Why on Earth would a literary agent possibly respond to this "antiquery"? An agent isn't reading their slush pile to find good writing generally, or engaging literary talent; they're trying to find *finished books* that they can sell to publishers. What you sent them is, by its size and nature alone, instantly recognizable as not one of those, so I would 100% expect it to go right in the trash. (Not because it's low QUALITY, I can't speak to that! But because it's simply not what they're looking for.)
You have to look at the process from the agent's point of view. (Try reading slush sometime; I have, and it's enlightening. Lots of places take volunteers.) The whole system -- guidelines, query letter, partials, etc -- is designed to absolutely minimize the amount of time the agent spends on any given submission, not because the agent's time is so precious in an absolute sense but because the volume of submissions (for a successful agent) is ENORMOUS. So it's all triage.
Didn't follow the guidelines? Usually an automatic pass; maybe there's huge literary merit there, but why take the time to read and find out when there's ten thousand queries that *can* follow instructions? Then the query letter isn't an opportunity to demonstrate authorial prowess, it's a filter for cranks, psychos, and no-hopers. Again, having been on the other side, you would not believe how many submissions fall into this category: conspiracy rants, borderline terrifying sex fantasies, Star Wars fanfic without the names even changed, and so on.
Even once you cut out all of that there's still more incoming slush then it is physically possible to read. The standard query letter format is designed to present the salient facts about the piece to the agent as succinctly as possible: genre, length, logline, style. Typically agents aren't just looking for "anything good"; their job is to know what editors want to buy, and curate their own list accordingly. So a given agent in a given month might be glutted with space operas but eager for a fantasy romance, or vice versa, and the query gives an easy way to sort them out.
Only then, after all these filters, comes the much more time-consuming step of figuring out which submissions are *actually good*.
Anyway, this is probably too long a comment already. Suffice to say that the results of your experiment are exactly what anyone who has some experience with the system would have told you to expect. In counterpoint, I know quite a few people who have sold books to agents on query and many more who have made it partway and gotten personalized responses, so there's quite a bit of reading going on.
My experiment did not prove that agents never read anything (not true) or that they are unethical (not true) or that they do not care, in the abstract, about literary quality—they probably do.
It proved that they don't read for quality—that quality writing cannot overcome the low social status of having to use a query letter in the first place. You are correct; this probably should not have been surprising.
You have basically argued my case—that there's too much volume, and that there's no level of quality at which the writing says, "I'm not volume." So why play? Why beg, on the off chance that an agent's heart will grow three sizes that day? Why dance like a monkey when it will probably amount to nothing—even if you get "a book deal" it will probably be trash.
Querying is a freemium product. The free tier is the glitchy version that mostly exists to piss people off and prime them for the upsell. The pricing tier at which you become a human instead of "volume" is where you find MFA programs, expensive conferences, and "manuscript assessments" from famous editors that may lead to referral. This is basically vanity press, though, and it's often a bad financial play—to spend $25,000 querying and get a $5,000 book deal? No financial advisor would recommend that.
I am not arguing that literary agents are bad people. I don't think they are. I'm arguing that the system is garbage and should be torn down so it can be replaced with something that actually works.
Also, it's their fault they are inundated with low-quality submissions. I'm not the one who spreads false hope.
The system certainly has its flaws! But the fundamental problem it addresses is just *really hard*. There are many, many, many more people who want to publish books then there are books that can be profitably published, and sorting out the best of them without spending an absurd amount of time and money is extremely difficult. Any system to do it will have flaws.
You're too hung up on the query letter, though. Query letters are not as important as you seem to think. Typically all you can expect of a query letter is to confirm to an agent that yes, this person is sane and wrote a book and yes, it's in a genre/style that they're looking for. The sample chapters have to do the rest of the work -- if anything should be obsessed over, it's those.
