Writing on Craig McCaskill

Introducing Posthorn: the missing layer in self-hosted email

Tue, 26 May 2026 00:00:00 +0000

If you self-host on a cloud VPS, you’ve probably noticed by now that some of your services can’t send email. DigitalOcean, AWS Lightsail, Linode, and Vultr all block outbound SMTP on ports 25, 465, and 587. It’s an anti-abuse measure: compromised VMs make great spam relays, and one bad actor can burn the reputation of an entire IP range. The official recommendation is to use a transactional HTTP API like Postmark. Or, if all you care about is a contact form, a SaaS service like Formspree. Both work for the apps that can speak HTTP. Some can’t.

Toothpaste Back in Tubes: On opportunity cost, art, and what stays human

Sun, 19 Apr 2026 20:51:03 +0000

Two environmental artists. Three months. Five hand-built buildings for an indie game’s city district.

Every facade custom. Every texture intentional. Three months of craft for five structures players will sprint past in eight seconds on their way to the next quest marker.

Or: a generation pipeline produces fifty convincing models in an hour. Variation, weathering, architectural detail. Good enough that players don’t notice. And now those two artists have three months for the work that actually makes a game memorable.

The Games Your Organization is Actually Playing

Sun, 08 Mar 2026 12:00:00 +0000

Or: before you blame the players, check the rules.

I once watched a team of smart, well-intentioned people spend six months optimizing a metric that didn’t matter. Not because they were confused about what actually mattered. Because the metric was what got discussed in leadership reviews. Everyone on the team could have told you what would move the business. They also knew it wouldn’t show up on the dashboard their boss checked every Monday.

The Case for Being Wrong Faster

Sun, 22 Feb 2026 12:00:00 +0000

Or: intellectual honesty is your competitive advantage.

In 2011, Netflix made a decision that, at the time, looked catastrophic. They split their DVD-by-mail service from streaming, raised prices by 60%, and watched 800,000 subscribers cancel in three months. The stock cratered from $300 to $77 in a single day. Saturday Night Live mocked them. Investors called for Reed Hastings to resign.

Blockbuster’s CEO at the time probably felt vindicated. They’d turned down the chance to buy Netflix for $50 million years earlier, confident in their retail model. As Netflix stumbled, Blockbuster doubled down on late fees and store expansions.

When Seeing Is No Longer Believing

Sun, 25 Jan 2026 12:00:00 +0000

Imagine you watch a video. Multiple angles, timestamped, from bystanders with phones. You see what happens. A confrontation. A shooting. The sequence of events is clear.

Then you read the official statement, and it describes something else entirely. Not a different interpretation of ambiguous footage. A different set of facts. The victim attacked officers, the statement says. The officer was hospitalized with serious injuries. Except you just watched that officer walk away unharmed, holster his weapon, and disappear around a corner.

The Bubble That Knows It's a Bubble

Sun, 24 Aug 2025 12:00:00 +0000

The four words that precede every crash: “This time is different.”

Except this time, the person warning about a bubble is Sam Altman, the CEO most responsible for creating it. When OpenAI’s chief executive warned last week that investors were “overexcited” about AI, markets reacted immediately. Nvidia fell 3.5%, Palantir dropped nearly 10%, and the selloff spread globally.

The warning came amid a cascade of seemingly supporting data. That same week, MIT researchers published findings that 95% of companies investing in generative AI were seeing no measurable returns. Apollo Global Management’s chief economist warned that current valuations exceeded even dot-com bubble peaks. And Federal Reserve data showed AI investment consuming more than half of America’s total capital expenditure.

Sensing and Sensemaking: The Hidden Skills of Product Management

Wed, 11 Jun 2025 12:00:00 +0000

Product management is built on a contradiction. We’re supposed to be data-driven, yet our most important decisions happen in the absence of clear data. We’re meant to be customer-focused, yet customers can’t tell us what they’ll want tomorrow. Navigating this requires two underappreciated skills: sensing and sensemaking.

These aren’t the skills we talk about at product conferences or test for in interviews. You won’t find them in job descriptions or performance reviews. They’re rarely taught in product courses or covered in certification exams. Most PMs develop them accidentally, if at all.

