I just spent an hour and a half in a delightful conversation with the shadowy figures who run publishing’s query system and who, by deciding which authors get the few literary agents who can place books, determine what everybody reads. No, I won’t tell you who they are. The first password is, well, you can find it online. Okay, it’s “Rosebud.” Never Google that. The second password I shall withhold, because they’ll kill me if I tell you. (Gurer vf ab frpbaq cnffjbeq.) It seems I have caused quite an annoyance by leaking the $25,000 price tag of a fair read, and they’ve decided to raise their rates. It now costs $35,000 to query literary agents. Move fast, before it goes to fifty. Big Query knows the poors are on the march, and the gates are closing.
I’m joking… probably. No one deliberately intended to build this system, and the $25,000 figure is an amortized cost derived from an observed market rate, not a literal ticket price. The query-industrial complex isn’t a conspiracy. It’s the natural consequence of a situation where there are thousands of desperate writers seeking traditional rescue—self-publishing is difficult, time-consuming, and often expensive—and far fewer (for lack of a better word) gatekeepers who are in the position to get books read.
There aren’t a lot of slots in traditional publishing. There are a lot of writers who want them. The number who write at a level I would consider peer to my own is small. But the number who can write well enough to meet traditional publishing’s current standard is not small. Supply, demand. It’s not worth being angry about.
Alyssa Matesic took exception to some of my findings about the publishing industry, which she called…
(0:35) Complete nonsense.
Well, that’s one less blurb I need when my novel comes out.
I’ve been misconstrued enough times that I ought to review my findings and, to eliminate misconceptions, clarify them.
The $25,000 Cost Is Amortized, Not Absolute
Some people are introduced to literary agents by their family publicists. Some people end up tutoring agents’ kids and get lead-title book deals in exchange for buying fake IDs and sitting through the same exams they took in high school. Some people cite obscure comps in queries that would ordinarily have gone unread like all the other queries, until a film adaptation is announced, an event that causes every query naming it before the film deal to get read. I won’t claim that these things never happen. I won’t claim that it’s physically impossible to sign an agent without expense. It’s extremely unlikely. The first law of probability is: You ain’t special.
No, that’s a lie. Probability is applied measure theory; we’d have to get into σ-algebras (and, therefore, set theory) to discuss what probability truly is. Because it’s a branch of mathematics, it has no opinion either way on whether you are special, and so there is no “first law of probability” to such effect. See, this is why I could never work in publishing. I’m perfectly capable of saying things that aren’t true. What I can’t make myself do is let them stand for long enough to form a useful deception.
You probably aren’t special. You probably are no exception to the 1-in-292,201,338 chance of buying a winning ticket for the next Powerball drawing. You might also buy one, and win, and it will not prove me wrong. If people who query for free have a 0.05% chance per annum of being fairly read, then some people will get lucky in the first week, because the world is big, but the median person using this strategy will spend… 1400 years. That is, for most of us, a professionally unacceptable delay. What are the precise numbers, when it comes to slush-pile rescues? We don’t know. We don’t know because they are so infrequent, the rate is basically unmeasurable.
The way you win in publishing is to build the social status or leverage not to rely on a slush pile in the first place.
There’s no formal process by which you can spend exactly $25,000 and receive one (1) fair read by a literary agent. It would do you harm to believe one does. Buying access is inherently stochastic, which makes it infuriating for those who must, because the other side always has a plausible justification (“it turns out we’re not supposed to be doing this”) not to deliver. No one puts referral editor on a business card. And yet we know that this is a common practice. Who pays $0.12/word for a developmental edit, when the benefit of referral is heavily implied but never guaranteed?
Not career self-publishers, who are trying to run a sustainable business and cannot afford speculative expenses.
Not traditionally published authors in good standing, who don’t need to buy the service online.
Not traditionally published authors in poor standing, who already know that even a famous editor can’t fix a bad sales history.
Not hobbyists, for whom this would be an unjustifiable expense.
Who does this leave? The sorts of people who think every book deal involves a six- or seven-figure advance. The sorts of people who think better plotting or better sentences can make up for a lack of the leverage and social status that publishing actually cares about.
