1. Precision Creates the Illusion of Accuracy The Washington Monument example is perfect: stating “305 feet 1 inch” sounds authoritative and credible because of its precision, but it’s completely wrong (that’s the Statue of Liberty’s height). The document warns: “don’t assume something is accurate simply because it is expressed with great precision.” The insight is that precision is easy to fake and hard to verify. Anyone can add decimal places to a made-up number. When someone says “we’ll increase revenue by 23.7%,” the precision suggests careful analysis, but it might just be a guess with extra digits. Precision without accuracy is worse than vagueness because it misleads people into false confidence. This is why financial projections five years out that specify exact dollar amounts are often nonsense disguised as analysis.

 

2. Accuracy Must Come Before Precision The document states clearly: “The focus should be accuracy first, precision after.” The insight is that most people do this backward. They spend time refining estimates to three decimal places before confirming they’re measuring the right thing. Getting the precise wrong answer is useless. This explains why projects with detailed Gantt charts often fail: teams precisely schedule tasks without accurately understanding what needs to be done. It’s like the precision of hitting a target exactly while aiming at the wrong target. Better to roughly estimate the right thing than precisely estimate the wrong thing. Accuracy defines whether you’re in the right ballpark; precision specifies where in that ballpark.

​

3. Excessive Precision Misrepresents Confidence in Your Data The document warns: “don’t misrepresent your level of confidence...by providing more precision than is justified.” The Washington Monument’s actual height is given as “554 feet 7 11/32 inches,” showing precision appropriate for a permanent structure with precise measurement. The insight is that precision should match confidence. If you measured something with a 10% margin of error, expressing it to 0.01% precision misrepresents certainty. When someone forecasts “127 customers next quarter,” they’re implying more confidence than “about 125 customers” even though his forecast method probably can’t distinguish between them. Honest communication matches precision to actual certainty. Mismatched precision is either incompetent (doesn’t understand measurement error) or manipulative (wants to appear more certain).

 

4. The Cost of Precision Must Justify Its Value The document’s final point asks: if something is less precise than desired, “what it will cost to gain greater precision.” The insight is that precision has diminishing returns. Going from ±10% to ±5% accuracy might cost $1,000. Going from ±5% to ±1% might cost $10,000. Going from ±1% to ±0.1% might cost $100,000. At some point, the cost of additional precision exceeds its value. Many projects waste resources pursuing precision that doesn’t matter for decision-making. If your choice between options A and B would be the same whether the cost difference is $98,000 or $102,000, spending money to narrow that range from ±10% to ±2% is wasteful. The decision only needs enough precision to confidently choose; anything beyond that is academic.

​

·       When I suspect someone is using precision to fake accuracy, how do I challenge them without seeming cynical? The warning about precision creating false confidence is useful, but if someone presents precise figures, questioning their accuracy can seem as if I’m accusing them of dishonesty. Should I ask about the methodology, request confidence intervals, or just privately discount precision?

​

·       If I’m asked to provide an estimate and don’t have enough information to be accurate, should I refuse or give a rough estimate with caveats? The emphasis on accuracy first suggests I shouldn’t guess, but sometimes people need some number to make decisions. Should I provide my best rough estimate clearly marked as uncertain, or refuse until I can be more accurate? What if refusing makes me seem unhelpful or indecisive?

​

·       How do I determine the appropriate level of precision for my context without either undershooting or overshooting? The document says precision should be useful, “not so crude as to prevent proper planning” but “not so fine as to be unnecessarily exact.” But how do I know what’s appropriate? Should I ask what level of precision people need for decision-making, or is there a rule of thumb for matching precision to purpose?