What Is a Percentile?
A percentile is a value below which a given percentage of observations in a dataset falls. If a customer's satisfaction score is at the 85th percentile, it means 85% of all respondents scored lower and 15% scored higher. Percentiles divide a distribution into 100 equal parts, providing a precise way to locate any individual value within the broader dataset. The 50th percentile is the median, the 25th percentile is the first quartile (Q1), and the 75th percentile is the third quartile (Q3). Percentiles are especially useful when raw scores alone don't tell you much, knowing that someone scored 78 on a test means nothing without understanding where 78 falls relative to everyone else.
Why Percentiles Matter
Percentiles give context to raw numbers. A Net Promoter Score of 32 might sound mediocre until you learn it's at the 75th percentile for your industry, meaning three-quarters of competitors score lower. In survey research, percentiles help you identify top performers, set benchmarks, and communicate results to stakeholders who may not intuitively understand standard deviations but immediately grasp "top 10%" or "bottom quartile."
How Percentiles Work
Calculation Methods
There are several methods for calculating percentiles. The most common is the percentile rank method:
Step 1: Sort the data in ascending order. Step 2: Use the formula to find the position:
L = (P / 100) × (n + 1)
Where:
- L = the position in the sorted dataset
- P = the desired percentile
- n = the number of observations
Step 3: If L is a whole number, the percentile is the value at that position. If not, interpolate between the two adjacent values.
Worked Example
You have satisfaction scores from 20 respondents, sorted in ascending order:
12, 15, 18, 22, 25, 28, 30, 33, 35, 38, 40, 42, 45, 48, 50, 55, 60, 65, 70, 80
Find the 75th percentile:
L = (75/100) × (20 + 1) = 0.75 × 21 = 15.75
The 75th percentile falls between the 15th value (50) and the 16th value (55):
P75 = 50 + 0.75 × (55 - 50) = 50 + 3.75 = 53.75
This means 75% of respondents scored below 53.75.
Find the 30th percentile:
L = (30/100) × 21 = 6.3
P30 falls between the 6th value (28) and the 7th value (30):
P30 = 28 + 0.3 × (30 - 28) = 28 + 0.6 = 28.6
Percentile vs. Percentile Rank
These terms are related but answer different questions:
| Percentile | Percentile Rank | |
|---|---|---|
| Question answered | "What score corresponds to the Pth position?" | "What percentage of scores fall below this value?" |
| Input | A percentage (e.g., 75th) | A specific score (e.g., 53) |
| Output | A score value | A percentage |
| Example | "The 75th percentile is a score of 53.75" | "A score of 53 has a percentile rank of 73" |
The percentile rank formula is:
Percentile Rank = (Number of values below X / Total number of values) × 100
If 14 out of 20 respondents scored below 50, the percentile rank of 50 is (14/20) × 100 = 70th percentile.
Common Percentile Benchmarks
Several percentile cutpoints have standard names:
- P25 (Q1): First quartile, bottom 25%
- P50 (Q2/Median): Middle value, splits the distribution in half
- P75 (Q3): Third quartile, top 25% starts here
- P10 and P90: Often used for "bottom 10%" and "top 10%" benchmarking
- P99: Used in performance monitoring (e.g., "99th percentile response time")
Percentiles and Non-Normal Data
One of the biggest advantages of percentiles is that they work regardless of the data's distribution shape. They don't assume normality, they're not distorted by outliers (unlike the mean), and they describe any distribution accurately. For skewed data, like income, response times, or spending amounts, percentiles are often more informative than means and standard deviations.
Percentiles in Benchmarking
Companies commonly use percentile benchmarks to contextualize their performance:
- "Our CSAT score places us at the 68th percentile of SaaS companies"
- "Survey completion time is at the 90th percentile, respondents are taking longer than 90% of comparable studies"
- "Product quality perception is in the bottom quartile compared to competitors"
These statements are immediately meaningful to stakeholders in ways that raw scores and z-scores aren't.
When to Use Percentiles
- Benchmarking performance against industry norms or competitors
- Identifying outliers: observations below P5 or above P95 are often examined as potential outliers
- Reporting results to non-technical stakeholders who understand "top 25%" more easily than "1.5 standard deviations above the mean"
- Setting thresholds for customer segmentation (e.g., top-quartile customers by spending)
- Summarizing skewed distributions where the mean doesn't represent a "typical" value
Common Mistakes to Avoid
- Confusing "at the 90th percentile" with "scored 90%": the 90th percentile could correspond to any score value depending on the distribution; it means 90% of respondents scored lower
- Assuming percentiles are evenly spaced in value: the difference between P50 and P75 in score units may be very different from the difference between P75 and P100, especially in skewed distributions
- Using percentiles with very small samples: with 10 observations, percentiles are too coarse to be meaningful; you need at least 30-50 observations for stable percentile estimates
How Quali-Fi Supports Percentile Analysis
Quali-Fi's benchmarking features display your survey metrics alongside industry percentile norms, showing exactly where your scores rank relative to comparable studies. The platform calculates percentile distributions for all continuous metrics and highlights items that fall in the top or bottom quartile.
Benchmark your results with Quali-Fi
Frequently Asked Questions
Is the 50th percentile always the same as the average?
No. The 50th percentile is the median, the middle value when data is sorted. The average (arithmetic mean) can differ significantly from the median, especially in skewed distributions. For income data, the median (50th percentile) is typically lower than the mean because high earners pull the mean upward.
Can someone be in the 100th percentile?
Technically, being at the 100th percentile would mean scoring higher than 100% of the population, including yourself, which is logically impossible. Some systems cap at the 99th percentile; others report 100th for the maximum value. The convention varies by context.
How are percentiles different from quartiles?
Quartiles are specific percentiles: Q1 = 25th percentile, Q2 = 50th percentile (median), Q3 = 75th percentile. Quartiles divide data into four equal parts; percentiles divide it into 100 parts. Quartiles are a subset of percentiles.