How the Company-Startup Thing Worked Out For Me in 2024

Every year, I post a retrospective about what it’s been like to start up a company. If you want to catch up, check out past posts in the life quest tag. Starting this year, the posts cover calendar years, so this one covers how 2024 went for me. When we left off the Year 12 (2023) post, I had just barely launched SmartPostgres.com with the goal of becoming “the Brent Ozar of Postgres.” I’d been really burned out on Microsoft data products, and I wanted to find an off-ramp to another technology.

The short story for 2024 is that broadening my technology & geographic horizons went great, but it made me even more thankful for Microsoft SQL Server and Las Vegas. The long story, well, deep breath – let’s get introspective and do some navel-gazing.

What made BrentOzar.com successful?

As a SQL Server-focused business, BrentOzar.com has a bunch of parts that evolved organically over the years:

• Consulting: lead generation, contracts, methodologies, slide decks to assist the process
• Training classes: pre-conference workshops, in-person classes, live online classes, and private corporate events
• Training videos: long (day+) training recordings, 10-60 minute recordings, 1-2 minute recordings
• Training in written form: blog posts, specialized mailers (like a 6-month DBA training newsletter), social media
• Subscriber base: about 100K email subscribers, about 40K YouTube subscribers
• Scripts, aka the First Responder Kit
• Apps: SQL ConstantCare, the Consultant Toolkit

All that stuff didn’t all materialize at once, obviously. We built the company up for over a decade. Initially, it was me, a few founding partners (shout out to Jeremiah, Kendra, and Tim), and we added employees and vendors/contractors over time. The BrentOzar.com you see today is the product of a lot of blood, sweat, and tears from a LOT of people, and there’s absolutely no way I could have built anything even remotely resembling it by myself. I’m thankful for their work, although of course they got paid for that, and I still try to credit them constantly. (I would never say I wrote sp_BlitzIndex or sp_BlitzCache – I simply wouldn’t have the talent for that.)

However, “a lot of people” also means “a lot of time and money” – and if I was gonna build a new company from scratch, I didn’t want to reuse the same plan, because it was expensive in a lot of ways. When I started up a new company aiming to achieve a similar level of success, I couldn’t reuse the same plan. I had to reconsider what to build to make SmartPostgres.com as sustainable as possible, quickly. To do that, I thought about:

1. What income-generating parts brought the most profit?
2. What could bring in the most customers in the door for #1?

I decided to focus on training revenue.

I came to the conclusion that for revenue, I should focus on building training classes. Done well, it’s purely passive income: you keep building more and more courses, and eventually you hit the point where the income is coming in even when you’re not working.

Software and consultants never hit that point unless you grow the business so large that you can afford to hire enough people to assign all the work out, and even then, you spend time managing those folks and their service delivery. I’m trying to work less, not more, so I ruled those options out. (Just to be clear, I do think there are a lot of interesting online SaaS angles in the Postgres market, and I’ll probably pursue some at some point, but to get the business profitable and sustainable quickly, I had to ignore those in 2024 for sure.)

So with that decision out of the way, how could I build a sustainable long term pipeline of customers for training material? It was time to make that decision.

And to bring in customers for that,
I decided to focus on a newsletter.

For inbound marketing, I focused on building the kind of newsletter I would wanna read myself. I like newsletters because people will trade their email address for it, and you own that list of email addresses. If you have a mailing list, it’s really easy to sell content to your captive readers.

Starting December 2023, I built the SmartPostgres.com blog & newsletter with two kinds of content:

• Short weekly recaps like this of the best 3-4 resources I found each week, plus one query tutorial. I thought the combination of the super-limited curation, PLUS the query tutorial, would be the magic recipe to differentiate me from other Postgres curations in this space like Postgres Weekly.
• Query exercises with a challenge one week, followed by answers the next week

Part of my reasoning with picking those kinds of content was that AI wasn’t a serious competitor for either one. Even if LLMs completely invaded content creation, they still wouldn’t do a great job on those two kinds of content.

I also wanted to build something very different than what I saw in traditional Postgres blogs. There’s a lot of Web 1.0 content out there, if you know what I mean: ancient web sites, mailing lists, non-interactive stuff. I wanted to attract a different kind of reader.

Aaaaaand… I gave up on the newsletter.

By summer, I decided my magic recipe for newsletter contents was wrong. It was simply too resource-intensive to come up with a useful, concise SQL tutorial each week, plus a query challenge and answer. Some of the tutorials and challenges took me an entire day to write.

I couldn’t justify that much time – that’s 30-50 days per year dedicated to growing the newsletter. Just for rough comparison, that represents $105K-$250K worth of Microsoft SQL Server consulting, depending on each day’s billable rate (like a single long term client vs individual emergency gigs.) And that doesn’t even include the time required to build courses! Given the choice between the two, I’d be much better off in the short term simply giving up on SmartPostgres, and focusing on SQL Server instead.

I could have skipped the code examples and just curated links, but… why bother when Postgres Weekly already exists? I do wish they’d do a little curation because some of the stuff feels like obvious corporate spam, but that’s my only complaint. I would rather just point people to Postgres Weekly, stick a fork in my own newsletter, and call it done. Time to move on to an alternative strategy.

