How Not to Write a Book
I'm thrilled to announce that my book, Lean Customer Development, has been published by O'Reilly and is available now!
It's a bit ironic, because writing this book is probably the least "Lean" thing I've ever done. Even having a baby (which I also did in the past year) was faster!
While I did work with many of you to understand your needs and get feedback and iterate (oh, the iterating!), it's taken a long time to get the product out the door -- much longer than any "product" I've ever launched before.
I think I wrote a pretty great book. I'm confident that will help a lot of people be more rigorous about testing their company, product, and feature ideas. But, of course, that's still a hypothesis that will only be validated once customers actually start buying it!
Would I do this again? I'm not sure. But in case I do, I've learned a lot about the process -- what I'd repeat, and what I'd change.
Here's what I did right:
- Iterated on the proposal. I had a few friends (including Eugene Kim) read my initial proposal, and their brutal feedback made it much, much stronger before it ever saw an editor.
- Pitched multiple publishers. It was amazingly exciting to be offered a book deal by another publisher, but if I'd taken the first offer I'd gotten, I would've missed out on O'Reilly! Being part of Eric Ries' Lean Series at O'Reilly has clearly been a better "fit" for the book. It gave me a lot of support from like-minded authors as well as strong PR support.
- Identified my target customer (reader) profile quickly, then validated it. Having a strong vision for who would benefit from this book helped me shape the content. In particular, it was important that the book be helpful for people with no user research or customer-facing experience. Without a strong sense of the reader, it would've been easy for this book to drift too much into "user research" territory.
- Finished my first MVP quickly (an outline of the planned chapters) and got feedback. Even though the proposal was accepted, I still showed the chapter outlines to a bunch of potential readers and collected email addresses. (Not as good as pre-orders, but an okay indicator of potential purchasing).
- Pivoted after doing customer development interviews. Yes, of course I did customer development interviews for a book on customer development! Initially the book was much more focused on startups and the needs of small rapidly-changing companies. But a few conversations with people in larger enterprise companies made me realize that they had some major pain points! I changed my target reader profile to be more focused on people with existing products and customers, which impacted the content I wrote.
- Sought out unique stories to tell. There are a handful of companies who have served as "lean startup success stories" -- that have been so successful that they're referenced repeatedly. I wanted to make sure that my examples were from a different crop of companies and organizations.
OK, that sounds good, right?
But here's what I did wrong, wrong, wrong:
- Procrastinated. Oh man, did I procrastinate. Yes, I completed the initial MVP quickly and did some iteration. But then I just... stopped. I let myself get paralyzed thinking about the amount of work needed and did very little writing for months.
- Didn't work in small enough "batch sizes". I wrote an outline and made changes after talking with my target customer. But after that I was inconsistent in seeking out feedback. I should have made a plan to talk to my target reader on a more regular basis -- biweekly or even monthly -- instead of waiting until I had entire chapters done.
- Worked alone. I should've found a writing buddy who would hold me accountable to writing each week. When I was about halfway done, O'Reilly paired me with a wonderful development editor who not only helped me with deadlines but helped with structure. But I should've found my own someone sooner -- they wouldn't even have had to read my writing, as long as they would know when I hadn't met certain word count or page count.
- Wrote out of order. I started with the chapter on "the interview" because it was so central to the book, and then I jumped around. But by not starting with chapter 1, I wasn't starting with the problem statement or the target customer profile! In my head, I knew what problem I was solving and for whom. But I hadn't written it down. I ended up rewriting the entire Preface, Chapter 1, and Chapter 2 twice. If I'd started with a Chapter 1 MVP, I think I could've avoided a lot of rewrite time.
- Procrastinated. Did I mention procrastinating? The O'Reilly author advice is to spend some time writing every single day. With a family and a full-time job, that wasn't realistic. But it took me too long to figure out what type of schedule was manageable. (One weeknight writing after work plus writing most of the day Sundays, as it turned out.)
Any way about it, I'm glad to be back to writing here on the blog! I'm looking forward to upcoming speaking engagements to promote the book and, of course, to continue teaching customer development principles both within my full-time employer Microsoft and lots of other companies looking to make their mistakes faster and cheaper.