We observe that a 1% chance of a fair read by a literary agent seems to be valued, as of March 2026, at about $250. If a conference can deliver a 10% chance of getting a fair read, its ticket price will go up by about $2,500. If an opportunity were found that delivered fair reads more cheaply, it would quickly become congested. Either its price would converge to the standing level, or it would deteriorate through crowding and neglect. Similarly, a venue where the price is too high will be abandoned. This is simply how markets work. And prices fluctuate. Querying is a stochastic toll road; you don’t get to know how much it will cost until you have used it and you find out.
Matesic has said some people exist who have queried for free and succeeded; she’s probably not wrong, even if her incentives force her to overstate the frequency. Survivorship bias is also a factor. People who query cheaply and hit will announce themselves, and will also round their costs down due to benign confabulation, differential affect biases, and the need to appease an industry that insists on being called a meritocracy despite refusing to do the work that one would require. Those who sink $50,000 into querying and get nothing—statistically, we expect a 13.4% failure rate at this expense level—are unlikely to say anything.
As for me, I’m not going to sell you a course, for the low, low price of $24,000 $5,000 $399, that contains the hidden incantations that force literary agents to read, because said sorceries do not exist. You need leverage or social status. That’s it. There are no shortcuts. You can buy them, or you can try to earn them. Leverage isn’t usually earnable; it either happens or it doesn’t. Social status is even less so, because social status’s vacuity is both the source of its power and its entire point, although it is often bought, with one caveat: methods of obvious buying tend to lose effectiveness over time.
I never said, as Matesic claims, that “everyone who has signed with a literary agent has paid.” This is an utter misrepresentation of the argument I have made.
I Didn’t Prove That No Literary Agent Has Ever Read; I Proved They Basically Never Read for Quality
The claim that no literary agent has ever read a query letter or manuscript in good faith is almost certainly false. Even if it were true, it would be impossible to prove.
The July Study has been discussed at length. I’ve already gone over why a query letter would have been the wrong instrument. When you’re conducting a study, you shouldn’t also be trying to get something out of the people you’re studying. That should be obvious. To keep it as scientific as possible, I counted engaged responses—not positive responses, not whether I got what I wanted. Also, it is impossible for a query letter to show significant writing talent, because optimal queries can be written by AI; agents have banned the practice for that reason. It should surprise nobody that soulless machines have mastered a soulless format.
What I have proven is that 200 of the top literary agents, even when exposed to better writing than most of them have on their client lists, will… not even know. Or care. You can write whatever you want to them; you’ll get the same response. To eliminate common alternative explanations, I used a large sample and extremely conservative parameters, putting the null hypothesis at an obscenely low (compared to the claims literary agents make about themselves and their attention to quality) engaged-response rate of 3.15%. To remove any accusation of emotional bias, I used AI to double-check my grading of responses. The final count was zero.
I did not prove that no literary agent has ever, in all of human history, read in good faith. I simply found, in the largest-ever signs-of-life study this industry has ever seen, no such signs. Even if you’ve invested the time it takes to write as well as I do, you will not be able to write your way into the social status required to get fairly read by a literary agent. What if you’re a better writer than I am? Well, good for you, but the bad news I have is that we are both well above the upper threshold of any literary agent’s instruments, which means we’re basically on the same footing, and so the conclusion does not change.
I have been proven wrong on only one front. The study has generated discussion, including by literary agents. It turns out it is possible to get them to read. The way you do it is you tell a few thousand people that agents don’t read. Then they will read. The career value of this is negative, and it has already been done, but it does pass the test.
I Have Never Advised Spending Tens of Thousands of Dollars to Get Published
Matesic says, “This article’s advice to spend tens of thousands of dollars on an MFA to get your foot in the door with an agent is completely nonsensical.” I never gave such advice to anybody. I have always been candid about the costs and risks of every strategy I’ve discussed. Not everyone who attends an MFA program gets an agent through a professor; the ones who don’t will have to explain how they ended up in the bread line of the query system. Also, not everyone who gets a book deal gets a good book deal. The median offer from a traditional publisher is probably worth less than what querying costs. Don’t spend $25—let alone $25,000—on this stuff if you can’t afford the loss.
I would usually advise a literary author with low opportunity costs (i.e., the median college graduate) to take a fully-funded MFA at a top program, if a slot becomes available. In that particular case, it’s probably a good idea to attend. It’s not a good idea for everyone. You should probably not exit a successful career as, for example, a doctor or a lawyer, to pursue a degree that offers no guarantees. You should probably not pay tuition, unless you’re absolutely certain you can afford it—definitely do not go into debt. You should probably only consider the top five programs if your goal is traditional publishing, since most programs have very few agent referrals to hand out.
