Writers who operate.

Occasionally folks tell me that I should “write full time.” I’ve thought about this a lot, and have rejected that option because I believe that writers who operate (e.g. write concurrently with holding a non-writing industry role) are best positioned to keep writing valuable work that advances the industry. This is a lightly controversial view, so I wanted to pull together my full set of thoughts on the topic.

The themes I want to work through are:

• Evaluating believability for operators is much easier than for non-operators
• The pursuit of distribution changes what and how authors write (e.g. pulls towards topics that are trending)
• How writing full-time anchors you on writers and audiences, whereas part-time writing allows a third balancing perspective (the folks you work with in the context of your industry work)
• Invalidation events happen in industry (e.g. move from ZIRP to post-ZIRP management environment) but it’s difficult for non-operators to understand implications with conviction
• Operating is an endless source of new topics (e.g. the topics in An Engineering Executive’s Primer are the direct outcome of my operating)
• Part-time writers can still get better at writing, although maybe slower than full-time writers

I’m not particularly interested in convincing someone else whether this is the right choice for them, but hopefully at the end you’ll understand my perspective a bit.

Examples

There are many writers out there who fit into the “writers who operate” archetype. A few examples: Charity Majors, Dan Na, Eugene Yan, Hunter Walk, Tanya Reilly.

Venture capitalists use “operators” to indicate folks who’ve worked in industry as opposed to in venture, but I don’t make that nuance here–working in venture capital is “operating” in my usage. Similarly, you could try to cohort various writers by the volume of their writing, but that’s not too important to me–someone who hasn’t written anything in the past three years probably isn’t who we’re talking about, but generally this is a broad church.

Believability

Believability is a Ray Dalio and Bridgewater idea, and experiencing some public scrutiny of late (e.g. Bridgewater Had Believability Issues), but at its core the observation still rings very true: we should weigh advice more heavily from folks who we have reason to believe. Cedric Chen has a few tech-centric pieces on Believability, that are interesting reads: Believability in Practice and Verifying Believability.

First and foremost, I appreciate writers who operate because they directly experience the consequences of their choices. Cedric’s second piece tells the story of “Q”, a widely read tech leader who’s had a mixed career, as an example of needing to verify believability. I agree with that observation, but the only reason we’re able to evaluate the advice at all is because that writer is an operator. If they weren’t an operator, we wouldn’t be able to evaluate their believability at all.

Operating is, for me, remaining accountable for what I write. What I write is a pretty direct reflection of what I believe and how I operate at the time that I write it.

Distribution shapes writing

As you watch new writers come onto the “scene,” you’ll often notice a shift from a genuine passion in a given niche to engaging in topical events and controversy. The reality is that it’s exceptionally hard to write something that generates a lot of discussion, and it’s even harder to repeat that formula consistently. After folks have the experience of writing a popular piece, they often get sucked into the desire to produce more, and this ultimately means seeking wider distribution.

Reliable distribution is a hard thing to find on the internet, and one of the most obvious opportunities for distribution is to engage in controversy. Write something controversial, engaging in an existing controversy, subtweet someone who did something dumb, whatever. The problem with this is that it pulls you out of picking topics, and instead towards picking positions.

Ultimately, I don’t believe you can say anything particularly novel or interesting in reaction to a trending topic. There are certainly takes that are more or less nuanced, but mobilizing the base is not advancing the industry.

This problem is even more acute when you’re trying to make a financial living out of your writing, because matching your message to your audience becomes that much more important. You’re going to spend even more time tuning your messaging to resonate with what the audience currently believes than you are on writing something new.

Taste is tribal

A year or two back, Brie Wolfson wrote a very compelling take on taste, Notes on Taste. Reading those notes, I want nothing more than to identify as someone with taste. However, perhaps out of jealousy, I’m a bit of a taste-skeptic. I view taste principally as tribal, and find that identity-through-taste is a frequent driver of boring takes and perspectives.

As an example, think about Marc Andreessen’s recent The Techno-Optimist Manifesto. Regardless of how you personally feel about the manifesto, I’m confident that you know exactly how you’re supposed to feel about it within each of the various tribes you participate in. Further, I’m certain that you knew what you’re supposed to feel about it without even reading it. That’s not a recipe for interesting discussion.

