Precision refers to how consistently you get the same result when you repeat a measurement or process. If you step on a scale five times and it reads 154.2 pounds every time, that scale is highly precise, even if your actual weight is 156 pounds. Precision is about repeatability, not about being correct. That distinction between precision and the related concept of accuracy is one of the most important ideas in science, medicine, and technology.
Precision vs. Accuracy
People use “precise” and “accurate” interchangeably in everyday conversation, but in science and measurement they mean different things. Accuracy describes how close a measurement is to the true value. Precision describes how close repeated measurements are to each other. You can have one without the other.
A classic way to visualize this: imagine throwing darts. If your darts cluster tightly together in the lower left corner of the board, you’re precise but not accurate. If they scatter loosely around the bullseye, you’re accurate on average but not precise. The ideal is both: a tight cluster right on the target. Importantly, precision doesn’t require you to know the correct or true value at all. It only tells you whether your instrument or method gives consistent results.
This matters in practical terms. A bathroom scale that always reads two pounds low is precise (the readings are repeatable) and can still be useful for tracking weight changes over time. But it’s not accurate. A scale that bounces between wildly different numbers each time you step on it lacks precision, which makes it nearly useless regardless of whether one of those readings happens to be correct.
Repeatability and Reproducibility
Scientists break precision down further into two subtypes. Repeatability is the tighter standard: it measures whether you get the same result when the same person uses the same equipment under the same conditions in a short period of time. Think of a lab technician running five identical tests back-to-back on the same sample with the same instruments.
Reproducibility is broader. It asks whether different people, in different labs, using different batches of materials, can still get consistent results with the same method. Reproducibility is harder to achieve because it introduces all the real-world variation that repeatability deliberately holds constant: different operators, different equipment, different days. When a measurement method has good reproducibility, scientists can trust that results aren’t just artifacts of one particular lab setup. The gap between these two types of precision is itself informative. If a method is highly repeatable but poorly reproducible, something about the process is sensitive to conditions that vary between labs, which tells researchers where to look for problems.
Precision in Medicine
The word “precision” has taken on a second, high-profile meaning in healthcare. Precision medicine is an approach to treatment and prevention that tailors decisions to an individual’s genes, environment, and lifestyle rather than treating everyone with the same condition identically. The NIH’s Precision Medicine Initiative defines it as accounting for “individual variability in genes, environment, and lifestyle for each person.”
One of the clearest applications is in how your body handles medications. Your DNA determines what types of drug receptors you have and how many, which can mean you need a higher or lower dose than most people. Your genes also affect how quickly your body breaks down a drug. If you metabolize a medication faster than average, it leaves your system sooner and may not work well at a standard dose. If you metabolize it slowly, the drug can build up and cause side effects. Genetic testing before prescribing certain medications can help match the right drug and dose to the right person.
In cancer treatment, precision approaches have produced some of the most dramatic results. Targeted therapies designed for patients with specific genetic mutations in their tumors have improved survival in several cancer types. In advanced melanoma, immunotherapy drugs that work by releasing the brakes on the immune system have substantially improved survival rates. For men with metastatic prostate cancer who carry certain inherited gene mutations, a class of drugs that exploits a weakness in cancer cells’ ability to repair their own DNA has improved both progression-free and overall survival. These treatments work because they’re matched to the biology of a specific patient’s disease rather than applied broadly.
The Diversity Problem in Precision Medicine
Precision medicine is only as good as the data behind it, and that data has a major gap. More than 90% of participants in large genomics studies have been of European genetic ancestry. This has led to a narrow understanding of disease biology and slowed the development of treatments that work for all populations.
The NIH’s All of Us Research Program is working to close this gap. Researchers analyzing data from nearly 250,000 participants discovered more than 275 million previously unreported genetic variants. Half of the genomic data came from participants of non-European genetic ancestry. The findings revealed significant variability in disease-risk variants between different ancestry groups, meaning that risk prediction tools built on less diverse datasets could produce misleading results for large portions of the population. Without diverse data, genetic risk scores developed for one group simply don’t translate reliably to others.
Precision in AI and Diagnostics
Artificial intelligence has pushed diagnostic precision forward in medical imaging. In one study, deep learning algorithms distinguished between low-grade and high-grade brain tumors on MRI scans with 93.2% accuracy as measured by area under the curve, a standard metric for diagnostic performance. AI software analyzing over 22,000 mammograms detected breast cancer with an 89.6% score on the same metric. For detecting a specific type of lung collapse on chest X-rays, an AI system outperformed radiologists.
These tools don’t replace human judgment, but they add a layer of consistent, repeatable analysis. That consistency is itself a form of precision: unlike a human reader whose performance can vary with fatigue, workload, or experience level, an algorithm applies the same criteria every time. The combination of human expertise and algorithmic precision is where diagnostic imaging is heading, with each compensating for the other’s weaknesses.
Why Precision Matters in Everyday Decisions
Understanding precision helps you evaluate information you encounter regularly. When a fitness tracker says you burned 347 calories during a workout, the number looks precise, but that doesn’t mean it’s accurate. If it consistently reports similar calorie counts for similar workouts, it’s precise and useful for comparing your own sessions over time, even if the absolute number is off. The same logic applies to home blood pressure monitors, kitchen scales, and weather forecasts. A tool that gives you consistent, repeatable readings lets you detect meaningful changes, which is often more useful than a single perfectly accurate measurement you can’t replicate.
Precision also matters when you read statistics in the news. A poll with a wide margin of error lacks precision. A study with a small sample size may produce results that would scatter widely if repeated. Recognizing that precision is about consistency, not correctness, gives you a practical framework for judging how much to trust any number you encounter.