“Write drunk; edit sober”

A quote often (and probably inaccurately) attributed to Ernest Hemingway. And if you take the quote too literally, you’ll miss the power of what it teaches.

We have at least two sides when it comes to creating something.

On one, we see endless possibility. We can create anything our minds conjure. The muses are everywhere.

On the other, our brains are great at tearing down all the bullshit, and finding the kernels of what’s efficient. What’s practical. What’s actually good. And that side often doesn’t like what it sees of the other.

When I create, I try to take “Hemingway’s” advice.

To begin something I want to create, a blog post, a software feature, a YouTube episode, I’ll start from a thread of optimism. A motivating book. A TED talk that has me inspired to teach. A workout where I was able to push just a little further than last time.

I hype myself up to get closer to that feeling where anything is possible.

From there, I throw tons of stuff onto my canvas. I’ll write pages of run on sentences. I’ll code just to get the idea working. It’s not the best stuff. It’s not even good. But the goal is to get something, anything on the page.

Eventually, I’ll take a break. I’ll get a solid night’s rest. A walk. Lunch. Something to mark the change because it’s time to switch modes.

I start to edit.

I take all these things I created and pare them down.

What you see as 500 words today, started with 1200. The code you see tomorrow is the result of two dozen versions.

One thing I’ve noticed about editing is that I know I’m getting closer to something decent when the editing starts to hurt. When it feels like I’m starting to cut bone.

I start to throw away the things that I had previously considered good, in order to make room for what’s sitting on the canvas now. I remove that anecdote I was originally convinced HAD to be there. Or I realize no one is going to need this feature after all, even if it was the thing that had me most excited to start.

My best work comes when I balance my two selves — the one who can do anything and the one who’s my strongest critic.

When I find a way to invite both of my selves to a project, but make them work separately, that’s when things get really good.

P.S. Please help spread this by clicking the below.

You should follow my YouTube channel, where I share more about how we run our business, do product design, market ourselves, and just get through life. And if you need a no-hassle system to track leads and manage follow-ups you should try Highrise 🙂


Get to the point

It’s the single best way to improve your day-to-day writing.


If you want to consistently improve your everyday writing, there’s one really straightforward thing you can do…

Get to the point.

When you get to the point quickly, your writing becomes instantly clearer. Clarity makes your writing easier to understand, easier to retain, and more enjoyable to read. All of that makes your readers happy.

Here are a few pointers in how do do that.

Avoid the long windup

The long windup is the most common (and most painful) mistake I see when reading.

Common offenders are things like a grand introduction about yourself, a discussion of why your post is important, a massive outline, or a long setup story that buries a one-line reveal.

You don’t need any of that, and your readers don’t want it either.

When readers run into a long windup they either 1) skim to find the main point or 2) just leave. Either way they’re irritated and are probably going to miss the valuable parts of your writing.

So let’s avoid that by keeping the following in mind:

  • You shouldn’t have more than a couple short paragraphs before you’ve stated your main point. If you do, start editing.
  • If you have any of those offenders I mentioned above, cut them and re-read your draft. I’ll bet it’s far clearer and more effective.
  • Assume your readers are there for a reason. If they’ve clicked in, they’ve already expressed interest. Give them what they want, not a bunch of setup.

Here are a couple good examples of short intros and getting to the point:

A direct, well-written pull request by Conor.

Work ethic by Jason F.

Edit for your audience

While writing and editing, you should repeatedly ask yourself two questions…

  1. Who, specifically, am I writing this for?
  2. What do I want them to know when they’re done reading?

If there’s anything in your writing that doesn’t support your answers, it’s time to get editing. You’ll find yourself getting to the point a lot faster and more effectively.

For example, let’s say your answers are…

  1. I’m writing this for Ruby programmers
  2. I want them to learn a few tricks I’ve learned over the years

Bam. Your writing is instantly narrowed in focus. You can tailor your writing specifically to programmers familiar with Ruby. And you can jump right into the tips and tricks, not a bunch of basics about the language.

Be direct, not incomplete or cold

I want to be clear — getting to the point/being direct is not the same as being cold, unfriendly, or incomplete.

Getting to the point shouldn’t come at the cost of watering down your supporting case. Be direct and get to your point, then support your case and tell a compelling story. Be mindful though — don’t add fluff. Be precise.

Along the same lines, being direct doesn’t mean you can’t be friendly (especially in messages or emails). Being warm never hurts — how you say something still helps in getting your point across.

Practice (at the right times)

Don’t worry, everyone struggles with getting to the point sometimes. Every early draft I write has extraneous fluff that ends up getting cut.

The good news is that I (and you!) have plenty of opportunities to practice every day. Every time I write — a message in Basecamp, an email, a pull request, or a blog post — is a chance to keep working at it.

But let’s be honest, I’m not constantly working on it. There are times where I need my writing to be sharp, and other times I just need to get it done. That’s OK!

It’s hard, time-consuming work to do your best writing. So don’t worry about doing it all the time. You don’t have to polish every piece of writing. Instead, pick your moments and really try to nail those.


