Ask HN: Considerations when asked to write a book?

[new999999937]

I've been approached by 2 different publishers to author a programming book, based on the strength of a few technical articles I've written. They are both established publishers - think O'Reilly, Packt, Manning, Wrox, etc - and it's darn flattering to be asked. I think it'd be a fun experience, and a feather in my cap. Though I'm under no misconception about the monstrous amount of work it entails.

To those who have been down this road: Before I get into bed with a publisher, what else should I be thinking about? What questions should I be asking? What is a fair, market-rate deal for a first-time technical author writing about a popular subject? What are the areas to negotiate? Thanks HN!

[patio11]

What is a fair, market-rate deal for a first-time technical author writing about a popular subject?

You're going to be shocked and dismayed by the offer they give you. Let's get that out of the way now. This is the model: they'll tell you to not do it for the money, in exactly those words, prior to explaining to you terms which imply that they're not doing it for ~92.5% of the money.

You will likely be offered something akin to a $5k advance and a 8% royalty rate on paperback sales with, perhaps, a modestly higher royalty rate on e-book sales. The advance is guaranteed contingent on milestones. The royalties first "earn out" the advance and then start getting actually paid to you. (i.e. You have to sell $5k/0.08 = $62,500 of books on those terms prior to receiving any additional money.)

Most tech authors do not earn out advances. You should expect to receive just the advance and then occasional minor payments ($1k to $2k) for foreign rights if the book turns out to be so popular that e.g. it gets translated into e.g. Japanese.

If you want to make money in writing books, it is very possible. You'll want to start collecting email addresses, start publishing more things which get more people interested in trading you their email address, publish the book yourself, and sell via your own site/email list. For bonus points, sell packages (book, book + videos, book + videos + code samples) and price much, much higher than you're comfortable with ($49/$99/$249 is popular and works well).

There exist numerous people on HN who have successful businesses doing the second model. Understand that the second model is far more akin to running a business than it is to writing books. This is true of writing books generally, but it is more obviously true when you don't have a publisher to blame for your marketing/sales outcomes.

[TheOtherHobbes]

I asked someone who does this. Somewhat transcribed comments:

Expect $0 to $3k from a really shitty publisher who is probably just going to waste your time, or a serious publisher who believes you'll waste the publisher's time, or both.

$3k to $6k from a publisher who has some clue, but the topic is extremely niche. ("Pentesting South African Wifi Networks with Monads")

$6k to $12k from a publisher with some clue for a book that will sell fairly well to a big niche developer audience - something like a developer guide to Android or iOS.

$12k to $25k for super-popular consumer tech books for mainstream topics like macOS, iOS, and so on.

Royalty rates are 8% to 12%, maybe 15% for someone with a track record. (8% and $5k is standard for mid-list unproven fiction. Tech rates can be higher.)

Watch out for comedy royalty rates on ebooks. (15% on a downloadable file? 50% or GTFO.)

True: advances don't usually earn out. And if you write more than one book the unrecouped advances get added together, so you will never, ever, earn out, unless a book becomes an unexpected best-seller.

Which it won't, because publishers do not promote books. There's a sell-in period aimed at store buyers before publication to get the book on physical shelves, but otherwise publishers do somewhere between exactly zero and almost zero marketing and promotion.

Bottom line: writing books on its own is largely a waste of time, unless you want to research a topic and get paid to learn it and don't mind producing a book to pay for the payment.

Writing books as shop windows for other tutorial content and/or consultancy can work very well.

People who know what they're talking about, can write to a deadline and make sense, and are willing to accept pocket money are incredibly rare. Publishers often trawl for new writers, and if they find someone who is all of the above and not too easy to flatter they may be willing to push the advance higher than usual to close the deal.

[prawn]

Had an experience that gels with this. Advance of $1k per chapter (I wrote 3 chapters of an 8-10 chapter book) then without the publisher pushing it aggressively, royalties beyond that were minimal. The content did not date too badly so much as the presentation of it did.

