textCrunch, No. 2: Looking Inside the Slush Pile

And your odds of getting out of it

The slush pile is the embodiment of all of publishing’s myths and romances. For aspiring writers, it’s about being discovered, plucked from the masses and identified as a budding talent. For agents, it’s about the thrill of the find, that frisson of energy that crackles with recognition: this could be something. And for all who practice the art and science of making books, it’s about (the myth of) meritocracy. That, all things being equal, talent stands a chance. Cinderella stories sell books, after all.

Any agent will tell you that finding a writer in slush is like finding a needle in a haystack. It’s so difficult, and with such diminishing returns, that even agents who maintain slush piles still look for clients elsewhere. Still, we like to talk about the needles—those books that made it, against the odds. We can name them: Catch-22 on the one hand, Twilight on the other. But we know quite little about the haystacks in which they’re found.

The slush pile itself is a bit of a mystery, one that fuels speculation and gossip. It’s a classic publishing research problem: on the one hand, there’s the “common knowledge” of publishing, a process that’s invisible to all but the insiders who hoard information. On the other, there’s a hunger for knowledge on the part of aspiring writers—especially about how to escape it. “The result,” I told my agent a few weeks ago, “is an information vacuum, where wild speculation and bias flourish. An information vacuum isn’t good for writers, it’s not good for the industry, and it’s not good for books.”

In this post, I’m not talking about needles. I’m talking about the haystack. This post will cover:

• The slush pile: how big it is, and a writer’s odds of being discovered in it

• What agents want to see in their slush piles

• The marketability of genres and categories

• The data we don’t have (but definitely need)

A quick note. There are two types of posts here at TextCrunch: those in which I recap and digest academic research, and those (like this one) that contain my own original data analysis. Moving forward, these types posts will only be available to paid subscribers. If you find this information helpful, enlightening, or otherwise worth supporting, please consider subscribing!

Slush has been on my mind because I recently joined the Book Industry Study Group for a webinar on “A.I. and the Slush Pile.”¹ I sketched out some simple questions in preparation.

Q: How many literary agents are there in the United States?

A: According to the best data (which we know is incomplete) around 1,500.

Q: How many agents read slush?

A: A tougher question to answer. Let’s say around 300 are entirely closed to queries, except by referral. That leaves us with around 1,200 who maintain slush piles.

But that doesn’t mean they read slush. Some agents’ slush piles are read entirely by assistants, who vet for promising submissions first. Some are open to queries only periodically. And while agents may have a slush pile, technically, that does not mean they pay it significant attention. Or any attention at all. Even those who do read slush likely still privilege clients that come in via referral, or whose work they’ve read in a literary magazine (another sort of referral). The slush pile may languish in a virtual corner for months, years.

Q: How many unsolicited manuscripts do agents receive each year?

A: Like almost everything else about publishing, it varies. It depends on the agent, her (and her agency’s) reputation, her specialty. In a 2012 article for Poets & Writer’s, Scott Hoffman of Folio Literary management estimated that Folio agents received, collectively, about 100,000 unsolicited queries a year. (Around 11,000 per agent.) In a 2014 article for the same publication, Susan Golomb (now of Writer’s House) estimated that she received about 20-30 new submissions daily; on the high end, that adds up to a little less than 11,000, not too far off from Hoffman’s estimate. (“And that was in 2012!” said Regina Brooks, president of the AAR and owner of Serendipity Literary Management, when I shared those numbers during the webinar.)

So let’s assume each of the 1,500 agents in the United States receives a similar volume. That’s 16,500,000 manuscripts in the collective slush pile each year.

Q: How many writers are found in the Slush Pile?

A: Vanishingly few. As a part of my book research, I observed an agency looking through unsolicited queries; in all, there was a .04% chance that one of the queries submitted would lead to an offer of representation. May the odds be ever in your favor.

For agents (and their assistants), the slush pile represents a volume problem: how to work through manuscripts in a timely fashion, on top of managing their existing and prospective clients. You can find some of the solutions offered up by Thad McIlroy at the BISG’s AI in Publishing seminar here. (And you can read Jane Friedman’s great write-up here).

I want to think about the other perspective. For writers, the slush pile represents an access problem: how to escape the dustheap and break into the industry?

There are countless barriers to access to publishing. Aside from skill—and, even rarer, talent— there’s the matter of credentials and networks of prestige.² There’s the matter of fit, which, as I wrote about last issue, often comes down to factors beyond a writer’s control, like their race or ethnicity, gender, or age.

Even setting aside these barriers, the playing field is not exactly level— not everyone stands an equal shot at being plucked from a slush pile because not every genre is equally desirable: there’s also the small matter of supply and demand. What sorts of books do agents want to read, sign, represent?

There are several websites where agents can identify the sorts of manuscripts they’d like to acquire—their “wish lists.” In many cases, a listing on one of these websites is an open invitation to be queried, a sign that an agent is actively reading, and taking on clients from, slush.³ This is all categorized according to genre or category. I collected data from one of those websites (and prettified it a little bit) to get a sense of the current genre demand.⁴

Here's what agents want.⁵

The graph above shows the genres most frequently sought-after by literary agents. Young Adult is the most frequent desired genre, followed by LGBTQ, Women’s Fiction, Middle Grade, and Commercial.⁶

Overwhelmingly, agent interest skews commercial. “Commercial” ranks higher than “Literary” in terms of the raw numbers (5^th vs. 8^th). More importantly, almost every other genre identified by name is a commercial genre. Realistically, we’re looking at “Literary,” “BIPOC Literary”, “Short Story”, and “Poetry” vs. Everything Else. Of course, there are lots of agents who still select into working on high-quality literary fiction; I’ve interviewed dozens of them for my book! But, because the majority of fiction bought and sold in the United States is commercial in nature, it’s not surprising that the majority of agents are likewise looking for commercially viable projects.