Remember, getting to the point means greater clarity, and clarity is king when it comes to writing. It saves time, avoids confusion, and enhances comprehension. Combine that with a strong supporting case and a friendly tone and you’ve got writing gold! ✍️💰

If this article was helpful to you, please do hit the 💚 button below. Thanks!

We’re hard at work making Basecamp 3 better every day. Check it out!

The writing class I’d like to teach


During Q&A at a conference I spoke at a few years back, someone asked me “What’s your take on the true value of a university education?” I shared my general opinion (summary: great socially, but not realistic enough academically) and ended with a description of a course I’d like to see taught in college. In fact, I’d like to teach it.

It would be a writing course. Every assignment would be delivered in five versions: A three page version, a one page version, a three paragraph version, a one paragraph version, and a one sentence version.

I don’t care about the topic. I care about the editing. I care about the constant refinement and compression. I care about taking three pages and turning it one page. Then from one page into three paragraphs. Then from three paragraphs into one paragraph. And finally, from one paragraph into one perfectly distilled sentence.

Along the way you’d trade detail for brevity. Hopefully adding clarity at each point. This is important because I believe editing is an essential skill that is often overlooked and under appreciated. The future belongs to the best editors.

Each step requires asking “What’s really important?” That’s the most important question you can ask yourself about anything. The class would really be about answering that very question at each step of the way. Whittling it all down until all that’s left is the point.

Maybe one day.

Transforming a screen with a few questions

Some scraps and learnings from the workroom floor

Recently I’ve been working on an update to Basecamp 3 that’ll give people a couple more options for how they sign up and log in to Basecamp. People will be able to choose from using a password, their Google account, their Facebook account—or all three.

Huh, why write about that, there’s nothing special there? I know, right!? I’m not sharing this because there’s anything groundbreaking in the feature, but because I found myself re-learning the importance of looking at my work critically, and honing things in by asking and answering a few simple questions along the way.

So, as we start offering multiple methods to log in to Basecamp, we need to also provide a place for people turn each method on or off. Let’s dive in and take a look at my first iteration on this screen.

Getting the mechanics in place

I started by carving out a screen to hold all the login options that are available to a person. I was thinking about this screen as an unfortunate (but necessary) evil for letting people toggle these login options.


Take one look at this first iteration and it’s clear that my attitude was showing through 😳. Sure, you can turn Google off. Yes, you can start using Facebook. Yeah, you can figure out which options are on or off…you just have to read every single word on the page!

The mechanics were there, but there was no soul, and no clarity.

What is this really about?

I’m still not clear what I’m looking at here or what the options actually are… Ok, I finally figured out I’m using Google…

…right now there’s a lot of parsing to do. You could have Facebook on, but you’d have to scroll through two other options first that are calling you to do something. It’s unclear if this screen is informational, or transactional, or both.

After getting the above feedback, I realized I hadn’t really stopped to think on what this screen was really about. Sadly, this is a mistake I’ve made many a time! It can be all to easy for me to start with the mechanics and then just leave it there.

So I asked myself, what is this really about? After a bit of thought, it was clear that I was hoping to do two things:

  1. Give current state of the world
  2. Offer any additional options

Explaining the current state of the world was something I was already trying to do in each individual login method’s copy, but totally failing to express holistically. If these two things were really the purpose of this entire screen, why not structure everything around them?


A big step forward. Now the screen tells me what’s set up right now, and what my other options are. But…it still lacks clarity and precision! I have to read the entire thing to actually understand it.

The first time around I missed the forest for the trees—state was expressed in the individual blocks, but not holistically. This time around I missed the trees for the forest—the screen has the right structure, but the blocks are confused and seemingly unaware of their context.

I mean, who wants to read three paragraphs just to get a sense of what there current options are?

How can I make this clearer?

Now that the screen’s structure reflected its purpose, it was time to tighten things up and double down on making it as clear and precise as I could.

Could I say this more efficiently?
The first step towards clarity was to remove a lot of words.

“Right now you can log in with…” becomes “Today, you log in with…”

“You can add the option to log in with…” becomes “Add more ways…” because folks already know what we’re talking about.

The paragraphs describing how “you’re set up to login…” for each active login method are just repeating what the heading tells us. Drop those.

Sometimes words are too wordy, and an image can speak more efficiently. Adding icons for each login method made it easier to see what’s on/off without even having to read beyond the headings.

Oh, and speaking of headings, we don’t need that “Login options” title up top anymore — gonzo!

(oops, just noticed that “login” should be “log in” here!)

The final (for now) product is a lot better. The whole and the parts are working together. It’s clear what the state of the world is, and I can easily see my other options. The words are precise and clear.

The questions I asked along the way

Like I said, no groundbreaking designs here, but I was surprised at what asking a few questions did to transform this screen.

Here are the questions I used along the way:

  1. What’s this screen really about?
  2. Is the overall screen oriented around that purpose?
  3. How about the individual building blocks?
  4. Could I say this more efficiently?
  5. Is there anything I could remove (or add) to make this clearer?





If you haven’t already, head over to Basecamp and start a free account. You can get going with either a password or your Google account—we haven’t gotten to Facebook yet.