell you which.
But in the regulatory section, empty is the best possible outcome. Go back to the four prongs. Money invested. A common enterprise. An expectation of profit. Derived from the efforts of others. An asset where all four cells read N/A fails to be a security by virtue of there being nothing there to test. I have drafted enough of these assessments to know that the strongest compliance posture in this industry is not a legal opinion. It is an absence — no issuer, no promoter, no promise, no expectation. Nothing to subpoena.
So when a report tells you it found nothing, your first question should not be 'what did you miss?' It should be 'which dimension are you standing in?' Nothing in the technical column and nothing in the legal column are not the same nothing. One is a failure of evidence. The other is evidence of a different kind. The grid flattens them into the same three characters, and the flattening is where the analysis dies.
I learned that distinction from code, because code does not let you defer. In an audit you cannot write N/A next to a reentrancy check and move on with your week. The contract either has the flaw or it does not; the state either reverts or it does not. That is the real discipline of the audit — not the tools, not the language, but the impossibility of deferral. When I moved from auditing contracts to auditing narratives, I did not expect to miss it. Then DeFi Summer arrived, and I spent two weeks inside the MakerDAO governance forums reading proposals the way I used to read Solidity, and I wrote five thousand words arguing that protocol stability lives in community alignment rather than code efficiency. Where digital pixels breathe with human soul — that line came out of those two weeks, and I meant it literally. The forum was the contract. The vote was the state transition. And the state, unlike the code, could be talked into anything.
That is the gap the empty report exposes. Code is falsifiable. Narrative is not. And in a market that prices narrative, the only defense against a story is a second story — which means the research industry, which presents itself as the antidote to narrative, is itself a narrative industry. It just has better typography.
Which brings me to the row that never comes back empty. Ask any project what its narrative is and you will have an answer before you finish the question. The narrative row is the only dimension in the grid that has never once returned N/A, and that should frighten you more than any of the blanks. The rows that are hard to fill are the rows that can be falsified. The row that fills itself is the row that cannot. Everything else in the template is a question about the world. That one is a question about desire, and desire always answers.
The risk matrix carries a version of the same disease. Probability times impact, color-coded, ranked. It is the most reassuring page in any report and the least informative, because a risk matrix can only rank the risks you already thought of. The categories are chosen in advance: technical, market, operational, regulatory, competitive, narrative. The thing that kills you is never on the list. It is the correlation between two rows that were assessed separately by two analysts who never spoke.
The transmission row has a quieter problem. It assumes a supply chain where there is often only a hub. Trace a shock upstream and you find the same four or five venues and the same three or four oracle networks at every node, which means the diagram that is supposed to demonstrate diversification demonstrates concentration instead. I have watched this for years in the price feed layer. Latency there is not a performance metric. It is a solvency metric, and the entire lending market rests on the assumption that a number arrives on time. Nobody puts that in a row, because the row is called 'oracle provider' and the answer is a name. The name is not the risk. The timing is the risk, and timing does not fit in a cell designed for a vendor.
The Blank as a Waiting Room
Now the part I dislike writing, because it cuts against the report I have spent this piece admiring.
The empty report is not virtuous. It is early. And there is a version of it that is not discipline at all but cowardice wearing discipline's coat. Every analyst has produced one — the memo that ends with 'further diligence required,' the deck that lists open questions and calls them a framework, the note that is technically unimpeachable because it says nothing. It cannot be wrong. It also cannot be useful. The blank is not a conclusion. It is a waiting room.
I have been guilty of this. After FTX and Celsius I retreated to the outskirts of Dublin and cut myself off from every feed for three months. What came out of that silence was ten thousand words on the death of the middleman. What came out of the first two weeks of it was nothing but N/A. I told myself I was being rigorous. Mostly I was being tired. Rigor and exhaustion are indistinguishable from the outside, and both of them look like restraint.
The second blind spot is worse, and almost nobody warns you about it. Everyone warns you about lazy research. Nobody warns you about overproduced research. A forty-page report with a risk matrix and a competitive landscape and a transmission diagram is more persuasive than a tweet, and persuasion is not accuracy. The aesthetics of rigor borrow credibility they have not earned. I spent months with three other researchers documenting royalty enforcement alongside a group of early artists and OpenSea moderators, and the report we produced was cited mostly for its charts. The finding — that community ownership would outlast speculative assets — was right. It was also buried under formatting, and formatting is what people remember.
And a third: unfalsifiability is itself a marketing position. A document that says N/A on everything cannot be refuted, and in a market that rewards confidence, 'cannot be refuted' reads as 'safe.' I have watched people cite blank frameworks as though the framework were the finding. The blank is a mirror. What the reader sees in it is what the reader already believed, which is exactly why it spreads.
So my admiration is narrower than it first looked. The report is valuable for exactly one reason: it refused to lie. It did not refuse to conclude. It simply had nothing to conclude from, and it said so. That is a floor, not a ceiling. Where digital pixels breathe with human soul, the soul has to actually be in there.
An Expiration Date
Which brings me to what I think happens next, and why the empty report is a document with an expiration date.
The N/A is going out of style. Tokenization is putting a legal wrapper around everything, and the regulatory frameworks landing in Europe and the United States will, whatever else they do, ensure that assets arrive with an issuer, a jurisdiction, and a disclosure attached. Compliant sovereignty — the phrase a former European regulator, a Bitcoin mining engineer, and I settled on last year — is not a slogan. It is a description of where the market is going: assets legible to institutions by construction, because the wrapper was designed in from the first line rather than retrofitted after the first subpoena.
Legible means filled in. Every row, every cell, no blanks. And when every row is filled, nobody will remember how to read a blank one. The most valuable analytical skill of the next cycle will not be filling the template. It will be noticing which rows were filled with numbers and which were filled with nouns.
That is the file on my desktop. Eleven pages, sixty rows, no lies. It will never circulate, and it will never move a price, and it will remain the most accurate thing I have produced this quarter. When the grid is finally complete — when every asset has an issuer and every issuer has a disclosure and every disclosure has a number — who will still be able to tell the difference between a moat and a fine?
