Learn how to write a enterprise guide — behind the scenes from THE COLD START PROBLEM at andrewchen

(above: Me in Sep 2023, a cheerful creator, discovering the Japanese translated model of my guide on the fantastic Daikanyama Tsutaya Books in Tokyo)

Expensive readers,

As lots of you realize, 2 years in the past I printed my first guide THE COLD START PROBLEM. It goals to inform the story of why some merchandise — YouTube, Instagram, Uber, Slack, Dropbox, and others — find yourself with a whole lot of tens of millions (and typically billions!) of customers, and to offer the definitive concept of community results which are sometimes referenced within the tech business, however solely superficially understood. It’s been successful, now in a dozen markets, translated into many languages (together with Japanese, Chinese language, Spanish, Russian, and so on).

Right here’s a screenshot of among the fantastic footage that readers took throughout launch week:

This was an superior expertise. However, I swear I’ll by no means write one other guide once more . (I suppose by no means say by no means, ha)

The artistic course of was a protracted, meandering path, and folk ping me every now and then as a result of they need to tackle a masochistic journey of their very own. So this publish can be in regards to the messy, annoying, behind the scenes main as much as writing a guide like this — it describes a bit the artistic course of, but additionally among the main milestones and classes realized.

Hopefully it will likely be helpful for somebody sooner or later who’s pondering of an enormous writing mission of their very own.

A quick abstract of what I’ll cowl:

• Month 0: At first, writing a guide looks like a enjoyable concept (till you determine it’s not)
• Month 1-6: Discovering an agent, writing a proposal, and opening up your Christmas presents early
• Month 6-12: Amassing and organizing the concepts — a lot of enjoyable chats, reconnecting with colleagues, speaking to nice individuals
• Month 12: Learn how to write the preliminary the define, then the mega-outline — discovering the system
• Month 12-24: The very messy center, the trough of sorrow, the laborious slog, adopted by trench warfare (sure, it’s 3-5 years to put in writing a guide)
• Month 24-36: Why you’ll really feel insecure in regards to the artistic course of
• Closing months: Simply ship it already

I’m additionally going to hyperlink to varied copies of intermediate content material alongside the way in which — sadly I can’t share every little thing (like interview notes, and so on) since some stuff must be confidential, however right here’s just a few fascinating bits anyway

OK — so let’s get began on the journey.

Month 0: At first, writing a guide looks like a enjoyable concept (till you determine it’s not)
I joined Andreessen Horowitz in mid-2018, I had already been writing on my weblog for 10 years and I used to be form of having some artistic boredom over it. At the moment, Elad Gil had simply printed his guide and we had a pleasant convo at MKT’s lounge a couple of week after his guide was out — he had superb issues to say in regards to the course of (he writes quicker/higher than me, behind Ubers it seems), and he mentioned it helped him so much professionally. At a16z, as you all know, Ben and Scott have each written incredible books as nicely, and it appeared to actually be nice for them professionally, so I assumed it may be a enjoyable problem to do the identical. So consider the motivation as 50% a artistic problem, and 50% seeing what it had carried out for different individuals.

I had two traces of pondering by way of selecting the guide matter. First, I’ve had good luck taking ubiquitous jargon and writing the definitive weblog publish on the subject — one thing I did with progress hacking, CAC/LTV, viral loops, and ideas like that. I had just a few concepts bouncing round in my head that felt like good candidates. “Energy customers” was one — a time period we use willy nilly, however with no sturdy theoretic underpinning. “Community results” was one other, since we had been speaking about it at a16z on a regular basis, however when it got here time to take a look at the metrics and reply the query — OK so does this product have it!?? — then it acquired slightly mushier. One other was “Product/market match” or “MVP” and increasing these ideas a lot additional.

The opposite line of pondering revolves round answering the large query — why? I made a decision my focus could be on one thing concentrating on a really small group of nerdy founders and executives, fairly than a large matter that may be extra mainstream. I might write about, say, profession recommendation or the right way to begin a enterprise (in a common sense), however felt like these could be too broad.

Ultimately, I picked the subject of community results as a result of it’s a genuinely vital matter, I felt like I had one thing to say, and I additionally felt prefer it might fold in a number of ideas from my prior work on progress. As soon as I began down the trail of selecting, I began to speak to individuals at a16z about it — they beneficial I begin with writing a guide proposal.

