Trusted by 10,000+ marketers

  • Private planning
  • Instant access
  • Publication insights

Find your next useful tool

Popular tools and guides

Measurement

Calculate percentage growth without hiding the baseline

Report the starting value, ending value, period and calculation alongside a percentage change. A large percentage can come from a very small baseline, and a zero baseline needs special handling. The number should help the reader understand the observation rather than make a chart look dramatic.

Illustration of an analytics dashboard

The short answer

Report the starting value, ending value, period and calculation alongside a percentage change. A large percentage can come from a very small baseline, and a zero baseline needs special handling. The number should help the reader understand the observation rather than make a chart look dramatic.

A worked example

In an explicitly fictional worksheet, visits rise from 20 to 30. The increase is 10 visits, or 50% relative to the starting value. A second example rises from 2,000 to 3,000 and also produces 50%. Showing the absolute values prevents the identical percentage from implying identical business impact.

A practical checklist

  1. Use comparable periods and the same metric definition. Record the actual source values before applying any formula or formatting.
  2. For a nonzero starting value, calculate (ending minus starting) divided by starting, multiplied by 100. Report the absolute difference as well.
  3. If the baseline is zero, describe the change from zero in absolute terms rather than dividing by zero. Add sample size and context when interpreting small counts.

What to avoid

Avoid using illustrative growth examples as client results or presenting a percentage without its denominator. Do not imply causation from a change alone. The same observed increase may coincide with advertising, seasonality, product changes or measurement changes that require separate investigation.

A useful follow-up

Can I report 100% growth from zero to one?

Not using the standard relative-change formula. Report that the count increased from zero to one and explain the baseline.

Editorial note. Review the stated author, sources and publication scope before applying this guidance. Examples are illustrative, not customer results. No ranking, citation or commercial outcome is guaranteed.

Continue with context

Related reading

A practical SEO measurement plan before publishing

Define the business question, metric and observation source before publishing a campaign report. Separate discovery, visits and completed actions so the report does not imply that they are interchangeable. A measurement plan should also specify unavailable data and the limits of attribution.

Read the guide ↗

Search Console and analytics answer different questions

Use Search Console for the search visibility and click information it provides, and analytics for the on-site behavior your implementation actually measures. The datasets have different scopes and methods. Investigate discrepancies rather than forcing their totals to match or treating either one as a complete account of user intent.

Read the guide ↗

Keep missing metrics separate from zero

Represent an unavailable observation as missing, not as a measured zero. Zero is a real value; missing means the evidence is not present. This distinction affects sorting, filtering, averages and the decisions readers make from a comparison table.

Read the guide ↗