(For SQL Server, I do even now still spend 1-2 days per week on inbound marketing – building blog posts, curating the newsletter content, recording Office Hours episodes, etc., but that pays off in >$1M/year in training sales, so it’s way easier to justify.)

Instead of building inbound marketing for Postgres, I tried something radical.

I tried pre-selling Postgres content I hadn’t built yet.

In August, I went to my social media with my early access sales pitch: pay $195 once, and for life, get access to any Postgres performance fundamentals classes I build, plus the ability to watch me stream live as I build those classes out, and watch the live stream recordings. I really had no idea if it would sell.

I sold $161K in the first 30 days.

Well, that changed everything right there! Instead of spending 30-50 days building inbound marketing for products I didn’t have yet, I could take that $161K and spend 30-50 days building out training material – which would then be easier to sell. It worked: I built Fundamentals of Select, Fundamentals of Vacuum, Fundamentals of Index Tuning, and commissioned Fundamentals of Python from Drew Furgiuele. I really hustled in that 2024 period.

Now, to be clear, I don’t want you to think SmartPostgres.com raked in a ton of cash after those first 30 days! As soon as the sales started rolling in, I went, “OMG, I priced that way too low.” I think I left a lot of money on the table. As soon as the promised early access dates finished, I shut off enrollment, then raised prices later. In the subsequent 10 months, I sold less than $25K because I was really focused on building the training material first. I wanted to be able to justify a much higher price tag after the material was built.

While I focused on SmartPostgres, I had to turn down the rest of my activities a little. After all, I can’t say “yes” to everything.

I did have to say no to some things.

I stopped doing proactive SQL Server First Responder Kit script additions just for the sake of adding stuff. Thankfully, Microsoft didn’t ship a new SQL Server version, so the scripts were more than good enough to get by through 2024.

I massively cut back on new SQL Server consulting clients, instead farming them out to a network of trusted friends that I’d worked with on previous projects. I took a small referral fee to send them the work, but I hardly took any new clients on myself – only the companies who insisted on working with me personally.

I also decided to be much more strategic about choosing conferences in 2024. I really, really love attending conferences, especially to give back to small local conferences, but I just had to make tough decisions to get Postgres content out the door. In a way, this sucked, because the in-person event circuit had just started to build back up enough that I could regroup with amazing friends (exhibit A: the above photo) – but I had to make tough choices.

That kinda sucked, because the post-COVID conference market was starting to ramp back up, and I was able to attend and speak at a few really amazing events.

My heart is really full with happiness when I look at the above photo, and that’s just one of the many amazing memories I made in 2024 as conferences started to come back to life.

Plus, a friend of mine from the Microsoft Certified Master program had branched out to start a completely new conference on how to do presentations. Boris Hristov had invited me to give a talk at the Present to Succeed conference, which was really flattering and I had to leap at the chance.

That conference was humbling! There were so many great sessions from really talented presenters, and it really inspired me to up my game with my own presentations. I’ve been proud of a lot of my sessions over the years, but this conference reminded me that my work is never done, and I gotta keep upping my game.

Even though I cut back a lot on work travel, I still managed to hit a bunch of places. The highlights of 2024 included Telluride Colorado, Hong Kong, Shanghai, Orange County CA, Nashville, Vancouver, Bulgaria, Cabo San Lucas, an Alaska cruise, San Diego, Boston Massachusetts, Anaheim California, Seattle, West Hollywood, and I finished up the year with a month in China and Japan before coming home to spend New Year’s Eve in Vegas with friends. My social media profiles say I love teaching, travel, and laughing, and the travel part is true for sure.

The more I roamed, the more I appreciated home.

I can type that now in retrospect, but I definitely didn’t recognize it in 2024. The more I played with different technologies and visited different countries, the more I appreciated the tech and places I’d built up the most familiarity with.

I continued traveling a lot in 2025, and working with different technologies, but my appreciation for the familiar grew in 2025 thanks to the ground work and experimentation I put in during 2024. Today as I type this in December 2025, I think the evidence is pretty clear: I’m churning out 3 blog posts and one long-form video per week on my SQL Server platforms through 2025, and I’m hardly publishing anything on my Postgres platforms in comparison. In 2026, my goal is to find a natural balance between those two things, because I really do love both of ’em tremendously. They’re both awesome.

I’m 52 years old, and I’m still cursed/blessed with a desire to know everything about every technology. Man, if I could wave a magic wand and suddenly know one thing, I wish I could have instant mental familiarity with the entirety of the Postgres source code. I want to know everything about how that one database works (which is easier than SQL Server, since it’s open source), and then I’m sure I would want to know everything about how every database works, and then how every caching layer works, and then how every cloud vendor’s implementation of every database and every caching layer works, and then how every ORM interfaces with those technologies, and….

Yeah, it just never ends.

But at the same time, I wanna do a decent job of work/community/family/personal life balance, and there’s a lot I wanna achieve across all of those. I’m not the kind of person who wants to leave a legacy or have buildings with my name on them, but I wanna enjoy all of the time that I can. That goal keeps me from even going close to source code.

So yeah, in 2024, I feel like I did pretty well trying to balance roaming away from home, plus keeping the lights on at home. If anything, I erred on the side of roaming. I look forward to the next blog post in this series when I look back at 2025, because for now, I feel like I erred to much in goofing off, and 2026 will involve me getting back to work. Let’s see how things look in the rear view mirror a year from now!