I doubt I am breaking news by telling anyone that the economic inequality in this country is absolutely insane. There are people for whom it is worth far more than $25,000 to be able to say that they are “bestselling novelists” even while losing money due to querying costs, self-funded book tours, publicists’ retainers, and own-buy operations. There are corporate SVPs using the external publicity a book confers to make EVP. There are motivational speakers whose books are loss leaders. There are wealthy housewives rejoining competitive fields who need to account for time. There are trust-fund kids who’ll be cut off by their daddies if they can’t show that they’re doing something, anything. People exist for whom it makes sense to lose tens or even hundreds of thousands of dollars on shitty books. In the market for fair reads by literary agents, you are competing with them. Sorry.
Spend zero dollars that you cannot afford to lose. And if you are financially fortunate enough that you can lose $25,000 and not mind the blow, then you should know that we live in entirely different worlds, and advice is mutually meaningless between us.
Alyssa Matesic’s Statistics Don’t Tell the Real Story
Matesic shows a pie chart (6:08) indicating that, in a survey of 145 literary agents, 60% claimed their average signed author has no writing experience. Let me explain why that’s not relevant. And here’s the part I don’t like writing about, because I feel like a bit of a dick in pointing it out, but it needs to be said. Most literary agents, in the context of one of the most hierarchical industries on Earth, do not matter. They might be fantastic at their jobs. They might be wonderful people. They might be exceptionally insightful readers. They might even—this would be negatively correlated to success in their industry, but I am sure it happens—have good taste. None of that means anything. Can they get editors to read? Can they get them to read quickly? Can they create concurrent interest so there is leverage and a good deal is actually possible?
There are probably ten thousand people calling themselves literary agents, and most of them can probably get a manuscript into a submission queue somewhere, but that’s not what counts. A smaller slush pile is still a slush pile. The percentage of them who can get pages fairly read by an editor is extremely low. This is a huge mess, ethically speaking. Agents are supposed to represent authors’ interests, but an agent who loses an author can get another one in an afternoon, whereas one who loses the ability to force editors to read is professionally dead. Therefore, when interests diverge, which side is favored? The publisher’s, obviously.
Laura McGrath found that half of all award-winning novels went through 25 literary agents. Publishers can’t unilaterally decide what wins awards, of course, but they determine what is noticed enough for juries to read. No book deal or positioning can guarantee awards, but bad deals and bad positioning can preclude them. And half the award-capable slots go through 25 people. That’s less than a subway car. I’ve taught classes larger than that. You do not get these agents by querying—they’ve heard sobbier sob stories than anything you can come up with, unless you’ve had a truly miserable life. The only way to get them is through your family publicist—don’t have one? neither do I; I guess publishing isn’t for you—or through some complex extortion I’d rather not develop because, honestly, for all the effort such a campaign would require, it’d be far better to do anything else. A normal crime would be easier and make more money. Getting a job would be easier and make more money.
I am sure there are literary agents out there who do read manuscripts in good faith, and who truly want the best books to win. They’re the ones who’ve just started. They don’t matter in publishing yet, because they don’t have any social capital, and if they stick with publishing long enough to accumulate any, they’ll turn into what all the rest of them have had to become—cynical, transactional professionals—in order to survive. There are people who enter publishing because they love books. If they still love books, it means they never got far enough into the industry to hear the conversations in which real decisions are made.
It feels shitty to make a “those people don’t matter” argument. As human beings, of course they matter. To publishing, most of them don’t—only a few dozen agents do. I also don’t matter in publishing, at all, and I write better than the vast majority of people in that industry have ever seen in the wild, a fact that makes me matter even less—less than zero—because it pisses them off when they find out.
Matesic says, at 7:12, that conferences “are not a fast track to getting an agent.” On this, I agree. The data bear that out. An upper bound for a $750 conference is a 3% chance of getting read by a serious agent; if you’re bad at tactics, or choose the wrong conference, it can be lower than that. Considering travel costs and time investment, you should probably only go to conferences if you actually want to go to those conferences.
The easiest, cheapest way to get a literary agent is probably to buy followers on social media. Be careful, though, because inexpensive bought followers tend to deliver terrible engagement, and that’ll do more harm than good. You’ll need to plan out your growth curve, because a spike will get you caught and, on some platforms, banned. And I can’t give precise advice on which services to use because I’ve never done it; it feels vain and mildly unethical. If you don’t get caught, it will almost certainly get you an agent, but you can still lose money.