This is particularly hard to navigate as a full-time writer, because you’ll become more focused on the tribes of other writers and the tribes of your audiences, and your standing in both is important to your success. As an operator, those tribes will matter to you, but fitting into their expectations is not essential to your success (and your survival, if this is your primary source of financial stability). There are, of course, other tribes you have to pay attention to from your operating work, but those tribes will vary across writers, such that in aggregate they allow for a broader expression.

Invalidation events

In 2020, Ranjan Roy wrote ZIRP explains the world, which is an interesting dive into how zero interest rate policy was shaping so many dimensions of the economy. Among other things, ZIRP created the conditions for hypergrowth companies and funded the industry’s shift towards larger teams driving revenue growth rather than margins. People operating in the industry today have felt this transition in layoffs, a slower hiring process, and a notable shift in the dynamic between employees and employers.

When I meet with industry peers, we spend most of our time discussing either tactical problems related to this shift (e.g. how do we benchmark costs properly to justify engineering headcount) or wondering if we should hide in a hole for several years hoping that the industry reverts to kinder time. Despite that, I see a large swath of folks pitching ZIRP-era content and strategies to struggling leaders.

The folks still making their ZIRP-era talking points aren’t bad people, but they are giving bad advice, and it’s because they’ve failed to recognize an “invalidation event.” Good advice is grounded in accurately diagnosing circumstances, and folks operating in the industry are best positioned to update their advice because they’re directly experience the industry’s changes rather than observing them from a distance.

It’s not that non-operators don’t detect these shifts, they certainly do, but it’s exceptionally challenging to quickly build confidence in a large change when operating on second hand information. Operators get a lot wrong too, but it’s my experience that self-aware operators will get direct information earlier and be in a better position to evaluate it.

Endless topics

Writing as an operator, I have a constant source of new topics. More than just any topics, these topics are the most challenging topics that engineering organizations and companies encounter. All three of my books are directly grounded in the topics I was struggling with at the time. An Elegant Puzzle focused on the challenge of managing within a hypergrowth company. Staff Engineer documented the various ways that senior engineers were finding leadership impact outside of management roles. An Engineering Executive’s Primer tracks what I’ve learned from operating in executive roles. There’s no way that I personally could have written these without the benefit of operating in those environments.

Conversely, I see folks who leave operating roles often fall into a rut of repeating topics. They want to say something, but they’re not encountering new problems, so they fall back onto their fixed experiences in the industry and come back with the same ideas.

Writing well and frequently

Occasionally folks make the assertion that it’s hard to improve as a writer if you’re only writing part-time. There’s a kernel of truth in this observation: writing up my notes on finishing my 3rd book, Primer, I described each book that I write as a separate education. Even on my third book, I’m still learning so much about how to write books. I’m not sure the ideas are getting better, but the books containing those ideas certainly are.

That being said, I’ve found that having the space to explore in my writing has created so much room for improvement that I wouldn’t have found writing under a structured publishing schedule. Free-form writing has allowed me to write when and where I have energy, and to stop writing where I don’t have much energy (e.g. I starting work on Infrastructure Engineering and then subsequently paused it). It’s also allowed me to experiment with formats and mediums: I’ve written this blog, written books, spoken at conferences, done a YouTube recording, and so on. If I was focused on very specific outcomes, I’d likely be experimenting less and trying to “exploit” the mediums more, which would focus my learning.

It’s possible I would have improved more as a writer if I did it full-time, but I’m confident that I’m not a meaningfully worse writer due to the part-time nature of my writing. I also lightly hold the belief that I’m a better writer as a result of not writing full-time. Writing on a schedule is, in my opinion, not at all fun. Further, most of my best writing is stuff that I originally think isn’t even worth writing down, which would translate poorly into a world where I need to predictably write good stuff.

Echoing my earlier comments, not trying to convince anyone to switch sides on this topic, and many non-operating writers are quite good. There are many techniques you can use to address the above topics (e.g. maintaining an active network in industry), but generally those techniques apply equally (or better) to writers who operator (e.g. writers can probably get access to any company in the industry, but you couldn’t convince me that’s not equally true for operators outside of–maybe–getting visibility into a small pool of direct competitors).