The Innovator's Dilemma: Why Good Companies Fail at the Very Thing They Do Best

Sun, 01 Jun 2025 12:00:00 +0000

Your product is crushing it. Customer satisfaction scores are through the roof. Your biggest clients are renewing multi-year contracts. The board is thrilled with your margins. And yet, somewhere in a garage or a dorm room, someone is building the thing that will make your product irrelevant. The scariest part? Doing everything “right” according to conventional business wisdom is exactly what will cause you to miss it.

Market leaders are perfectly positioned to fail. Not because they’re lazy or stupid, but because they’re doing exactly what they’re supposed to do. They listen to their best customers. They chase high-margin opportunities. They optimize what works. And then some startup launches a “toy” that doesn’t even have half their features, and eighteen months later, that toy owns 30% of the market.

How to Think About AI

Thu, 29 May 2025 12:00:00 +0000

New tools don’t just change what we can do, they change who wins. The printing press didn’t eliminate scribes overnight, but it made literate societies dramatically more powerful than illiterate ones.

Every technological shift creates a predictable pattern and early adopters gain disproportionate advantage over those who wait. What makes AI different isn’t the technology itself, but the speed at which this advantage compounds. Jensen Huang captures this dynamic perfectly: “You will not lose your job to AI, but will lose it to someone who uses it.” The CEO of NVIDIA isn’t making predictions, he’s describing a competitive reality that’s already reshaping how work gets done.

Old Code, New Voice: How Technical Debt Led to Better Storytelling

Sun, 25 May 2025 12:00:00 +0000

I’ve been writing again. After nearly four years of letting my website collect digital dust—publishing only a few posts while my day job consumed all my creative energy. I finally felt the itch to both write and fix the stagnant site that housed my thoughts.

The site functioned—posts appeared, links worked, the basics held together. But navigating it was like searching through a poorly organized filing cabinet. I had scattered tags on posts but had no way to browse by topic. No tag pages, no meaningful organization, no help for anyone trying to find related content. For someone who spends their days thinking about user experience, this was painfully ironic. You know the old saying about the cobbler’s children having no shoes? I was living it. Here I was, making shoes all day for everyone else, while my own website felt like it was designed by someone who’d never heard of information architecture.

Writing is still thinking

Sat, 24 May 2025 12:00:00 +0000

Something humbling happened last week: three hours of my life disappeared into a blank document with barely a paragraph to show for it.

I sat down to write a product strategy document – something I thought I’d mastered after years of practice. In meetings, I can usually help teams navigate the tough questions: what to build next, how features connect to business goals, which metrics matter most. I’ve built a career partly on my ability to clarify complex concepts and find structure in ambiguity.

The Science of Saying No - Frameworks and Execution

Mon, 24 Feb 2025 12:00:00 +0000

Every company claims to be disciplined about prioritization; most roadmaps look like feature buffets with everything marked “high priority.” The space between strategic intention and daily execution is littered with good ideas that never shipped.

The irony wasn’t lost on me: I was writing about prioritization while completely failing to prioritize my own content. What started as a focused piece about saying no had metastasized into an everything-and-the-kitchen-sink manifesto covering organizational psychology, decision frameworks, political dynamics, and implementation tactics. Classic scope creep, just with words instead of features.

The Art of Saying No (Or At Least, Not Yet)

Mon, 17 Feb 2025 12:00:00 +0000

The math of product is simple but brutal: too many demands, not enough resources. Your backlog isn’t just full—it’s a graveyard of good intentions. When every feature is called “critical,” nothing actually gets built well. This isn’t about managing tasks better. It’s about survival. Because when everything matters, nothing really does.

The True Cost of Poor Prioritization

Poor prioritization costs more than just time and money—it destroys trust. Teams that jump between half-finished projects develop organizational ADHD. Engineers grow cynical watching yesterday’s “must-have” feature become tomorrow’s forgotten project. Executives start to doubt the entire product team.

Escaping the Build Trap: How Effective Product Management Creates Real Value

Fri, 13 Dec 2024 12:00:00 +0000

You ship feature after feature, your development team works overtime, your backlog is always full—yet customer satisfaction remains flat and business metrics barely budge. Sound familiar? You’re not alone. Most product organizations have fallen into what Melissa Perri calls “the build trap”: the dangerous cycle of equating more features with more success while losing sight of actual customer value.

