Night Water: Year Four
A quantitative look at the fourth year of this newsletter
On Wednesday, February 12th, Night Water turned four years old. I took it the pediatrician and she confirmed that it’s hit all of its developmental milestones—catching a large ball most of the time, naming colors of items, and pretending to be something else during play.
If you’re relatively new here: I started Night Water because I love blogging and wanted to force myself not only to write but to just get that writing out there into the world. Over time, it developed into a regularly scheduled newsletter, with branding and lore and subscribers who weren’t my friends and family. My first post, a fundamental text on Babyland General Hospital, went out to 15 people. This post is going out to over 1600.
This year, I helped answer questions from those subscribers like “Why is Noah Kahan so popular?” and “What is this motorbike singing to me?” I introduced them to radio stations like Shonan Beach FM and Antioch Broadcasting Network, and even subjected them to my own Scrubs-themed radio show. And, as always, I put my body on the line to test out the best and worst vegan fast food so my readers don’t have to.
I’ve always appreciated it when the newsletters I read pop the hood up and let readers take a look at the engine, which is why, every year now, I do the same to Night Water. So let’s take a look at the fourth year of this newsletter through four big questions.
How much Night Water did we get this year?
Here’s how it all broke down:
- 40 Night Water emails
- 3 email reruns
- 38 new public posts
- 1 web-only post
- 14 Day Soda posts
- 1 guest post
- 2 podcast episodes
- 1 Scrubs-themed radio show
Emails: Last year, I changed the promised Night Water schedule from “every Tuesday” to “most Tuesdays.” It was a small change meant to make the publishing schedule more sustainable for me. That’s worked out for the most part—I still delivered 40 Night Water emails, including 3 reruns, a solid 77% of the year.
Day Soda: I wanted to focus more on bringing value to paying Day Soda subscribers in Year Four, and I think (hope??) I’ve done that. Day Soda subscribers got an additional 14 exclusive posts this year, so if you were a Day Soda subscriber for the full year, you got 54 emails total—104% of the year. Plus, it’s all available on the web, so new Day Soda subscribers can go back and read that full exclusive archive.
Web-only Post: While all of Night Water is available on the website—and many people read it there!—I haven’t done much web-exclusive content in the past. I experimented with one web-only post this year—November’s “Swifties have shown us the future of Bluesky”—for a few reasons:
- It was time-sensitive. There was a mass migration of X users to Bluesky post-US elections, and I wanted to ride the wave of excitement around that without waiting for the regular Tuesday publish date.
- It was a niche topic. While I’d written about Bluesky in the main newsletter in the past, there’re only so many times you can write about the technical features of decentralized social networks before getting self-conscious.
- I was curious if anyone would read it without the distribution power of the newsletter.
On that last point, at least, it was a major success—it was the second-most read new article last year, and fifth overall, and also earned Night Water a mention in one of my favorite newsletters, Garbage Day.
Podcasts: Based on listenership, it seems like you guys don’t care for the podcast episodes. That does not matter to me in the slightest, for what is a straight white male without a podcast? Nothing, that’s what.
How many people subscribed to Night Water?
Night Water started Year Four with 906 subscribers and ended it with 1660—that’s 754 new subscribers, or an 83% increase. I was aiming for between 2000 and 5000. Somewhere in that range, if a certain percentage of those folks converted to paid subscribers, there would be enough revenue to support this newsletter and help it grow.
This year’s Night Water anniversary not only marks four years of the newsletter, but one full calendar year of being hosted on Ghost. After spending a lot of words telling people to get off Substack, I had to put my money where my mouth was and move Night Water off the platform. That meant literally paying for hosting, amongst other small costs, and giving up Substack’s network effects. Those network effects are probably overplayed—I spent quite a lot of words making that argument in last year’s annual wrap-up—but I had to find a replacement if I wanted to aggressively increase the free subscriber base.
For that, I turned to paid advertising.
Meco: Meco is a free newsletter reading app for iOS and Android. There’s a Discover page in the app for newsletter recommendations—as a newsletter, you can pay to be featured here. In theory, this is a great place to find new subscribers. It’s a captive audience who love newsletters enough to download a whole separate app for them. But the flip side of that is that if Meco is not sticky and they stop using it, that subscriber just falls off the map. The worst case scenario is someone downloads Meco, excitedly subscribes to a bunch of newsletters, and then never opens Meco again.
Based on my numbers, that scenario plays out more likely than not. I’ve paid $144.50 for 289 subscribers ($0.50 a pop). 269 of those members are still subscribed to Night Water. Seems good on the surface—low churn!—until you look and see that 60% of them have never opened a single email. Yikes. Only 19 subscribers have open rate of 50% or higher.
Meco introduced a new ad product late last year that applies a level of engagement pre-screening before a subscriber is added to your list. That should help cut down on short term dead weight. Otherwise, you just need to be prepared to cull a lot of inactive subscribers.
Refind: Refind is a curation service that delivers great articles in subjects you’re interested in via email or their app on a schedule you choose. It’s an über-newsletter, a links email on steroids. They have a massive user base who have given the company clear signals about the topics they’re interested in. That makes it relatively easy to advertise to them without using any creepy, invasive targeting.