It follows that "query letter clinics" and "manuscript assessments" are mostly scams. Any that cost $25,000 are *definitely* scams, as is any famous editor who charges for possible referrals. They generally will not increase your chance of getting accepted. I 100% agree with you that this is bullshit, but people love to prey on desperate would-be writers. But it's not an unlockable paid fast track; they can't deliver what they promise. (MFAs and conferences are a different kettle of fish, kind of a mixed bag.)
That's fair. My issue with the query system is that we no longer need it at today's technological level.
Consider AI. Can it read for quality as well as a skilled human? Hell no. Can it provide a better signal, for the first-pass determination of whether a submitter is worth an agent's time at all, than the query letter? Easily. So let's replace the query letter and the social proof factors with something flawed but, at least, impartial.
I strongly suspect that the failing systems that currently exist are kept in place because there's a financial incentive to do so. I agree that it's extremely unwise to invest $25,000 into the query-industrial complex, but people do it. This means that if an agent fails in that career, there's money to be made selling introductions. If this cottage industry of query support didn't exist, the query system itself would have been replaced five years ago.
The query-support system is mostly just parasites. If there was a better system for sorting through submissions, the agents of my acquaintance would happily adopt it. I'm sure some agents *are* already using AI for a first pass, and many already use human assistants for the same function.
Consider what that means for your experiment though. The AI or assistant gets instructions like "go through this slush pile and bring me all the romance novel submissions", or whatever the agent is looking for. An essay, however well-written, is not a novel submission, so the chances are high it never even crosses the agent's desk.
What you're saying proves my point, though. They don't read for quality. They read for other things. Quality isn't really a big deal from a commercial standpoint; if it's not there, a trade house will hire a book doctor or ghostwriter.
I'm not arguing that literary agents are unethical. It's not unethical not to read someone; I'm not owed it. I'm only arguing that the claimed meritocracy is dead. We should admit it, and then decide whether meritocracy is worth bringing back (in my view, it is) and how to do it.
I mean -- if your point is "publishing is a commercial enterprise", then yes, that's true! The system is trying to select the "best" books in the very narrow sense of "books that will make the most money". If sales don't line up with literary merit, they choose sales every time. But, given that we expect the audience to *pay* for the books, it's hard to imagine how it could be otherwise?
But I LOL'd at your point about book doctors and ghostwriters. This, roughly, *never* happens. (The only time you'd get those would be if you were *already famous* and what they mostly wanted from you was your name.) Ghostwriters cost money! The other end of being profit-maximizing is cutting costs. For years now writers have been complaining about getting *less* editorial support, that publishers want books in perfect shape from the start so they need an absolute minimum of (expensive) editorial attention.
Remember those huge volumes. If they think your book isn't 'quality' (=profitable) they're not going to rewrite it. They're going to *pick another book to buy* from the thousands available to them. Worries about editorial interference come later, when you actually have a career and a sales track record they want to leverage.
That's why I used N = 200. I considered all the factors, including that an agent might recognize the quality and relevance, but not reply since there is no call-to-action. After all, it wasn't a query—it's deliberately off-format (because the format is intentionally hostile to talent.)
Ultimately, no study can prove that agents categorically do not read, which I do not believe to be true. I think it's extremely rare that they read, and the fact that an N = 200 study found no signs of life is indicative of this.
yes, i don’t dispute the experimental implications.
as a sole datapoint; in my experience
as a hiring manager i did my best to read and consider as many resumes as possible, however if i chose to reject i used the form letter rather than crafting a reply. i imagined my time was better spent considering more applicants than on the rejection letters.
as a job applicant i am expecting a form letter of rejection so i seldom read them myself. i very much appreciate the closure of rejection rather than the uncertainty of no response.
i loosely understand the meaning of N=200, however i admit i am completely unfamiliar with the connection between hypothesis 1 of “they don’t read” to hypothesis 2 “the nature of their reply correlates to hypothesis 1”.
regardless, love your writing, both quality and content.
You make it sound like querying is new. It’s not. I've been published for 27 years and in the 6 years before a publisher bought me, I sent query letters. I have enough rejections to paper a wall in my office, and I’m not kidding. It’s the way publishing has worked for a long time and it actually works.
