2025 Year in Review

Welcome to the annual review for Ned in the Cloud! Believe it or not, this is the eighth installment of the “Year in Review” series. Frankly, I’m impressed with myself for having kept this up for almost a decade. Going back to that first post, my goals for 2018 are hauntingly familiar:

• Build a successful cloud practice
• Learn to program in Go or Python (or both!)
• Keep creating new Pluralsight courses
• Blog on a weekly basis
• Podcast on a weekly basis
• Speak at more user groups and conferences

Granted, that first bullet point Build a successful cloud practice doesn’t really apply anymore. In 2018, I was working at a VAR and trying to build up our cloud services offerings. I quit that job in 2019 to work for myself. Everything else? Still applicable! I supposed the fact that I am a full-time content creator and technical educator makes things like creating content and speaking in public a pretty consistent set of goals.

If the theme of my 2024 was acceptance, and the theme of 2025 was commitment. In some ways, I had left the revenue generating portions of the Ned in the Cloud business on autopilot while I tried out new things. In 2025, it was time to refocus on building revenue and updating my aging content. As we’ll see shortly, I ended up doing a LOT.

High-level Review

I won’t share actual revenue numbers, but Ned in the Cloud saw a revenue drop of 19% YoY. Since 2022, revenue has dropped every year. If I were a publicly traded company, people would be selling my stock and talking about hostile take-overs. But the thing is, despite these revenue drops, I am still making a very comfortable living from the company. As it’s only employee and sole owner, I really only have to answer to myself. Am I still happy with what I’m doing and making enough money?

Yes, yes I am.

What accounted for another drop in revenue? The biggest factor was Pluralsight revenue, with a drop of 8%. Day Two DevOps revenue also dropped in half! I’ll address each of these in their own sections.

The overall lessons are these:

1. Pluralsight revenue is pretty important and I’ve produced a ton of new courses this year to course correct (pun intended)
2. Podcast sponsorships are harder to acquire due to an abundance of supply
3. My YouTube channel needs some TLC

Let’s dig into these topics and more in detail.

Pluralsight Courses

Since the biggest chunk of my revenue comes from Pluralsight, I should probably start with that. First, a little context about how Pluralsight works for the uninitiated. When producing courses for Pluralsight, there are two primary ways to get paid. There is an upfront payment for completing a course, and a quarterly revenue share for the course based on viewership.

The general idea is that Pluralsight brings in X amount of money in revenue for subscriptions. Of that revenue pool, they determine how much of that revenue is from video courses. That revenue is divided by the total number of hours watched on the platform. You get a percentage of that revenue based on how many hours your course was watched and your revenue share percentage, which is negotiated when you sign up to build the course.

The bottom line? The more popular the course - i.e. the more hours it is watched; the more money you make from a course. The more popular Pluralsight is, the more revenue they generate, and the bigger the revenue pool is to distribute to authors.

I’m sure I’ve explained this before, but I won’t make you navigate to another link.

A few years ago, Pluralsight made two big changes. First, due to financial issues, they cut all author’s revenue by 25%. That was a sweeping cut across the board for all authors. Was I pissed? Hell yeah, I was pissed. You can read my feelings in this post.

The other big change was a lifetime earnings cap for a single course. The general idea was learners want fresh and new courses, and there were authors who had created a popular course many years ago and have been coasting on the revenue ever since. Pluralsight felt this was unfair to new authors, as these old courses were taking a significant share of the revenue pool without contributing back.

If we’re being totally honest, I think Pluralsight just wanted to force these authors to create a new version of the course with a lower revenue percentage. You can dress it up however you want, but the simple truth was authors created content for Pluralsight. People watch those courses on Pluralsight as part of their subscription. Pluralsight is making money from those courses, and they want to limit or eliminate how much authors were earning based on an arbitrary cap.

From a business perspective, I get it. Pluralsight has been through a lot in the last 5 years. Going public, going private, TechEd market consolidation, multiple rounds of layoffs, etc. It has been a bumpy ride financially and organizationally. The fact that they are looking to cut costs is not surprising, and the author rev share is a big cost, especially for courses that have been around for years and received no updates.

From a moral perspective, I don’t think it’s the right thing to do. These courses are still delivering value to Pluralsight, shouldn’t the authors continue to get paid?

The lifetime earning limit is actually pretty high, and only the most popular courses are likely to hit it. If you don’t have a course in the top 100, then there’s nothing to worry about. But you know who does have a course in the top 100 and wasn’t paying close attention? THIS GUY.