Deadlines were brutal, especially incorporating edits and requests to expand/etc. That was a brain-melting experience. Sourcing and providing screenshots was particularly annoying.

Be prepared for it to be tougher than you expect and (as Patrick said) not to earn anything after the advance. Just when you feel like you're done, they'll come back with edits that double your workload.

Before long in my writing experience, I had "THINK ABOUT THE MONEY" taped to the top of my monitor. It was the only way I got through it. Interesting to do it once, but I'm not sure that I'd do it again for a publisher.

[zrail]

> There exist numerous people on HN who have successful businesses doing the second model. Understand that the second model is far more akin to running a business than it is to writing books.

I've done this twice now (links in profile if anyone is interested). It is very much a business for me. The first book led to an ongoing consulting gig with a wonderful client, along with a smattering of additional clients over the last few years. The first book has also earned north of $50k since publication, of which I've been able to keep 97% minus minor costs like hosting.

I have not published with a traditional publisher. O'Reilly approached me about publishing something like MMP as a traditional book, but they weren't interested in conciseness nor were they willing to pay anything close to what I was making self-publishing so it never went anywhere.

[paulajohnson]

There is also a ... "business model" ... for technical books which goes like this. Get someone with a PhD to write a book on something. Around the world there are about 300 libraries of record which are legally required to hold a copy of everything published. So you can sell them a book that nobody wants for $200 each. That is $60,000 of sales guaranteed, with minimal requirements for copy editing and other forms of quality control. Rinse and repeat for as long as the supply of stage-struck PhDs holds out.

[acangiano]

Entirely agreed on your second point.

Regarding your first part, I must mention that The Pragmatic Programmers are the exception to this rule. They offer 50% of royalties which adds up fast.

[jseliger]

Great post. I'll add that Philip Greenspun's article "The book behind the book" is also excellent: http://philip.greenspun.com/wtr/dead-trees/story.html. It affected my own life and decision making too: http://seliger.com/2011/03/06/why-youre-unlikely-to-see-seli....

Greespun wrote his piece in 1997 and updated it in 1999. Except for the growth of the second model, too little has changed.

[baudehlo]

I'd be shocked if anyone got a 4 figure advance. I got a few hundred dollars for an XML book in the height of the XML times. Ultimately the book never got finished but everyone I know who has written a tech book (unless it's a huge seller) says expect nothing more than lunch money from it, plus the employment opportunities.

[Xorlev]

As a first-time writer, you might not even be offered an advance (contingent or not). I certainly wasn't. My contract was merely royalties.

[cheez]

TL;DR self publish

[jonshariat]

I'm currently an author with O'Reilly, writing about the design space. (http://shop.oreilly.com/product/0636920038887.do)

I am in the final stages after over a year and half. I was in your position not too long ago and researched as much as I could before accepting, so I will skip things you will find in a quick Google search.

Negotiating:

1. Everything is negotiable, remember that

2. Under promise on #pages, they ask you for an estimate but lowball it because thats how they price it and then will ask you to fill it.

3. Get in writing what they will be contributing.

Choosing the topic:

1. Choose a topic you are passionate about

2. Choose a topic you know a lot about

3. Choose a topic you can write a lot about

4. Don't be afraid to tweak the topic half way in.

5. Choose a topic that will sell in a year and half.

Writing the book:

1. Write an outline, write the first chapter, throw it away, and rewrite the outline agin.

2. Its better to lose work than to keep going in a direction that isn't working

3. You'll be busy but be reading other books as much as you are writing.

4. Wake up early or stay up late, 0 distractions is the best for writing

5. Talk out loud, like youre presenting to an audience to get unstuck from writers block.

6. Get feedback as soon as possible.

Marketing the book:

1. If you can, get in writing what marketing they will do for the book.

2. Negotiate on how many free books you can get to give away

3. Sales on Amazon are important, direct sales there, esp on launch day.

4. Start a newsletter (yay more writing!)

Other:

1. This is a second job

2. It does open a lot of doors

3. Consider a co-author for your first book. I added one towards the end and wish I had done it sooner. Its easier to collaborate and bounce ideas off each other and keep each other accountable.