Agents know what they’re looking for, and a query letter lets them scan it to see if the story grabs them. Every agent I know (and I know many) are HOPING to find the next big thing. It’s how they make a living, and they love reading. They love discovering a voice they want to share with others. Most agents and editors will tell you they can determine in the first few pages if a project is for them. They have the experience of reading thousands of queries, and they read them with a hopeful eye.
Getting published is hard, but it’s always been hard. Back when I was first published, there was no viable indie option. It was vanity press or waiting for a publisher to like my writing and story enough to take a chance. I think it’s actually too easy to publish these days. Anyone can throw a book up on Amazon before it’s really ready.
And books aren't getting worse--they're getting better. I'm in awe of so many great stories out there. I read about 3 books a week on average, even when deep in the throes of writing my own next novel. Lisa Gardner is a favorite in the trad world but there are plenty of others who are automatic buys for me. A fave indie author is Kyla Stone. There are many great indie writers too. The field is vast and open. I talk to aspiring writers every day, and many of them WILL make it.
I lost my dear agent of 25 years, Karen Solem, this year. Karen always said, “If you want better sales, write a better book.” Ouch. But she was right. We can’t control much, but we can work hard on craft and story. We can spend hours, days, and weeks noodling on ideas and characters until we find THE ONE that hits others the right way. It’s in our court.
Here’s another truth: no one knows what book will hit the market in a big way. Not agents, not editors. It’s a mystery. The agents do their best and so do editors, but no one really knows. If they knew, every book would sell well. As authors all we can do is craft the best book we can, then pull our ego out of the way and get to the hard work of editing it to make it better. Even after all these years, I am still constantly reading a book on craft. I tell writers to do the best they can, edit and polish, then start querying. In the meantime, start the next book! You learn to write by writing. Attend conferences and learn more about the writing process and get to know the agents, editors and other writers. Encourage others and be encouraged by them. It’s a process but it can be a fun one if you embrace it.
I agree with much of what you're saying, but publishing was completely different in the late '90s. The last time querying worked was the early 2010s in YA dystopia. It's not accurate to say querying never works—only that it only does when publishing decides to clear the whole bench for one subgenre, as in did in the wake of the Hunger Games.
The query system is a relic. We don't need it anymore because, while an AI read isn't nearly as useful as a deep read by a skilled human, it provides better signal than a query letter does. If we can't give a fair read to everyone—and, given how few literary agents there are, and now many lottery players are out there, it seems we can't—then technology is the way to go.
This is a fascinating article which confirms what I already suspected. As an aspiring writer I’m shocked at how poorly literary agencies come across. Obsolete information, incomplete bios, constant obsession with demographics, such as the ‘disadvantaged’ - does that mean having an English degree is bad then? And then there are the ones who brag about their own book deals - forgive my cynicism but I wonder how they got those?
Maybe my work is good and maybe it isn’t - though as a small publisher did once agree to take one of my works on I don’t think it’s totally without merit! - but if I (and others in my position) don’t want to chuck hundreds and thousands of pounds/dollars down the toilet by sucking up to agents at conferences or getting beta readers/editors /query letter ‘doctors’ to do services that likely won’t be engaged with either, where do we go next!?
We have to figure out self-publishing. The problem is that self-publishing is about 100,000 different strategies, some of which work and some of which do not, and it depends on the book. What works for one title will flop for another. But any trade deal where you're not going to be a lead title, you should avoid at all costs.
If money's no object and you can get the editors you want, always self-publish. But some books are going to cost a lot to self-publish, while some cost very little, and it's hard to know in advance which is going to be the case.
Sorry, but ... what?
Why on Earth would a literary agent possibly respond to this "antiquery"? An agent isn't reading their slush pile to find good writing generally, or engaging literary talent; they're trying to find *finished books* that they can sell to publishers. What you sent them is, by its size and nature alone, instantly recognizable as not one of those, so I would 100% expect it to go right in the trash. (Not because it's low QUALITY, I can't speak to that! But because it's simply not what they're looking for.)