Initially, like Meco, Refind would charge you for every new subscriber, regardless of whether or not they opened emails or immediately unsubscribed. But in the middle of last year, Refind tweaked their model to only charge for engaged subscribers—those who stayed subscribed and opened at least one email in the first seven days. Of the 501 subscribers I've gotten through these ads, 37.3% were "engaged," and at $0.50 per engaged subscriber, you can see how the savings add up.
Some of those “unengaged” subscribers do eventually engage. Out of 501 initial subscribers, 421 are still subscribed, and only 161 have never opened an email. That’s only half of everyone who counted as unengaged—and that I didn’t get charged for. On the other end of the spectrum, 95 subscribers have an open rate higher than 50%. Again, that’s roughly half of the total number of engaged subscribers I paid for.
Between Refind and Meco, I’d definitely suggest experimenting with Refind, especially if you’re a newsletter that, unlike Night Water, sticks to a particular topic that would appeal to their audience. Will any of these efforts actually lead to Night Water hitting some revenue goals? I still have to wait and see, but knowing which sources bring in long-term engaged subscribers is good first step.
How many people actually read Night Water?
One downside of aggressively growing your list, especially from sources where you know a good portion won’t engage with the newsletter, is that it will push your open rate down. Just take a look at this delightful line chart:
There’s a pretty clear trend there. Luckily, this is just a vanity metric for me—I’m not selling ads or proving my worth to some higher ups, so I don’t need to keep my open rate up for appearances. But if I did, it would be as simple as cleaning up all of those unengaged Meco and Refind subscribers from my list—something I’ve been meaning to do anyway.
Of course, open rate is not necessarily an accurate representation of who is opening your newsletter or even reading it. Plenty of email programs obfuscate that metric in some way for privacy reasons—Apple Mail, for example, reports every email you receive as open, while services like Hey and Proton do the opposite. You have to take the data with a grain of salt.
Night Water doesn’t just exist in email form, however—there’s a whole secondary audience of web readers, primarily coming from Google and social media sources. Night Water saw over 35,000 unique visitors in Year Four, with 64% of that traffic coming from Google alone. Around 10% came from social sources, with Reddit bringing in the most visitors.
Here are the top five Night Water posts of Year Four, based on web traffic:
5. The Monster at The End of This Book ending explained (617 pageviews)
This was one of my favorites of the year, a stupid little parody of those “ending explained” videos you see on YouTube for every movie, regardless of how straightforward the plot is. Weirdly enough, the vast majority of views have come from Google, suggesting that there is a significant portion of the population that truly needs this Sesame Street book explained to them.
4. Time is running out for the perfect Austin Powers legacy sequel (782 pageviews)
Listen, I’m glad the message is getting out there, but at the end of the day, I only need one pageview: that of Mr. Michael Myers himself.
3. Meet the developer saving the iPod’s click wheel games (976 pageviews)
Reddit provided a little under 30% of the total pageviews here, which adds up—I found this developer’s preservation work through Reddit, and there are several active iPod enthusiast communities there. My only regret is getting slightly scooped by Ars Technica, so I wasn’t first out the door with my interview.
2. Swifties have shown us the future of Bluesky (1226 pageviews)
As I mentioned above, this was a web-only post, and it rode the wave of organic web traffic almost to the top. 28% of readers came from Reddit, where there is a large Bluesky enthusiast community, and 22% came from the mention in Garbage Day. Oddly enough, a mere 3% came from Bluesky itself, though it’s possible some Bluesky traffic got categorized under Direct.
1. It’s time to bring back the iPod (5042 pageviews)
If there’s one thing that writing Night Water has taught me, it’s that people love retro audio devices. The vast majority of this traffic came in one fell swoop shortly after publication. Around 42% of visitors came from Google—primarily from Google Discover, an articles recommendation engine that’s part of Google Search. Another 24% came from Hacker News, where it hit the front page for a short period.
It’s not just new posts that get web traffic, though. “It’s time to bring back HitClips,” a post from 2021, got 5955 pageviews last year alone, primarily from Google Search. Google continued to send tons of traffic to older posts last year, posts like “We need to talk about bedside water carafes” (4668 pageviews) and “Every vegan thing I ate at Greggs this month, ranked” (1526 pageviews).
Google reports that results from Night Water showed up in over 2 million searches in Year Four. The top query? “Lego loco.” I knew I wasn’t the only person who missed that LEGO train game.
What's next for Night Water?
It’s a pretty grim time for the open web, not gonna lie—the AI ouroboros is slowly but surely reducing the web into its own waste product, and we’re all swimming in it.
But—and this might just be me blowing smoke up my own ass—I think that just makes human-written blogs and newsletters like Night Water more valuable. I certainly value them more. Whether it’s a single-author newsletter like Where’s Your Ed At or a collective like Hearing Things, I see independent media as an oasis in a desert of AI-generated listicles.
I hope Night Water is like that for you. My primary goal here is entertainment, to be fun, weird, idiosyncratic, and most of all, refreshing. I still have subscriber and revenue goals and all that—it’d be great to at least break even, even better to be able to grow Night Water a little bit and get back to featuring more guest writers—but I’m not stressing too hard about that. As long as someone wants to read Night Water, I’ll be here writing it.