[mcbits]

> Negotiate on how many free books you can get to give away

What's a realistic order-of-magnitude number there? 50? And do you have to agree not to sell them?

[raesene6]

One point there really resonantes with me as someone who reads a fair number of tech books.

"5. Choose a topic that will sell in a year and half."

makes it hard (I think) for books to be written about very fast moving technical topics (e.g. Docker/Kubernetes/CoreOS). By the time you've written the book, it will inevitably be out of date.

The approach that publishers like Manning and O'Reilly take of publishing beta books helps a bit but still these topics are not a recipe for long-term saleability.

[pmiller2]

I can tell you have actual experience here. These are the same things I've heard people I know who have written books say.

[swalberg]

I've written 3 technical books. I highly recommend talking to an agent. I used studiob.com. Not only will they negotiate money on your behalf, but they'll get crappy clauses taken out. Stuff you've never heard of like "cross accounting clauses". My last contract had a right of first refusal for my next book in their opening offer. The acquisitions editor even laughed when we asked about it because she didn't know it was there.

The agent will take 15% and believe me it's worth it.

Whatever deal you cut, pretend like it's not going to earn out its advance. Because it probably won't. As everyone says, you're not doing it for the money. You're doing it because you love the topic, love to write, and to be seen as an expert in the field. Writing the books themselves earned me very little money. Follow on work, like articles, and using it in salary negotiations, more than paid for itself.

I'd suggest you figure out the book you want to write then shop it around the publishers (again, agent helps). And be picky about the publisher, talk to other authors that used them. They're not all the same. When one publisher hires editors and pays their technical editors and another expects a lot of self-editing and offers a free book to the tech reviewers, the outcomes will be vastly different.

[drcode]

As an author of several books with an "established publisher" I guess my main advice is that you know 1000x more about your domain than your publisher and you have a better idea than them what the book should be like so that it will sell a year and a half from now.

The best way to write a financially-successful book is not to negotiate the best royalty rates possible, but instead to make sure you write a book people are actually going to want to buy once the book comes to market... make sure you get a reasonable contract, but after that use any remaining leverage you have to make sure you get to write the right book for your market.

[JohnStrange]

That sounds like very good advice. Let me add to this that it is extremely important for a programming related book to double and triple check for technical accuracy, and in particular check again and again that code snippets actually compile if they are intended to be complete.

Don't rely on any reviewers from the publisher to find mistakes.

[timbutlerau]

As someone who's currently 1/3 of the way into doing exactly what you've been asked... I'd advise you to really strongly consider the time it takes. I was wrong by a factor of 4. Thinking I could spend an extra few hours a week takes hours a night.

Expect no support, their staff are only there are a conduit to move things around. The other thing that shocked me is how crude it all is. The publisher I'm engaged with only has email and word docs. No form of document management nor version control outside of manual naming of the documents.

Even if my book sells well, it won't cover the cost of the time if I'd simply consulted that many hours. If you are considering it for the money, don't accept. If you want to study a subject in detail and get partially paid for it, then it might work out.

[asadjb]

100% this.

What I thought would be a quick few hours a week thing turned into taking up all of my free time over the weekends. And writing the first draft isn't even the most difficult part. In the review stage, you'll be revisiting stuff you wrote more than 5 months ago and we amazed at what a poor job you did then. Of course, time constraints make it impossible to re-write the whole thing, which is something we programmers like doing so much! So you do the best you can and move on.

And the tooling is non-existent. Word documents and emails are what I had to deal with. There were multiple instances in which I was editing the wrong version of the chapter, or the editor was looking at an older version, etc. I hope better tools exist for this process.

[joelhooks]

I've published two books through publishers and the actual writing experience was very similar to the book that I self-published.