What’s driving the interest in these specific commercial genres? Is it because they sell reliably or are trending upward? Is it because there is a hole in the market that these agents hope to fill? Because they aren’t receiving many submissions in these genres or categories?

I suspect that part of this has to do with the agents’ identities. YA and MG are so desirable because many of the agents with profiles are on the younger end of their career, and therefore more likely to select into working on these genres. By the same logic, the high ranking of Women’s Fiction might be a result of the composition of the profession (which is, and always has been, comprised mostly of women).

These preferences, therefore, translate into an agents’ best sense of what could work. For an aspiring writer, this is really useful information. While one data point might tell us something about an individual agent’s taste and whether or not you should query them, when we zoom out, we get something like a collective assessment of saleability.⁷ Romance = very marketable. Cyberpunk = Tougher sell.

So while there is no secret to magically rising to the top of a slush pile, these data give us some useful information about your query might be received, and how desirable (read: marketable) your project might be.

Genre isn’t really categorical or straightforward. Case in point: Romantasy. Or take this deal, posted this morning:

How would you categorize this book, according to the categories above? RomCom, of course. But the topic—climate change—could also earn this book an Eco-Fiction categorization. At the very least, I think we can agree that this book is doing a little of both.

The graph below is a heatmap, which shows us agent interest in the combination of genres. Here’s how to read it: tiles that appear in green are not frequently combined; tiles in red are very frequently combined; yellow, occasionally combined.

Let’s set aside the needle/haystack metaphor for a moment, and think, instead, about archery. Think of each of these tiles as a target. The green tiles are relatively small targets, challenging to hit. The red tiles are larger targets, and (in theory), easier to hit. Looking at the darkest red points, we see that Thrillers and Women’s Fiction frequently go together—props to Gillian Flynn, Paula Hawkins, Lisa Jewell, Ruth Ware, Lucy Foley. At each of the darker orange-red tiles, you easily can imagine the bestseller that fills that niche.

This gives us a more granular picture of a project’s desirability. Say, for instance, you’ve written Dark Academia Novel about members of The Ravens, the brooding, scarlet-clad members of the Hawthorne Academy football team. You could find Gothic on the y-axis and Sports on the x, and see that there’s just one agent who might be interested. Unless you’re querying that specific agent, it might be a challenge. (I doubt that there’s much overlap between readers of Donna Tartt and Friday Night Lights, but I would be very happy to be proven wrong!)

But, of course, this isn’t the way that reading works. These graphs shouldn’t be taken as The Final Word on a writer’s chances of being signed by an agent. Hitting the “right” genre target is irrelevant, if the writing isn’t great, if the story doesn’t crackle. Two genres, chosen at random, might not “work” together—until they do. According to that my lovely heatmap, “RomCom” and “EcoFiction” are a pretty unlikely pairing (I don’t know about you, but climate anxiety doesn’t exactly spark warm fuzzy feelings for me). But Chelsea Curto’s book was just sold as a part of a two-book deal, at auction, with foreign rights sold in four territories (so far). Literature isn’t data, after all.

And just because more agents are searching for the next great YA Romance doesn’t mean that it will be easier for you to be published in that genre and category. Here’s what I don’t know: I have no idea how popular these categories are for aspiring writers.⁸ I have no idea how many writers are querying in these categories. I have no idea how competitive the field is for aspiring writers in each of these categories. More agents may want to work on Women’s Fiction and Romance—but it’s equally plausible that more writers do, too. You may have a bigger target to hit, but you might have to beat out many many more archers, and their 16,499,000 arrows, if you want to hit it.

This is the part of a newsletter when a snake-oil salesman might prey on the hopeful writer’s despair, and try to sell some magic formula guaranteed to help her escape a future languishing in a slush pile somewhere. Some perfect query letter, some irresistible hook for your first five pages. I don’t have a formula or a solution, or a template to sell you. I just have more data, more information, and I’m committed to sharing it. I believe that more information is good for writers, good for publishing, and ultimately, good for books.

So, please: instead of shelling out your money for some magic beans, save your money. Spend your time honing your craft as a writer and reading as widely as possible. It might not lead to publication, but—I promise—it will not be time wasted.

1. May I remind you: I have a very specific set of skills.

2. Full disclosure, I found my agent via referral

3. It’s worth acknowledging that many established agents do not have profiles on these sites, and that this is often a way for newer agents to drum up interest from prospective clients. So we likely see some bias in the results due to age (i.e., younger agents are more likely to acquire Children’s/MG/YA).

4. When creating a profile, agents can select as many pre-identified genres as they like. Some of these genres are extraordinarily specific, and some (frankly) don’t make much sense.

5. This is a Mount Rushmore sized Caveat: this is not a representative sample of all literary agents! Just those who have profiles on these websites and are open to queries.

6. Some readers might object to these genre categories—“Young Adult” and “Middle Grade” are age groups, not genres, per se. This is an artifact of the data.

7. In theory, of course. In practice, the book might be a stinker.

8. Any suggestions for how I might find this out? I fear that, until an agent or agency gives me access to their slush pile—ahem—this will remain a bit of a mystery.