You have to look at the process from the agent's point of view. (Try reading slush sometime; I have, and it's enlightening. Lots of places take volunteers.) The whole system -- guidelines, query letter, partials, etc -- is designed to absolutely minimize the amount of time the agent spends on any given submission, not because the agent's time is so precious in an absolute sense but because the volume of submissions (for a successful agent) is ENORMOUS. So it's all triage.
Didn't follow the guidelines? Usually an automatic pass; maybe there's huge literary merit there, but why take the time to read and find out when there's ten thousand queries that *can* follow instructions? Then the query letter isn't an opportunity to demonstrate authorial prowess, it's a filter for cranks, psychos, and no-hopers. Again, having been on the other side, you would not believe how many submissions fall into this category: conspiracy rants, borderline terrifying sex fantasies, Star Wars fanfic without the names even changed, and so on.
Even once you cut out all of that there's still more incoming slush then it is physically possible to read. The standard query letter format is designed to present the salient facts about the piece to the agent as succinctly as possible: genre, length, logline, style. Typically agents aren't just looking for "anything good"; their job is to know what editors want to buy, and curate their own list accordingly. So a given agent in a given month might be glutted with space operas but eager for a fantasy romance, or vice versa, and the query gives an easy way to sort them out.
Only then, after all these filters, comes the much more time-consuming step of figuring out which submissions are *actually good*.
Anyway, this is probably too long a comment already. Suffice to say that the results of your experiment are exactly what anyone who has some experience with the system would have told you to expect. In counterpoint, I know quite a few people who have sold books to agents on query and many more who have made it partway and gotten personalized responses, so there's quite a bit of reading going on.
My experiment did not prove that agents never read anything (not true) or that they are unethical (not true) or that they do not care, in the abstract, about literary quality—they probably do.
It proved that they don't read for quality—that quality writing cannot overcome the low social status of having to use a query letter in the first place. You are correct; this probably should not have been surprising.
You have basically argued my case—that there's too much volume, and that there's no level of quality at which the writing says, "I'm not volume." So why play? Why beg, on the off chance that an agent's heart will grow three sizes that day? Why dance like a monkey when it will probably amount to nothing—even if you get "a book deal" it will probably be trash.
Querying is a freemium product. The free tier is the glitchy version that mostly exists to piss people off and prime them for the upsell. The pricing tier at which you become a human instead of "volume" is where you find MFA programs, expensive conferences, and "manuscript assessments" from famous editors that may lead to referral. This is basically vanity press, though, and it's often a bad financial play—to spend $25,000 querying and get a $5,000 book deal? No financial advisor would recommend that.
I am not arguing that literary agents are bad people. I don't think they are. I'm arguing that the system is garbage and should be torn down so it can be replaced with something that actually works.
Also, it's their fault they are inundated with low-quality submissions. I'm not the one who spreads false hope.
The system certainly has its flaws! But the fundamental problem it addresses is just *really hard*. There are many, many, many more people who want to publish books then there are books that can be profitably published, and sorting out the best of them without spending an absurd amount of time and money is extremely difficult. Any system to do it will have flaws.
You're too hung up on the query letter, though. Query letters are not as important as you seem to think. Typically all you can expect of a query letter is to confirm to an agent that yes, this person is sane and wrote a book and yes, it's in a genre/style that they're looking for. The sample chapters have to do the rest of the work -- if anything should be obsessed over, it's those.
It follows that "query letter clinics" and "manuscript assessments" are mostly scams. Any that cost $25,000 are *definitely* scams, as is any famous editor who charges for possible referrals. They generally will not increase your chance of getting accepted. I 100% agree with you that this is bullshit, but people love to prey on desperate would-be writers. But it's not an unlockable paid fast track; they can't deliver what they promise. (MFAs and conferences are a different kettle of fish, kind of a mixed bag.)
