Percentages That Sound Right but Tell the Wrong Story
Percentages can make tiny changes sound huge—or hide real risk in plain sight. This guide shows you how to read percentage claims with a critical eye so the math matches the story, not the spin.
Published on
Math
A headline says "Sales up 200%." Another says "Risk reduced by 50%." Both can be technically true and still mislead. Percentages are one of the most persuasive—and most abused—tools in communication. They sound precise, so we often forget to ask: percent of what, compared to what, and over what baseline?
This guide walks through the main ways percentages distort reality: small baselines that inflate drama, peer comparisons that hide context, and stacked or cherry-picked rates. Then it gives you a simple checklist so the math matches the story, not the spin.
The small-baseline trap
Going from 1 to 3 is a 200% increase. Going from 1,000 to 1,002 is a 0.2% increase. Same absolute change in both cases (2), but wildly different percentages. When the starting number is tiny, percentage changes look huge even when the real-world impact is small.
To read a percentage claim critically, always ask for the baseline and the absolute numbers. If someone says "we doubled sign-ups," ask: from what to what? Two to four is very different from 10,000 to 20,000. For a proper treatment of when to use percent change vs. a symmetric comparison, see our post on percent change vs. percentage difference.
A quick gut-check example:
A landing page goes from 3 sign-ups to 9 sign-ups. That is a 200% increase, but it represents only 6 additional sign-ups. Another page goes from 4,000 to 4,400 sign-ups. That is only a 10% increase, but it means 400 more customers. If you only look at the percentage, the tiny experiment looks like the bigger win.
When you see a big percentage on a small base, mentally translate it back into absolute counts: How many more people, dollars, or events is this really?
Peer comparisons without context
"We're 50% better than the competition" can mean many things: 50% higher revenue, 50% fewer complaints, or 50% more features. Even when the metric is clear, the comparison is often unfair—different time windows, segments, or definitions. Percentages make the gap sound scientific without revealing whether the comparison is apples to apples.
In experiments, the same pitfall appears: the formula you choose changes the number. Our article on A/B testing and which percentage formula to use explains when to use percent change vs. percentage difference so your results stay honest.
Health headlines and "50% risk reduction"
Medical news often reports that a treatment "cuts risk by 50%." That sounds dramatic, but it might describe a change from 2 people in 10,000 having a side effect to 1 person in 10,000. The relative risk was cut in half, yet the absolute difference is only 1 in 10,000.
To interpret these numbers correctly, always look for both relative risk (the percentage) and absolute risk (how many people are actually affected). A 50% reduction from 40% to 20% is totally different from a 50% reduction from 0.02% to 0.01%.
The same logic applies to "side effect is 200% more likely" claims. If the base rate goes from 1 in 1,000,000 to 3 in 1,000,000, that is a 200% increase but still extremely rare. Percentages attached to tiny probabilities need a second look before they drive big emotional reactions.
Stacked rates and cherry-picked windows
Multiple percentage changes are often quoted in sequence: "Up 20% last quarter and 15% this quarter." That doesn't mean "up 35%" overall—percentages don't add that way. Compounding matters: 20% then 15% on the new base is different from 35% on the original.
Similarly, picking a favorable start or end date can make any trend look good or bad. "Down 30% since last year" might mean "since the all-time peak last January"—a very different story from "down 30% versus the year-ago average." When you see a percentage, ask: over what period and from what starting point?
Worked examples: turning headline math into real numbers
Suppose a subscription app claims: "Churn dropped 25% this quarter." Last quarter, 8% of users canceled each month. This quarter, 6% canceled. The math is:
(6% − 8%) ÷ 8% × 100 = −25%
The percentage change statement is mathematically correct, but the absolute reality is simply "2 fewer people out of every 100 cancel each month." When you translate the percentage into a concrete count, it is easier to judge whether the improvement matches the marketing language.
Next, imagine a press release that says: "Cyber attacks up 300% year over year." If reported incidents went from 5 to 20, that is a 300% increase—but it might be mostly better reporting or a bigger customer base. Use the Percentage Change Calculator to run the numbers, then ask: how many more events is this, and what changed about how we count them?
Percentage vs. percentage points
When a rate moves from 5% to 7%, the change can be described in two ways: it increased by 2 percentage points or by 40%. Both are mathematically valid, but they sound very different.
| How it's reported | What it means |
|---|---|
| +2 percentage points | The rate itself moved from 5% to 7%. You add or subtract points directly. |
| +40% | You're measuring relative change: (7% − 5%) ÷ 5% = 40%. The new rate is 40% higher than the old rate. |
In public communication, "percentage points" is often the clearer phrase when you're talking about interest rates, tax rates, or approval ratings. Reserve "% change" for when you explicitly want to describe the relative growth of one rate compared with its starting value.
How to read percentage claims so the math matches
Use a short mental checklist: (1) Baseline—what is the denominator? (2) Absolute numbers—what are the actual values? (3) Comparison—is this percent change over time or a symmetric comparison between two things? (4) Window—what time span or segment is being used?
Putting percentages in context is a skill. For everyday use, our real-world guide to percent change and our everyday math calculators show how to measure relative growth and choose the right tool so your own numbers tell an accurate story.
Summary: When Percentages Mislead
- Small baselines: Tiny starting numbers can make "+200% growth" sound impressive when it might mean only a handful of people or dollars. Always pair the percentage with the absolute change.
- Peer and risk comparisons: "50% better" or "50% risk reduction" says nothing about what is being measured, how it is defined, or how common the outcome is. Look for both the exact metric and the absolute risk (e.g., 1 in 10,000 vs 1 in 100).
- Stacked and headline rates: Percentages don't add across time—20% then 15% is not 35%. Decode claims like "up 300%" by asking how many more events that really is and whether counting or the audience changed.
- Percentage vs. percentage points: A move from 5% to 7% is +2 percentage points but a 40% relative increase. Use percentage points when you mean the rate itself changed, and % change when you mean growth relative to the starting rate.
- Your checklist: Baseline, absolute numbers, comparison type (change vs. difference), time window, and whether the claim is about risk, rate, or count—so the math matches the real story, not just the headline.