Month 1-6: Discovering an agent, writing a proposal, and opening up your Christmas presents early
The workforce at a16z was very useful, and within the first few weeks I labored with Hanne Winarsky (now at Substack!) and others to begin assembly brokers, which is how I finally met Chris Parris-Lamb from Gernert who additionally represented Peter Thiel for Zero to One, and Pete Buttigieg for his guide. I despatched him the next guide proposal with a placeholder identify, MOONSHOT. The proposal normally kinda reads like a marketing strategy:

• Overview
• Chapter summaries
• The market
• Writer bio
• Aggressive books

We shortly agreed to work collectively, and that we might strategy varied publishers to solicit gives. The precise strategy was form of enjoyable, truthfully a extra environment friendly model of what we do in enterprise capital. Chris ran the entire thing, and the method appeared like the next:

• Chris approached publishers and despatched alongside the guide proposal
• They learn the proposal (thanks!) and requested for half-hour of time
• We acquired on the decision they usually requested me detailed questions, displaying that they had really learn the proposals — UNLIKE a typical startup/VC course of the place the founder makes use of the time to current
• Later, they submitted a suggestion (I feel 7 did?)
• Chris then took the highest half of the gives, and gave them a second probability to bid once more
• The highest 2 bids had been shut, however I selected to work with Hollis Heimbouch at Harper Enterprise

I selected Hollis as a result of she’s legend within the business, and labored with Jim Collins, Clay Christensen, Satya Nadella, and others on their most well-known books — a16z had additionally labored along with her for Ben’s earlier guide and it went nicely. My advance was excessive mid six figures, which I used to be advised was superb for a first-time creator, and could be paid out in components because the guide progressed (one half at signing, the subsequent on the draft, the subsequent at publishing, and so on).

The complete means of doing this was possibly 3-4 months? I’ve described the early days of this as “opening up your Christmas presents early” since you get all the nice vibes up entrance of promoting the guide, with out the work of really writing something. However quickly I used to be going to pay the worth!

Month 6-12: Amassing and organizing the concepts — a lot of enjoyable chats, reconnecting with colleagues, speaking to nice individuals
The remainder of the primary yr was fairly enjoyable as nicely — I noticed I wanted to do a number of major analysis, so I began reaching out to individuals I revered, asking them for brief interviews. Thanks to Li Jin who tag teamed with me on many of those interviews, the place I requested open-ended questions, heard tales, and tried to put in writing as a lot of it down as potential.

Readers need to hear opinions. The sharper and funnier, the higher, and I had a concept that if I might accumulate all of it, then that in itself might be the bones of a guide. Thus, I wrote down pithy, opinionated statements each time I heard them — something that may be a very good tweet would even be a very good title or a very good opening paragraph. Opinions like, “launching with Techcrunch is silly” or “by no means construct a social community, it’s simply too laborious” — these are gold.

All of the interviews went right into a spreadsheet tracker like this, which linked to particular person notes for every, plus slightly abstract.

Ultimately, I ended up with 200+ interviews from individuals within the business, and pages and pages of opinions and ideas. It was absolute chaos. However I might additionally inform there was one thing fascinating in there. I ultimately interviewed some senior of us within the business — the founders/CEOs of Slack, YouTube, Twitch, Tinder, Dropbox, Zoom, Linkedin, and will others — these all ended up being tremendous enjoyable, and had been the showcase tales within the guide. Getting time with these of us ended up being among the most memorable moments whereas writing the guide.

Month 12: Learn how to write the preliminary the define, then the mega-outline — discovering the system
When you’ve got a whole lot of pages of random notes from interviews, plus pages of analysis, and a jumble of concepts in your individual head — what do you do? You want some form of organizing precept that makes all these concepts readable. I figured there was most likely a system in among the greatest enterprise books on the market, and so I re-read Lean Startup, Crossing the Chasm, Innovator’s Dilemma, and plenty of others.

What you discover it that the bones of the guide typically look one thing like this:

• Opening story
• Describe an enormous drawback/dilemma/query
• Current a framework
• Undergo one a part of the framework
  • Begin with an anecdote
  • Then describe the speculation
• Undergo one other half
• Then one other half
• Then once more…
• Conclusion

This isn’t all enterprise books, however look, it’s fairly ubiquitous. And so I assumed I’d begin by structuring my preliminary define form of like this, which is how I ended up with the next quick model of the define. The primary guide define.

Btw, Ryan Vacation has an amazing dialogue of how he wrote his guide, with tons of images, and I need to hyperlink that right here. He has a photograph of a field representing each matter/concept on this guide, every one in a notice card, categorized into sections:

I form of ended up doing the digital model of this, the place I created a doc that I referred to as my “Mega Define” — the place I took each opinion/level that I needed to make within the guide, and constructed out the primary 2-3 ranges of bullets in a a lot bigger model.