For the self-published book I followed he basic path set forth in Nathan Barry's Authority[0]. Financially speaking it was definitely more lucrative to self-publish, but I've got access to a large audience to sell to.

There is another approach that I've seen that I find interesting and that is of the "open source" variety where the book is given freely and later published by a major publishing house. You Don't Know JS is a recent example.

My personal experience with OReilly was a good one. They send you a framed cover and it was a decent experience. We got no advance and the pay was peanuts, but it served to get me recognized at the time in a specific space.

Sure, money isn't the only motivator when setting out to write a book, but it's definitely a motivator!

[0] http://nathanbarry.com/authority/

[StevePerkins]

Look for any "right of first refusal" clauses in the contract.

I was approached by Packt a few years ago, to write a book in a fairly niche topic (http://amzn.to/29LR9Ly). Their reputation is not the most stellar, but I put a lot of work into making a high-quality book with source code on GitHub and live examples running on Heroku. So I "beat the odds", I guess... earning several times the (small) advance, and getting some positive reviews on Amazon and Goodreads. The money doesn't justify the time that I put in, but it's tremendous resume fodder and played a large role in taking my career to the next level with my next job change.

However, my biggest regret is that I signed a clause giving Packt the right of first refusal on my next two books. Realistically speaking, I doubt I'll ever get around to writing another technical book again. If I do, I'll probably pitch it to Packt and then self-publish if they don't want it. But it sucks being locked in like that.

Truth be told, if I had known then what I know now then I probably would have just self-published in the first place. There's no real money in book authoring no matter which route you take. Publishers don't really promote books, and you're really on your own anyway for author support during the writing process. I doubt that most employers looking at my resume would care whether the publisher was Packt or Leanpub, so I probably would have gone the latter route just for the freedom.

Of course, if one of your publisher options is O'Reilly, then maybe that's another discussion.

[drb311]

Steven I'm managing editor at Packt. Thank you for publishing your book with us.

If you want to publish your next work yourself or with another publisher just mail us and ask. I'm sure we wouldn't enforce that clause against your will.

[przeor]

hello Steve, could you tell more about "signed a clause giving Packt the right of first refusal on my next two books".

I am also with Packt and I don't see any clause like that in my contract :-) (I've finished 350 pages from 380 goal, my book is 90% finished).

I am interested in details, what and why you had to sign smth like that?

Regards, Kamil

[calcsam]

Most technical authors don't write a book for the money -- they write a book for the prestige (and proportionally higher consulting rates).

[spdustin]

This is likely to be the more useful answer here. The biggest benefit comes from listing "author of [book you can buy at bookstore]" on your CV or web site. The marketing power of your name being seen by potential employers (or their employees who point to you as the domain expert of a topic). Not from actual royalties past the royalty advance which, as others have mentioned, would be rare.

Of course, the value of that would depend on if you trade on your name; e.g. if you're a freelance consultant.

[ghaff]

Even if you're not a consultant but otherwise have a job role that relates to your profile in the industry in some way or other, books are valuable.

I would add that, in many cases, a book is a book. People have short attention spans today anyway so for topics that don't require 100s of pages of code examples and screenshots etc. shorter may be just as good as longer. There are a lot of short books out there these days and they can be just as valuable as giveaways and credibility enhancers as a doorstop is.

[julian_t]

Absolutely. When I was starting out on my own I got the opportunity to write for Wrox Press. When talking to clients and going for contracts, it was definitely worth more than being one more guy with an MCSD.

Also, as Dr Johnson noted, writing a book is an excellent way to find out how little you know about a topic ;-)

[clifanatic]

Can confirm: I didn't publish for the money, I published for the prestige. Still waiting for all the prestige to roll in...

[MarcScott]

I wrote a kids programming book. £1.5k advance and 8% royalties if I remember correctly.

My biggest takeaway, is forget your tool chain. I wrote it in org mode, and exported to Libre Office and then Word. As soon as the first draft was returned I ended up battling with Word the whole time, for edits and redrafts.

