// description
The principle that among competing explanations or solutions, the simplest one — the one with the fewest unnecessary assumptions — should be preferred. Often stated as "don't multiply entities beyond necessity." Simplicity is the default; complexity needs justification.
// history
Named for 14th-century English friar William of Ockham, who formalised parsimony as a guiding principle in philosophy and theology. The principle predates Ockham (Aristotle stated something similar) but his formulation became definitive. Occam's Razor is foundational in scientific method, where simpler hypotheses are preferred until there's reason to add complexity. In business and design, it underpins principles like "less is more" and the drive to eliminate unnecessary features and processes.
// example
Your email open rates dropped. You could theorise about algorithm changes, sender reputation issues, competing newsletters, time-of-day psychology, and deliverability. Or you could apply Occam's Razor: the simplest explanation is that your recent subject lines have been boring. Test that first before investigating the complex alternatives.
// katharyne's take
I invoke Occam's Razor constantly when I hear myself (or other creators) spiralling into complex explanations for simple problems. "My sales are down because the algorithm changed and competition increased and buyers are price-sensitive and..." — or maybe your thumbnail is bad. Start with the simplest plausible explanation. Fix that first. Then get complex if you need to. Most problems are not as complicated as we make them when we're anxious about them.
// creative uses
- When a KDP book stops selling, apply Occam's Razor before overhauling everything: check the cover first. A bad cover is the simplest explanation for low clicks. Update it before changing your keywords, your subtitle, your description, or your price.
- Use it on your Midjourney workflow when images consistently miss what you're aiming for: the simplest explanation is usually the prompt itself — you've described too much or described the wrong thing. Simplify the prompt before troubleshooting model versions or parameters.
- Apply it when your course has low completion rates: the simplest explanation is usually that Module 1 is too long or too slow. Cut the first module in half before designing a new onboarding sequence, gamification, or accountability system.
// quick actions
- Identify a problem in your business you've been overthinking. Write the three simplest possible explanations for it — no jargon, no multi-variable theories. Address the most likely one this week before building a complex solution.
- Review your current funnel or launch sequence. Count how many steps a buyer has to take from discovery to purchase. If it's more than three, apply Occam's Razor: can you remove any step that isn't directly necessary?
- The next time you build a Midjourney prompt that isn't working, remove half the descriptors and run it again before adding more. Fewer elements almost always outperform more when the core subject is unclear.
// prompt ideas
I've been overthinking this problem and I need you to apply Occam's Razor to it. The problem is: [describe the issue — e.g. "my Etsy conversion rate dropped this month" or "my course has low completion rates"]. Give me the three simplest possible explanations in ranked order of likelihood, tell me the single most plausible one based on what I've described, and give me one concrete action I can take this week to test it before trying anything more complex.
My current [process / product / system] feels overcomplicated. Here's how it works: [describe it]. Apply Occam's Razor: identify every step or element that isn't strictly necessary for the core function. Then give me a stripped-down version that achieves the same outcome with fewer moving parts. I'd rather start simple and add complexity only when needed.
Help me simplify my Midjourney prompt. Here's what I'm currently using: [paste your prompt]. Strip it down to the minimum elements needed to get the core image I want — subject, style, and one mood descriptor. Then give me three variations, each adding just one additional element, so I can test which addition actually improves the output rather than adding everything at once.