I’m Not Trying to Discourage You
Often, after I write these real-talk essays on publishing, I’m accused of trying to discourage other writers.
You’ve got me. I want all the query letters to myself. I send out thirty thousand per day already—I’m the reason why everything sucks—but that’s not enough of the pie for me. I want 100 percent of it. I’ve already hired a harem guard for all the literary agents I’m planning to sign, because once I’ve cleared away the competition, they’ll have only me. Phase 1 of my nefarious plan was to use my massive platform to discourage all the querying writers in the world, and I’ve clearly done that because nobody is querying anymore; in Phase 2, I’ll collect every last literary agent. I even have a golden toilet for Andrew Wylie. I don’t intend to use any of them to publish anything—why would I?—and whatever I force the agents in my strictly nonsexual harem to read will probably not even be my book.
In all seriousness, I think you should do what you want to do. If you want to write, write. If you want to read, read—we need readers more than we need writers. If you want to tell your 179,000 YouTube subscribers that I’m an asshole, then you have my blessing to do that. You have one life. It’s not my place to tell you how to use it.
I enjoy writing. I even enjoy the revision process. I’m doing final edits (and have been “doing final edits” for longer than I care to admit) on a novel, and I don’t mind the work itself, but I doubt I would ever have started a book if I’d known what a fucking nightmare it is—whether you self-publish or use a traditional publisher, it’s truly awful—to get one discovered, and almost every honest writer will tell you the exact same thing. The really good books exist because people make mistakes. I say this not because it is always a mistake to write, but because it is costly to write that much and that well unless you already have institutional support, and institutional support is basically never given to people who have anything interesting to say.
Whether to use a traditional publisher is a business decision, no less and no more. You are not a better writer, nor a worse one, because you sign with Random House instead of rolling the dice as a self-publisher. I won’t refuse to talk to you if you sign a trade deal; if we were friends before you did so, we’ll be friends after you do it. Some traditional deals are worth taking. Some are not. Make sure that an actual lawyer—not your agent, who (as already discussed) relies on relationships with publishers for her career, and is therefore forced to favor the other side’s interests—reviews the contract before you sign it, but don’t be afraid to take an opportunity if one comes.
The news I bring is truly not that dismal. For most of human history, a newborn had about a fifty-fifty chance of reaching age ten. Billions of people have lived who never had the opportunity to own or read books, let alone write them. There are more than a dozen wars going on right now. The fact that people in traditional publishing are bad at their jobs is, as far as social calamities go, quite minor. It is true, as I have shown, that text does not matter in publishing. It will not be rewarded; it has not been for decades. It still matters to readers. Find them. How? If I knew…
Conclusion
The last thing I’ll say is that I have nothing against Alyssa Matesic. Since the world is big, and I can’t know in advance who will read what I write, I must say this: for the love of God, don’t bother her in any way. She is doing a job, just as I am doing a job.
She says her book is $21. Should you buy it? If you intend on querying literary agents, and you believe it’ll improve your chances of getting a fair read by more than 0.084 percentage points to do so, expected value says yes.
I have the equivalent of an MFA from a school that was hard to get into. I worked for years in intelligence. I've had short fiction published in journals that are hard to get into, and I've even won an award or two. I wrote a novel about working in intelligence, but it's not a spy thriller. It's a literary fiction/satirical novel. With so many novels that don't sell for shit, I figured someone would think this combination of proven writing chops and new material would be worth looking at. I barely got a response from dozens and dozens of queries. I finally paid to go to a writing conference, where I pitched it to three carefully chosen agents. All three said they were impressed with the pitch and wanted me to send it to them. One never responded. One said three times he was busy but would eventually get back to me but never did. One read a little bit and said it wasn't for her and gave a boilerplate reason. I may not have written Gravity's Rainbow, but I think what I've got is at least different from every other book and might have had a better chance of success than a lot of what gets traditionally published. I finally just self-published it as a way of forgetting about it, because nothing makes sense.
Supply and demand. Everybody's writing a book, and there are fewer and fewer readers of books.
If you are average, if you write in a specific genre, you will be lost in the crowd. And it's a very large crowd.
Don't go large and get lost in the crowd. Go small and local. Develop your own personal network and work from there. Yes, I said work.