Also, designers don't read code like programmers. Indentation was often messed up, incorrect quotes used and line breaks added to suit formatting. One particularly badly formatted piece of code had me ripping my hair out. Everything beneath the first line was messed up. The designer came back to me a bit confused, saying the only problem he could see was the first line of code was 4 spaces too far to the left. Being accustomed to an absolute left limit on code, defined by the first line, it hadn't occurred to me to view it like this.

[andrewl]

I've heard (but can't verify) that the Pragmatic Programmers have their own tool chain, and that it's quite good. It will, for example, check that code snippets are syntactically correct.

I have a few of their books, and they're pretty good. Their store is also very straightforward and friendly.

Does anybody have experience writing for them?

[lsc]

>My biggest takeaway, is forget your tool chain. I wrote it in org mode, and exported to Libre Office and then Word. As soon as the first draft was returned I ended up battling with Word the whole time, for edits and redrafts.

Ah yes, this is another part of the negotiation that I would have re-done, had I to negotiate again. Publishers will accommodate you if you push hard enough.

[julian_t]

I'm starting a self-publish project, and am looking at using Pollen[1] along with some CSS based on Edward Tufte's ideas[2].

I'm not a designer by any means, but I'm having fun with this, and the result is looking pretty reasonable so far. (And learning one programming language while writing a book on another is kind of cool)

[1] http://docs.racket-lang.org/pollen/ [2] https://edwardtufte.github.io/tufte-css/

[simonsarris]

I did this with Pearson (SAMS imprint) and wrote HTML5 Unleashed. My first time = $10K advance and 12.5% (I think, forget) royalty rate. I'd gladly answer any questions.

> what else should I be thinking about?

Time. I spent a better part of a year of my free time researching and writing the book. Its almost like I skipped a year-ish of life. Of course, the next time I write a book will be a lot quicker.

In terms of hours worked, its not a lot of money. If you like writing (and I do), it may still be worth it. Like most projects, the second one you do will probably be much quicker and feel a lot better than the first!

By the way everyone I worked with at Pearson was an A+ friendly, helpful person. I have a much less favorable impression of Packt, but never wrote for them.

[drcode]

> Of course, the next time I write a book will be a lot quicker.

That's what I thought, until I wrote my second book.

[paulbjensen]

I've just spent 18 months working on a title for Manning which is due for publication soon: https://www.manning.com/books/cross-platform-desktop-applica...

- Writing a book is a huge undertaking - make sure that you have the time to do it.

- Find a good environment to write the book in - if you have kids/pets/others that will interrupt you, try to find somewhere else to do the writing - Marcus Hammerberg (the co-author of Kanban in Action) wrote his in coffee shops, with the side effect that he gained 10lbs, so factor that in.

- Try to pause on blockers if you can - time spent on coding/writing does not always result in the equivalent amount of book being produced.

- Check what the tech landscape for your book's subject is like - if it's very fluid then plan around it - since I started my book, Node.js was forked to create IO.js, Node Webkit was renamed to NW.js, IO.js merged back into Node.js, and then Electron emerged on the scene and now overshadows NW.js (which is ironic given that they have something of a shared history).

- Keep your editor informed - I've had quite an eventful 18 months and it helps to let them know what's going on in your life.

- Writing a book is a great feeling when you see your name as a published author.

I think the one takeaway is that it is like doing a dissertation whilst also doing a job - ask yourself if you can commit to that. If you can, great. And find the right motivation too, because you are putting at least a year of your life into this project.

[antoaravinth]

Out of curiosity how much does Manning pays for the first time author?

[Siecje]

I just bought your book. It still talks about IO.js. Are you planning on updating that?

[d23]

I went into it for the "feather in my cap" and "fun experience" aspect as well. It's grueling, stressful, and the piddly amount of money you'll make for it doesn't do much for motivation. I stopped about 3/4ths of the way through writing it and finally admitted to myself that I couldn't go on.