That's fair. My issue with the query system is that we no longer need it at today's technological level.
Consider AI. Can it read for quality as well as a skilled human? Hell no. Can it provide a better signal, for the first-pass determination of whether a submitter is worth an agent's time at all, than the query letter? Easily. So let's replace the query letter and the social proof factors with something flawed but, at least, impartial.
I strongly suspect that the failing systems that currently exist are kept in place because there's a financial incentive to do so. I agree that it's extremely unwise to invest $25,000 into the query-industrial complex, but people do it. This means that if an agent fails in that career, there's money to be made selling introductions. If this cottage industry of query support didn't exist, the query system itself would have been replaced five years ago.
The query-support system is mostly just parasites. If there was a better system for sorting through submissions, the agents of my acquaintance would happily adopt it. I'm sure some agents *are* already using AI for a first pass, and many already use human assistants for the same function.
Consider what that means for your experiment though. The AI or assistant gets instructions like "go through this slush pile and bring me all the romance novel submissions", or whatever the agent is looking for. An essay, however well-written, is not a novel submission, so the chances are high it never even crosses the agent's desk.
What you're saying proves my point, though. They don't read for quality. They read for other things. Quality isn't really a big deal from a commercial standpoint; if it's not there, a trade house will hire a book doctor or ghostwriter.
I'm not arguing that literary agents are unethical. It's not unethical not to read someone; I'm not owed it. I'm only arguing that the claimed meritocracy is dead. We should admit it, and then decide whether meritocracy is worth bringing back (in my view, it is) and how to do it.
I mean -- if your point is "publishing is a commercial enterprise", then yes, that's true! The system is trying to select the "best" books in the very narrow sense of "books that will make the most money". If sales don't line up with literary merit, they choose sales every time. But, given that we expect the audience to *pay* for the books, it's hard to imagine how it could be otherwise?
But I LOL'd at your point about book doctors and ghostwriters. This, roughly, *never* happens. (The only time you'd get those would be if you were *already famous* and what they mostly wanted from you was your name.) Ghostwriters cost money! The other end of being profit-maximizing is cutting costs. For years now writers have been complaining about getting *less* editorial support, that publishers want books in perfect shape from the start so they need an absolute minimum of (expensive) editorial attention.
Remember those huge volumes. If they think your book isn't 'quality' (=profitable) they're not going to rewrite it. They're going to *pick another book to buy* from the thousands available to them. Worries about editorial interference come later, when you actually have a career and a sales track record they want to leverage.
perhaps i missed it, but what if the agent did take the time to read then did not take the time necessary to craft a meaningful reply?
what if they assume their reply would not be read, so instead just used a form letter?
therefore is your hypothesis really proven?
i love the experiment and the method.
thank you for writing.
M
That's why I used N = 200. I considered all the factors, including that an agent might recognize the quality and relevance, but not reply since there is no call-to-action. After all, it wasn't a query—it's deliberately off-format (because the format is intentionally hostile to talent.)
Ultimately, no study can prove that agents categorically do not read, which I do not believe to be true. I think it's extremely rare that they read, and the fact that an N = 200 study found no signs of life is indicative of this.
yes, i don’t dispute the experimental implications.
as a sole datapoint; in my experience
as a hiring manager i did my best to read and consider as many resumes as possible, however if i chose to reject i used the form letter rather than crafting a reply. i imagined my time was better spent considering more applicants than on the rejection letters.
as a job applicant i am expecting a form letter of rejection so i seldom read them myself. i very much appreciate the closure of rejection rather than the uncertainty of no response.
i loosely understand the meaning of N=200, however i admit i am completely unfamiliar with the connection between hypothesis 1 of “they don’t read” to hypothesis 2 “the nature of their reply correlates to hypothesis 1”.
regardless, love your writing, both quality and content.
best,
M
This got dumber the more I read.
You sent a non-query letter to 200 agents and received form responses. From this, you conclude agents “don’t read.” Is it possible you don’t think?