My course, Terraform - Getting Started, had been in the top 100 courses pretty much since I created the first version in 2017. I have refreshed the course about every 2 years ever since: v2 in 2019, v3 in 2021, and v4 in 2023. The v4 course was the first version affected by the lifetime earning limit, and wouldn’t you know it, in spring of 2025 I hit the cap.

There was no klaxon, no warning, no email or reminder. I looked at my quarterly payment and it was way off. That’s how I found out. Part of that is on me. I should have been keeping a tally on my own and tracking how close I was to the limit. But I also think it’s pretty fucked up that no one at Pluralsight thought to set up an early warning system for courses that are approaching the limit. For a single course that represents 50% of my income from Pluralsight, having that amount drop to $0 without warning was quite a shock.

I was already in the middle of finishing the Vault Associate Certification courses and couldn’t really stop to update the Terraform - Getting Started course. I had contractual deadlines I had already agreed to. So, I basically lost out on three months of revenue from my most popular course due to this cap. You want to know why my Pluralsight revenue was down 8%? This is why.

And it would have been much worse if I hadn’t kicked my course production schedule into overdrive this year. In 2025, I created 18 new courses for Pluralsight. That’s not a typo, EIGHTEEN new courses. That was nine courses for the HashiCorp Vault Associate certification, eight courses for the Terraform learning path, and one course for the foundations of computing.

Every single one of those courses had an upfront completion payment, and it was those payments that kept me from having a far larger dip in my revenue.

Entering into Q4 of 2025, things had stabilized. My new Terraform - Getting Started course was in place and the older version was retired. The course itself is shorter and I’ve taken topics from it and my Terraform - Deep Dive courses and spread them across the new Terraform learning path. It’s unlikely I’ll hit the lifetime learning cap for the course as quickly, although I am now keeping my own tally just in case.

I do not expect to keep up this frantic pace of course creation in 2026, although I have agreed to create courses for the updated Terraform Associate (004) certification. To be completely honest, I kinda let Pluralsight sit on cruise control for all of 2024, and in 2025 I had to make up for that.

YouTube Videos

You may have noticed that my YouTube video output dropped off severely in 2025. We’ll talk about actual stats momentarily, but essentially my focus on Pluralsight course creation kind of sucked all the air out of the room for other projects. I still managed to create a couple of videos around Terraform, but most of what I posted was interviews and livestreams. Things that do not require me to write a script or create a demo.

Don’t get me wrong, I was still writing a ton of scripts and demos. I was just writing them for Pluralsight and not YouTube. I have all my scripts in markdown files, and the total word count is about 350,000 words. I wrote 350k words about HashiCorp Vault and Terraform in 2025. I’m like the Brandon Sanderson of HashiCorp. Where’s my goddamn Shardblade?

ANYWAY

My point is that I had to shift my focus to Pluralsight to bring my revenue number back up. YouTube simply doesn’t pay well enough, even with sponsored content. Now that I’ve righted the Pluralsight ship, I can return my attention to creating YouTube stuff about Terraform and other things.

In terms of my top performing videos released in 2025, they are as follows:

What I glean from this is that people are most interested in practical Terraform demos and use cases, not theoretical stuff. That’s always been true.

My top videos overall for 2025 are mostly from previous years:

I still find it hilarious that my What is PXE Boot? video from five years ago continues to rack up the views. It was before I upgraded all my A/V equipment and put together a professional looking YouTube set. It’s just me, sharing what I’d learned about PXE boot because I was trying to get it set up in my home lab.

Just goes to prove, you never know what will be successful, but 101 content is always a safe bet.

As my overwhelming workload from Pluralsight dies down, you’ll begin to see more YouTube content being published. I really want to dig into Terraform Actions and Terraform Search. There’s also the Terraform Basics video series that could use some love. Stay tuned.

Livestream on MAIN

At the end of 2024, my buddy Marino Wijay posted about doing a learning livestream about AI. I said I’d be game to do one together, and thus Livestream on M.A.I.N. was born. The M.A.I.N. stands for Marino, AI, and Ned. I thought I was being clever. I’ll leave you to make your own judgement on that account.

Starting on January 16th, we began our weekly livestream to talk and learn about AI. The idea was pretty simple. Let’s integrate AI into an application and use AI to help us write it. In the process we can learn more about AWS Bedrock, Python, and Agentic AI. For the first seven episodes, that was all we worked on. Building out a podcast processing applications that used Amazon Bedrock, Lambda, and Python to process MP3 files from a podcast episode.

As Marino and I started getting busy with work things, we invited guests to come on and talk about their experiences building AI stuff. We had my Day Two DevOps cohost Kyler Middleton chat about the AI Slack bot she’s building for work. Later we brought on John Capobianco, Nnenna Ndukwe, Rizel Scarlett, Kat Morgan and others to talk about what they’re doing in AI. I learned a ton in the process and I hope you did to.

