Harpy Glossary

PCT (Percentage)

Amazon & D2C glossary · Harpy Media

PCT (Percentage) is the unit Amazon’s reporting speaks in: sales growth, conversion rate, Buy Box percentage, in-stock rate, defect and return rates are all expressed as proportions. It is the standardisation that lets metrics from different categories and periods be compared at all.

What is PCT?

PCT (Percentage) is the unit Amazon’s reporting speaks in: sales growth, conversion rate, Buy Box percentage, in-stock rate, defect and return rates are all expressed as proportions. It is the standardisation that lets metrics from different categories and periods be compared at all.

It looks too simple to need explaining, and it is where a great many misreadings originate. Percentages hide their base, so two figures that look comparable can be measuring entirely different things — and a proportion moving in one direction can mean the opposite of what the headline suggests.

Where the metric does its work

Growth percentages compare periods: week on week, month on month, year on year. Conversion rate measures the share of visitors who buy. Buy Box percentage tracks how much of the time your offer held the featured position. In-stock rate measures availability, and defect and return rates quantify post-purchase problems. Each is a proportion — and each is only meaningful against its base.

The value of percentages is comparability and clarity: they let a seller compare performance across categories, spot trends quickly, and identify where effort should go. The risk is that the same properties make them easy to misread — a change in a proportion can be driven by the numerator, the denominator, or both.

Reading percentages without being fooled

Two traps. First, base rates: 100% growth from a tiny base is arithmetically true and commercially meaningless, and high rates of change on small numbers are noise. Second, aggregation: averaging or summing percentages across different datasets produces numbers that describe nothing — a blended rate that does not correspond to any actual customer behaviour.

The habit that fixes both: always recover the underlying counts. Sellers who convert percentages back into units, sessions, and orders make better decisions than those who manage ratios directly, because the counts tell you what actually happened. Percentages are how the platform reports; counts are how businesses are actually run.

PCT = (Part ÷ Whole) × 100Example: 200 units out of a category total of 1,000 gives a 20% share. Before acting on any percentage, recover the base it was calculated against.

In practice

A seller reviewing their account sees a 40% rise in session growth from a new campaign and plans to scale it. They convert the percentage back to counts: the increase is 40% on a base of a few dozen sessions a day, within the noise band for the listing. Rather than scaling a campaign on a meaningless rate, they let it accumulate data — and spend the budget where the counts are real.

⚠️ Watch out. Comparing and combining percentages as though they were counts. A seller averages category conversion rates into a blended figure that matches no actual customer segment, then sets targets against it. Because every rate was computed on a different traffic base, the blended number is fiction — and the reporting built on it drifts further from reality the more carefully it is calculated.
💡 Harpy tip. Convert to counts before you decide. When a percentage moves, ask what the numerator and denominator did — units, sessions, orders — and whether the change survives that translation. Then judge it against volume: rates on small bases are usually noise, and rates on large bases are where the leverage is.

How Harpy Media helps

Metric discipline runs through our reporting work: percentages always read back against their base, blended rates avoided, and the underlying counts recovered before any decision is taken on them.

PCT FAQ

What does PCT mean in Amazon reporting?

Percentage — the proportional unit Amazon uses across sales growth, conversion, Buy Box share, in-stock rate, and defect metrics. It standardises metrics so they can be compared across periods and categories.

Why are percentages misleading in Amazon reporting?

Two reasons: they hide their base, so small-volume changes can look dramatic; and they cannot be meaningfully averaged or combined across different datasets, because each was calculated on a different base.

What should I look at instead of percentages?

The underlying counts — units, sessions, orders, returns. Convert every percentage back into the numbers it was derived from before acting, and judge whether the movement is real relative to the volume involved.

Want these numbers watched for you, every week?

Book Free Consultation

New guides, straight to your inbox.

Practical D2C playbooks as we publish them. No fluff, no spam — unsubscribe anytime.