You make it sound like querying is new. It’s not. I've been published for 27 years and in the 6 years before a publisher bought me, I sent query letters. I have enough rejections to paper a wall in my office, and I’m not kidding. It’s the way publishing has worked for a long time and it actually works.
Agents know what they’re looking for, and a query letter lets them scan it to see if the story grabs them. Every agent I know (and I know many) are HOPING to find the next big thing. It’s how they make a living, and they love reading. They love discovering a voice they want to share with others. Most agents and editors will tell you they can determine in the first few pages if a project is for them. They have the experience of reading thousands of queries, and they read them with a hopeful eye.
Getting published is hard, but it’s always been hard. Back when I was first published, there was no viable indie option. It was vanity press or waiting for a publisher to like my writing and story enough to take a chance. I think it’s actually too easy to publish these days. Anyone can throw a book up on Amazon before it’s really ready.
And books aren't getting worse--they're getting better. I'm in awe of so many great stories out there. I read about 3 books a week on average, even when deep in the throes of writing my own next novel. Lisa Gardner is a favorite in the trad world but there are plenty of others who are automatic buys for me. A fave indie author is Kyla Stone. There are many great indie writers too. The field is vast and open. I talk to aspiring writers every day, and many of them WILL make it.
I lost my dear agent of 25 years, Karen Solem, this year. Karen always said, “If you want better sales, write a better book.” Ouch. But she was right. We can’t control much, but we can work hard on craft and story. We can spend hours, days, and weeks noodling on ideas and characters until we find THE ONE that hits others the right way. It’s in our court.
Here’s another truth: no one knows what book will hit the market in a big way. Not agents, not editors. It’s a mystery. The agents do their best and so do editors, but no one really knows. If they knew, every book would sell well. As authors all we can do is craft the best book we can, then pull our ego out of the way and get to the hard work of editing it to make it better. Even after all these years, I am still constantly reading a book on craft. I tell writers to do the best they can, edit and polish, then start querying. In the meantime, start the next book! You learn to write by writing. Attend conferences and learn more about the writing process and get to know the agents, editors and other writers. Encourage others and be encouraged by them. It’s a process but it can be a fun one if you embrace it.
I agree with much of what you're saying, but publishing was completely different in the late '90s. The last time querying worked was the early 2010s in YA dystopia. It's not accurate to say querying never works—only that it only does when publishing decides to clear the whole bench for one subgenre, as in did in the wake of the Hunger Games.
The query system is a relic. We don't need it anymore because, while an AI read isn't nearly as useful as a deep read by a skilled human, it provides better signal than a query letter does. If we can't give a fair read to everyone—and, given how few literary agents there are, and now many lottery players are out there, it seems we can't—then technology is the way to go.
This is a fascinating article which confirms what I already suspected. As an aspiring writer I’m shocked at how poorly literary agencies come across. Obsolete information, incomplete bios, constant obsession with demographics, such as the ‘disadvantaged’ - does that mean having an English degree is bad then? And then there are the ones who brag about their own book deals - forgive my cynicism but I wonder how they got those?
Maybe my work is good and maybe it isn’t - though as a small publisher did once agree to take one of my works on I don’t think it’s totally without merit! - but if I (and others in my position) don’t want to chuck hundreds and thousands of pounds/dollars down the toilet by sucking up to agents at conferences or getting beta readers/editors /query letter ‘doctors’ to do services that likely won’t be engaged with either, where do we go next!?
We have to figure out self-publishing. The problem is that self-publishing is about 100,000 different strategies, some of which work and some of which do not, and it depends on the book. What works for one title will flop for another. But any trade deal where you're not going to be a lead title, you should avoid at all costs.
If money's no object and you can get the editors you want, always self-publish. But some books are going to cost a lot to self-publish, while some cost very little, and it's hard to know in advance which is going to be the case.
Brilliant. Thank you. I'd like to read you and I am subscribing.