Some other folks have put out information about the advances and royalties, and it's about what I saw as well. If you're doing it for the money, I think you'll be disappointed, but at least you can gauge that before accepting. If you go the self-publishing route, I'm sure the amount of money you can make can increase if you have the right network, but you may lose that sense of "making it" by being published by one of the big names.

[amenghra]

I was a reviewer for a Packt book (http://droppdf.com/v/UTl8X). The review process was terrible, they ended up not taking my feedback into account and the book was published with lots of technical mistakes.

The final book is essentially a random tutorial from pre-existing web content. Avoid them.

[halfdan]

While I was working on a video course for Packt, someone from their company approached me and asked whether I'd be free to review a video course.. (mine)

They are badly organised and most of their communication happens through Word docs randomly spread over multiple Dropbox folders.

[erroneousfunk]

Yes, Packt is awful. Do you enjoy spending hours on the phone with overseas tech support employees who speak very little English, don't care about their jobs, and have no actual skills? Now just imagine that, but you're trying to edit a book with them, and you're doing it around their local office hours. That's what writing for Packt is like.

[rdl]

I'm friends with the owner of No Starch Press, but I'd consider their published royalty schedule to be quite fair and a good sign of the upper-middle of the industry.

https://www.nostarch.com/writeforus.htm

15 percent royalty with no advance 12 percent royalty with $5,000 advance 10 percent royalty with $8,000 advance

[bhouston]

I've been approached quite a few times over the decades.

In my opinion you will make very little money from writing the book (unless it is a rare O'Reilly blockbuster) and it will take a horrible amount of time.

I think if you do the calculation based on expected income versus the time required to write the book, it is likely be close to minimum wage. Or it sure seemed close when I did the calculation for myself.

I think that those books make money for the publishers, but not much for the authors, especially if you can make market wages in a major market, or possibly more if you can get stock or some other form of potentially lucrative compensation.

[philip1209]

I haven't followed through, but just as an FYI - many people who have written blog posts have been approached by publishers. They don't put all of their eggs in one basket - for a given topic, they go through the "outlining" stage with multiple people.

[TylerE]

I would say it depends a LOT on who it is. Packt books for instance, are mostly....not very good.

[subsection1h]

By "not very good", do you happen to mean that Packt is a content farm that produces low-quality information written by cheap labor?

[shanemhansen]

I was approached to write a book for them (Mastering Go). I'm going to have to pass.

[parisidau]

I've written 10+ books for O'Reilly, and one for Wiley/For Dummies, back in the day. It's a great experience, and I'm still writing for them now. We really enjoy it, and gain credibility, clients, and speaking opportunities from it. The money is fine, but the opportunities are great. Once you've done one or two, it gets easier and quicker.

[kkapelon]

I have authored a book for Manning (Java testing with Spock) and was approached like you are (they noticed some of my technical articles)

My advice

1) Unless you are going to write a big hit, you are not going to earn match. You write a technical book for prestige, not money. Writing a technical book for money is the wrong reason to write it. Stop now.

2)You should really know your topic well. I mean REALLY know it.

3)The amount of effort it will actually take will be 6x the effort you think it will take.

4)Make sure that you have enough free time and you have discussed the matter with your significant other (and that she/he approves)

[ninjakeyboard]

I wrote a book for one of the pubs you mentioned and have talked to several other authors. Know what you're getting yourself into - you're going to be responsible for all editing so it is A LOT OF WORK. Never ending work. And it will never be good enough but dates move forward.

You don't make any money publishing with a publisher - not directly from the book, anyway. Maybe a dollar or two an hour. Do not do it for the money. Once you're done, however, you can demand a higher salary as an 'expert' - say an extra 10% or 15%. Or double your income if you move into contract or consulting. :)

One STRONG recommendation is to not promise the publisher first stab at any future books. Some publishers have this in their default author contract - that you can't publish with another publisher without first offering the publisher the opportunity to publish first (maybe for your next 3 books). That clause is one I would demand be removed from the contract.