Right here’s the primary web page, so you may get a way:

I’ve linked all the Mega define right here if you wish to peruse — it’s 30 pages the place every web page wanted to most likely be 10x’d. That’s, 1 web page of define = 10 pages of written prose, which I shortly discovered as I started to put in writing the primary few chapters. There’s a humorous George RR Martin dialogue (he’s the Recreation of Thrones man) the place he talks about how some writers are Architects, and a few are Gardeners. The architects do what Ryan Vacation and I each do — we’ve some chaos in the beginning, which we attempt to ruthlessly suppress, and use some organizing rules to place it collectively. As soon as there’s a construction, then that’s like a basis of a constructing — the architects then write, flooring by flooring, and construct the entire thing. Plus some polish on the finish. It seems that GRRM describes himself as the opposite archetype, the gardener, the place you form of plant some fascinating factors right here and there, then revisit them as you write. However that’s why his books are superb and take 10 years to put in writing.

Month 12-24: The very messy center, the trough of sorrow, the laborious slog, adopted by trench warfare (sure, it’s 3-5 years to put in writing a guide)
This entire center part after the primary yr provides me PTSD so I gained’t dwell an excessive amount of on it, and simply cowl the teachings realized. The mechanics of this part are fairly easy — you actually simply must translate the mega define bullet by bullet into pages of written prose. However listed below are all the issues you’ll face:

• Your regular instruments usually are not good for writing a guide. Most writing that you simply do every day, like e mail, may be composed of some paragraphs. That’s straightforward. In case you want an extended doc, then you definitely might need a number of sections that include a number of paragraphs every, and also you’ll use Microsoft Phrase or GDocs. However what if in case you have a guide with 7-10 components that include 5-10 chapters every, that include 3-4 main sections that themselves include numerous paragraphs? And what if midway via, you understand all of the stuff you’re writing in a single part ought to really belong as a chapter in one other part? Additionally what if you wish to do a phrase depend of various chapters or sections? It’s all a ache. Ultimately, I used Ulysses which no less than has the idea of nested folders, after which every chapter could be a folder that may then include information containing every half. The app then sync’d all of it to a bunch of Markdown information in a Dropbox, in order that I might work on it from a number of places
• You write on a pc, and your laptop could be very distracting. You want a browser to do analysis, however your browser can also be the place you’ll be able to examine what’s taking place on social media. You’ll be able to’t absolutely flip off the web, since you’ll want to do analysis. And typically you’ll want to go to YouTube to look at an interview, however proper subsequent to the video you’re purported to be watching is a gadget evaluation for one thing you would possibly need to to purchase. So what do you do?
• Distraction free gadgets and treating your self like a child. Ultimately I began to attempt to purchase a bunch of various instruments to maintain myself centered. I purchased a plexiglass timer secure factor and I’d lock my private telephone away for an hour or two at a time. I purchased a separate laptop computer, and put it in a unique location, and turned on all of the child-safe filters in order that I couldn’t go to Reddit, Twitter, and so on. If I wanted to lookup analysis, I’d typically simply print out pages and pages of it, in order that I’d keep analog and never fiddle. I purchased a sequence of e-ink Android tablets referred to as the BOOX that would run a Markdown editor, hook up with Dropbox, and will pair a pleasant keyboard.
• Say goodbye to holidays, weekends, and night time. To hit the deadlines I had set for myself, I ended up changing a number of my holidays and weekends into writing time. It’s laborious to put in writing for greater than, 3-4 hours in a row, so you continue to can go someplace good and sunny — however I discovered that I wanted to get up, work out, and get writing earlier than midday, as a way to make progress. You get your evenings, however it’s powerful. And weekends are like that too.

Right here’s a humorous photograph of certainly one of these kSafe timers I’d cover my telephone into throughout my writing instances — by the tip, I had 5 (!!!) of those in varied writing spots, in order that if I used to be feeling within the temper I’d throw my telephone in:

I’ve to confess, it was a grind. Not straightforward in any respect. If there was a degree the place I might have gotten caught and give up, this is able to have been it.

Month 24-36: Why you’ll really feel insecure in regards to the artistic course of
One of many craziest issues about writing a guide is that it’s such an extremely solitary expertise, and there’s finally a degree the place you’ve written sufficient that you simply really feel sorta okay about the place it’s going, however nobody else has seen it but. And so it would suck. However you’re truthfully undecided. I acquired acquired to this about 2 years into writing the guide. I had written the primary ~10 chapters (out of 35), and I had a number of questions for myself:

• Is that this guide any good?
• Am I saying stuff individuals already know?
• Or is that this guide too nerdy, and going into particulars which might be pointless?
• Are the tales really fascinating, or too apparent? Have individuals heard them already?