Most product teams know the pattern: quarterly planning becomes a feature auction where everyone bids for their priorities. Leadership pushes for velocity metrics, sales champions their deal-closing features, and engineering just wants a stable roadmap. The result? Endless debates about what to build, while the fundamental question—whether we should build it at all—goes unasked.

Playing Favorites - How AI Shapes Our Knowledge Without Us Noticing

Sat, 25 May 2024 12:00:00 +0000

AI is making us all smarter in exactly the same way, and that might be the problem. What happens when the most accessible knowledge becomes the most uniform? As AI increasingly curates what information we see, are we trading the beautiful messiness of diverse human thought for algorithmic convenience?

Andrew Peterson’s April 2024 research paper “AI and the Problem of Knowledge Collapse”1 reveals how our AI tools are quietly filtering out the diverse perspectives that drive innovation, creating a future where we all know more but think more alike. Here’s why that matters for all of us.

Beyond Metrics - Why OKRs often miss the human element of work

Tue, 02 Jan 2024 12:00:00 +0000

In the quest for clarity and alignment, modern organizations increasingly embrace structured goal-setting frameworks like OKRs (Objectives and Key Results). But beneath the surface of these seemingly rational systems lies a deeper question: Are we measuring what truly matters, or simply what’s easy to measure?

The Metrics Paradox

Here’s the hard truth about metrics: we’ve become obsessed with counting what we can count, not measuring what actually matters.

Look at the quarterly OKR ritual playing out across corporate America. Executives launch the process with fanfare. Teams huddle in conference rooms crafting objectives. Spreadsheets bloom with key results. Dashboards multiply. Everyone nods along.

The Great Enshittification - How Product Evolution Becomes Product Regression

Wed, 23 Aug 2023 12:00:00 +0000

In 2012, Netflix was a meticulously organized catalog that helped you discover films and shows across genres, decades, and continents. In 2015, Spotify was a music lover’s sanctuary that put your personal library and thoughtful exploration front and center. Fast forward to today, and these platforms—along with YouTube, LinkedIn, and countless others—have morphed into something unrecognizable to their early adopters.

The culprit? A relentless pursuit of “innovation” that prioritizes engagement metrics over user intent, growth at all costs over genuine utility, and the marginal user over the dedicated enthusiast.

Truth in a Post-Truth World, Revisiting Frankfurt's On Bullshit

Sun, 30 Jul 2023 12:00:00 +0000

Last week, philosopher Harry G. Frankfurt passed away. As a young adult, his essay “On Bullshit” completely changed how I thought about truth and lies, especially given a personal history with family who had a tenuous relationship with the truth. When I heard the news of his death, I felt pulled to read it again, wondering if it would hit differently after a couple decades of life experience.

Harry G. Frankfurt, May 29, 1929 - July 16, 2023

What makes them click?

Tue, 24 Jun 2014 12:00:00 +0000

Here are the slides for the lightning talk I gave at Conversion Conf 2014, aimed as an introduction to psychology in design, why it matters and what drives us to make decisions.

The goal of this talk was to give people without a design background a quick intro into psychology and how large an impact it can have on your products.

The Elements of User Experience

Sun, 20 Apr 2014 12:00:00 +0000

For anyone looking to get into User Experience, I can’t recommend enough that you read Jesse James Garret’s book The Elements of User Experience. It provides a great framework for thinking about design and gives you a structured rationale and toolkit to deal with just about anything. This is the book you read to find out just how little you actually know about UX, and what those gaps are. Jesse takes the idea that the customer is always right and shows you concisely how to find out how to fulfil your users wants, needs and goals – even if they don’t know them yet.

Mental Models

Tue, 06 Aug 2013 12:00:00 +0000

Imagine you have never used an iPad before, but I tell you that you can play chess on it. Before you turn it on, you will already have started building a mental model of how the chess game will look, how you will interact with it, and what you will be able to do. This is a different way of thinking about chess from someone who has already used an iPad, from someone who is a chess master playing on a physical board, and from the designer of the chess game itself.