These are not ethical beliefs. They are survival mechanisms built across 15 years of working inside large organisations, early-stage startups, and my own security and manpower business, where I learned what it costs to run low-margin, high-friction operations.
Three tiers. Tier 1 principles are non-negotiable: breaking them ends the business. Tier 2 are proven heuristics: follow them unless you have a specific reason not to. Tier 3 is still developing: provisional, subject to revision.
Tier 1: Non-Negotiable
Survivability over elegance. If the business survives, problems can be fixed. If cash runs out, the game ends. I do not accept contracts where cash flow is negative, payment terms exceed 45 days, or there is no written SLA. No exceptions for “strategic relationships.”
Revenue quality over revenue size. Not all money is equal. Large contracts with bad terms are liabilities dressed as growth. The filter: if new revenue divided by new complexity is less than current revenue divided by current complexity, reject it. Complexity means margin, payment days, and custom requirements combined.
Systems fail before people do. When operations break, blaming individuals avoids the real problem. Before blaming anyone, I ask: can I reproduce this failure by following the current process exactly? If yes, fix the system first. The failure hierarchy runs: bad system, bad incentive, bad timing, bad person. Work through it in that order.
Luck is a variable. Preparation is the constant. Execution buys a lottery ticket. Luck picks the winner. My job is to survive long enough for luck to find me. Before any experiment, I confirm two things: the business survives if this fails completely, and enough runway exists for 10 to 15 attempts. If either answer is no, I reduce the bet or defer.
Independence is the only asset I will not trade. I define enough by my ability to say no, not by the number of zeros in my account. I keep a six-month cash reserve. I keep margins high enough to lose any single client without panic. I track the time cost of revenue in Rs per hour of founder attention. I do not accept debt that removes my ability to walk away. I do not optimise for revenue if it decreases decision freedom.
Tier 2: Follow Unless You Have a Reason Not To
Consistency compounds. Genius burns. I cannot sustain peak performance every day. A system that requires it is fragile. I design for average effort repeated daily, not occasional brilliance.
Fast growth is often fast cancer. Systems need time to harden. When a client scales five times in under 90 days, I audit: is this growth profit-driven or burn-driven? Can their business survive if funding stops? Do I have systems for this scale? If I am unsure on any of these, I cap exposure.
Tools reduce error. They do not replace judgment. I use AI and software to remove single points of failure, not because they are interesting. The test: if this tool disappeared tomorrow, would the business break? If yes, the dependency is too tight. Build redundancy or own the risk explicitly.
Execution over ideas. A business plan is a guess. The terrain changes. The map must change with it. I do not defend past decisions. I update when new information arrives. Attachment to old plans is ego pretending to be strategy.
AI as cognitive infrastructure, not cognitive replacement. AI is an editor, not an author. It tightens drafts, surfaces gaps in arguments, and organises messy notes. It does not decide what to build, does not choose between options, and does not carry accountability. Before using AI for anything, I ask: do I already know what outcome I want? If I am using it to avoid thinking rather than to accelerate thinking, I stop and think first.
Tier 3: Still Developing
Every framework is a snapshot. These principles are empirical, not eternal. Contexts change. When reality disagrees with a principle, the principle gets revised. This page will contradict itself over time. That is not failure.
Principles have expiration dates. I do not assume these will hold a year from now. AI’s rise has already made several constraints I once treated as fixed look provisional. This page exists so I notice when my thinking becomes obsolete before the business does.