And to be trustworthy, you form of don’t know till you’re taking a half accomplished model of the guide and ask just a few trusted mates to learn it. I acquired a bunch of very superb suggestions — thanks particularly to Lenny Rachitsky, Sachin Rekhi, and many people at a16z for taking the primary crack — and it was additionally the primary time my writer and agent had been studying it. I acquired a bunch of helpful conceptual suggestions, for instance that the primary few chapters felt slightly gradual to get into the motion. It felt too theoretical at components. There have been sure particular subjects that felt trite. Some sections felt repetitive. And so forth. Brutal honesty is what you want right here. Ultimately I additionally felt like, beneath the scruff, was a guide that I’d actually take pleasure in studying myself, and that it simply wanted to be tightened.

I’ll say, probably the most painful refactoring taking place on this interval. As I neared completion of tough variations of all the varied chapters, I ended up with a roughly 100,000 phrase guide (which is regular, seems). Generally it takes 3-5 years to completely get to this, and the very fact I had a demanding day job and was capable of end in ~3 years — that’s nice. But when my fear was that if I needed to considerably rewrite parts half manner via, it might grow to be a 5 years course of, which I’ve realized is just not unusual. This sort of refactoring notably comes when you may have a full size guide and then you definitely determine to mix a chapter or two. Or to take a theme that’s showing in just a few spots, and make it into its personal part. After which it’s a must to replace every little thing within the guide in order that it flows correctly. It’s best to do with a weblog publish, or a doc, or one thing like that, however with a 35 chapter guide — that turns into a heavy raise. However so it goes.

Closing months: Simply ship it already
By the tip of the writing course of, I used to be useless drained. Truthfully, I acquired to some extent the place I used to be each concurrently feeling good in regards to the supplies, notably the primary few chapters that I had polished up. But additionally the method was lengthy and arduous and I used to be prepared to simply ship it. The issue with books, nevertheless, is that they’re actually developed in a waterfall course of for good motive — when you submit the guide, and it’s printed, that’s that!

One enjoyable backwards and forwards occurred as I began to work on selecting the ultimate cowl. I labored with a designer who had carried out a number of work on Stripe Press books, which I at all times cherished — nevertheless, they’re boutique operation which supplies them a number of latitude on what may be carried out, and the designs typically had very small textual content on the duvet (in any case, the title can be someplace on the internet web page in a digital-first expertise, proper?), or prescribed bizarre supplies. It was a negotiation to determine what was really potential.

I additionally realized that the majority the US hardcover books are printed at one firm (loopy???) and right here’s an excerpt about that from a Vox article:

Most guide printing occurs within the US. Books with heavy shade printing, like image books, are despatched to China, however as a way to hold the price of delivery low, most publishers do the remainder of their printing domestically. That’s getting an increasing number of tough to handle.

Till 2018, there have been three main printing presses within the US. Then certainly one of them, the 125-year-old firm Edwards Brothers Malloy, closed. The remaining huge two, Quad and LSC, tried to merge in 2020, however then the Justice Division filed an antitrust lawsuit. Quad responded by getting out of the guide enterprise totally; LSC filed for chapter and offered off plenty of its presses. Smaller printers have continued to function, however the infrastructure to maintain up with the demand for printed books in North America is in shambles.

Loopy proper? Couple different fascinating issues I realized on the finish:

• You solely want ~10,000 preorders to be a bestseller — far under what it was
• There are tons of books that grow to be best-sellers as a result of individuals purchase many, many copies of their very own books — typically through nameless networks of patrons to obscure what’s taking place (I didn’t do that, btw)
• The US is just not really the first marketplace for enterprise books, no less than by models — it’s China. There’s normally 3:1 ratio of books offered there versus the US
• The typical guide has 250-500 books offered in its lifetime (!!!) — and possibly the median is extra like just a few thousand. However both manner it’s fairly low

Anyway, as the ultimate months approached, I traded a bunch of revisions with Hollis and her workforce at Harper Enterprise. Regardless that I used to be very drained at this level, I had the unimaginable assist of Olivia Moore at a16z who did a as soon as over on the finish, that actually polished issues up, in addition to my agent Chris, and plenty of others. There are manner too many individuals to thank, so I encourage you to take a look at the acknowledgements 🙂

It was solely within the ultimate months that I began to consider advertising and marketing the guide. I even have a ton of notes there 🙂 Will share extra later. Within the meantime, hopefully y’all discovered this fascinating! It was a very good 3+ years of my life and there’s lastly sufficient distance to mirror now.

PS. Get new updates/evaluation on tech and startups

I write a high-quality, weekly e-newsletter masking what’s taking place in Silicon Valley, centered on startups, advertising and marketing, and cell.