As for the application we were building? That never really got past the basic functionality stage, but it served its purpose. Marino and I learned a ton about how Bedrock functions and how to build out Lambda functions to interact with the service and different models.

Our last episode was on September 11 with Cat Hicks. We didn’t realize it was going to be the last episode, but I suppose we shouldn’t have been surprised. Conference season was getting into full swing, and Marino and I were super busy with our own stuff.

Will Livestream on MAIN come back in 2026? Maybe? Honestly, Marino and I haven’t talked about it and I’ve been too in the weeds with Pluralsight stuff to say for sure. It might actually be a good fit for the Packet Pushers network, which doesn’t currently have an AI focused show. Speaking of which…

Day Two DevOps

Day Two DevOps continues unabated for a sixth full year! We have racked up over 2.2M downloads since the podcast started in January 2019 and there’s no sign of stopping. While the focus of the podcast has changed somewhat over the years- as cloud computing has become less of a new idea and more of a given- we continue to discuss practitioner focused challenges.

Going through the episode downloads for 2025, I wanted to do a fair comparison to determine which episodes were most popular, so I used the number of downloads after 90 days. That necessarily means anything released after September won’t make the list. I’ll include them as contenders for 2026.

Looking at this list and the other episodes that attracted the most interest, I have a few conclusions:

1. People are still looking for explorations of deep technical topics focused on the practitioner
2. Security is still a major concern in 2025, especially with the rise of AI
3. AI by itself is not all that interesting

As Kyler and I plan for episodes in 2026, I think we need to focus less on AI for AI’s sake and more on practical solutions and dissecting practitioner problems. If you have a solution, problem, or technology you’d like us to explore, be sure to reach out.

Starting in 2026, Day Two DevOps will begin publishing full video for each episode as well. We’ve been capturing the video for a while and using some of it for YouTube Shorts, but didn’t have the necessary resources to edit and publish full episodes. Packet Pushers has recently hired two new full-time employees to assist with editing, and all the shows will begin publishing video.

That’s actually part of a larger trend with Packet Pushers expanding out to twelve podcasts in the network. It includes amazing shows like N is for Networking, The Cloud Gambit, and Life in Uptime. It’s exciting to be part of a growing network!

Unfortunately, the general explosion of podcasts across the industry has led to a drop in sponsorships for Day Two DevOps. There’s so many shows out there, all vying for a piece of the advertising pie while marketing budgets seem to be relatively static. The good news is that Packet Pushers is also bringing on a full time outbound sales person, something that is sorely needed with so many shows.

Outbound sales is not one of my core strengths. Hell, I’m not even very good at inbound sales, i.e. people coming to me asking to sponsor the show. I am intrinsically bad at sales. While at KubeCon this year, I heard Ethan pitch Packet Pushers to potential sponsors and I realized how unprepared I am for that portion of things. This is why you hire people. To do things you’re not good at, and honestly don’t want to be.

When it comes to Day Two DevOps, I think we’ve got the right hosts, the right topics, and a great reputation. I need someone else to tell that story better than I can.

Chaos Lever

Much like my YouTube presence, Chaos Lever fell victim to the over-abundance of work I had on projects that actually make money. Chaos Lever has always been a labor of love and has never earned any real money for me and Chris. We did it because we both like talking about technology and talking to each other.

In June of 2025, we decided to put Chaos Lever on hiatus for the summer. That hiatus has extended into the fall and now winter. We’ve been doing the podcast on a weekly basis for over three years and I just didn’t have the time to keep doing it. In 2024, I hired Humble Pod to help me edit and promote Chaos Lever, hoping that we would build up a big enough presence to attract sponsors. That didn’t happen and I had to cut Humble Pod loose due to costs.

As I said then, Humble Pod was doing an amazing job and it is no fault of theirs that Chaos Lever didn’t become an overnight success. At a certain point I just have to face the fact that two dudes rambling about technology into microphones is just not that unique or fascinating in 2025. There are a ton of other podcasts out there doing something similar, and we simply haven’t won the attention economy lottery. Maybe we should try being fascist douchebags and kissing the Trump Administration’s ass? That seems to be working for a lot of bro-pods.

Then again, I also have to be able to live with myself, so maybe not.

While we’re on hiatus, I am trying to pitch Chaos Lever to a few publishing companies to see if they would want to adopt us like the adorable little puppies that we are. Realistically, this is the only way I see the show coming back in the near-term.

Training Classes

After Pluralsight, the biggest part of my revenue comes from delivering live training. That’s right! You could have me as an instructor if you want to learn about HashiCorp Vault or Terraform and sign up for the right class.

For the last several years, I have been teaching classes through River Point Technology, delivering both custom private trainings and public trainings through HashiCorp Academy. With the acquisition of HashiCorp by IBM, all HashiCorp Academy training is now delivered through IBM’s partners. That means I had to negotiate with a new entity to keep teaching those classes.

All public classes were put on hold during the transition, so I basically didn’t teach any Academy classes for three months. But I’m glad to say that things are back on track and I’ve taught two classes for this new company. They seem to be offering the public classes once a month, giving me the opportunity to apply to teach up to three classes every month if I wanted to. That’s a bit much, so I’ll be sticking to about one class a month or so.

I also had another training company reach out to me about teaching private Terraform classes. So far, I have taught two classes for them and there’s a distinct possibility that is going to ramp up a lot in 2026.

This year I finally decided to become a Microsoft Certified Trainer (MCT) as well. This makes is possible for me to teach Microsoft classes through training partners. I just got the MCT badge in December, and I plan to start looking into training partners this January.

Out of all the revenue areas for Ned in the Cloud, training is the only one that grew year over year. It makes sense for me to put more time and energy into the thing that is actually growing at the moment.

Books

I’m very excited to say that in 2025, I became an official O’Reilly author. I can’t even remember what the first O’Reilly book was that I read, but I can tell you they have been integral to my career development. Fifteen years ago, I would have never thought that I would be an O’Reilly author too, but last year that changed with a simple LinkedIn message.

One of their Acquisitions Editors reached out to me about a sponsored guide Microsoft was looking to do around running Linux on Azure. She wanted to know if I’d be interested in writing the guide. I said yes, on the condition that I could have Chris Hayner co-author it with me. Of the two of us, he’s more versed in Linux and I’m deeper into Azure. Between us, I thought we could churn out a pretty good book. O’Reilly gave the OK and we were off to the races.

I guess we did a decent job, because before the book wrapped, they asked if we wanted to write another guide about picking a Linux flavor. That’s about 75% done at this point, with Chris doing the lion’s share of the work while I tried to get my Pluralsight stuff done on time.

In 2026, they have another project already lined up for us, but we don’t have a ton of details yet. My hope is to keep growing this relationship and eventually publish a book with a cool animal on it.

Outside of the O’Reilly stuff, it appears that I need to update the certification guides for both Vault Associate (003) and Terraform Associate (004). I helped write and review the questions for the Terraform Associate (004) exam and I’ll be creating six courses on Pluralsight about it, so updating the guide shouldn’t be very hard.

Other Appearances

Did I do other things in 2025? Yeah probably. Here’s a short list of stuff I can recall:

• HashiTalks 2025
• Cloud Field Day 22
• HashiCorp Ambassador Thought Leader Series
• Platform and Cloud Cafe
• IaC Conf
• Iac Conf Live
• The Cube at KubeCon
• HashiCorp Ambassador Roundtable
• Edge Field Day Showcase

Okay, wow, so it was a lot of other things. Funny. You forget how busy you actually were while in the thick of things.

Deprecations and Failures

This is a fun section with a fun name. What did I fail at in 2025? And I don’t mean that in a super-negative way. One of my three pillars is embracing failure, so it’s time to give my failures a big ole hug!

YouTube Channel

I had planned in the beginning of 2025 to publish Terraform Tuesday content three times a month and also explore some other topics, like Home-labbing and WASM. Needless to say I utterly failed on all counts. If you’ve read this far, you know I sacrificed building the channel in lieu of creating Pluralsight content. No regrets on that front, but still counts as a failure.

Chaos Lever

2025 was supposed to be a year that Chaos Lever grew up and started making some money. While I feel like the quality of the show never faltered, we still weren’t able to monetize it. Considering how much time it took up in my week to write, edit, and produce; I couldn’t justify continuing with the show due to other priorities that actually make money. Still a failure though.

WebAssembly

Another year has passed and I am no further in my process to use or really understand WebAssembly. Maybe 2026? I wouldn’t bank on it.

Certification Guides

I meant to update the Vault Associate guide this past year, and never quite got around to it. Now that the new version of Terraform Associate is out, it’s time for me to update both. Why didn’t I get the Vault one done? Once again, Pluralsight strikes. I did create a ton of courses for the Vault Associate (003) exam, so I am well suited to update the guide.

Looking Forward

What’s coming for 2026? That’s what the Planning for 2026 post is for silly! Just as a preview though, I’m planning to finish about 10 Pluralsight courses, start delivering some Microsoft Training, and renew my focus on the YouTube channel.