You saved an app’s estimated revenue last month. Today the number is higher. The percentage change is easy to calculate. Deciding what changed takes more care.
Perhaps the time window moved from a calendar month to a rolling thirty days. Perhaps the country filter reset. Perhaps the vendor revised its historical model. Perhaps the app really did grow. The calculator cannot distinguish these explanations for you.
Keep the measurement recipe
For every observation, save the source, app identifier, platform, geography, metric definition, currency, period covered and time retrieved. A screenshot is helpful context, but a small structured record makes comparisons less dependent on memory.
Only calculate a growth rate when the numerator and denominator measure the same thing. If last month’s figure covers one country and this month’s covers the world, the result is a filter-change percentage, not business growth.
Watch the starting number
In a hypothetical example, moving from $100 to $300 is 200% growth. Moving from $100,000 to $150,000 is 50% growth. The first percentage is larger; the second absolute increase is much larger. Neither tells you profit or retention.
Always show the starting value alongside a dramatic percentage. If the starting value is zero, an ordinary percentage growth formula is undefined. Describe the absolute change instead of displaying infinity as a business achievement.
Treat revisions as revisions
If a data provider changes an earlier estimate, preserve the original observation and record the new version. Quietly overwriting history makes it difficult to tell whether your alert detected market movement or a measurement update.
For idea research, look for a sustained pattern and supporting evidence before declaring a breakout. A launch, seasonal event or short promotion may explain an isolated jump.
The useful output is a dated, comparable series with an honest interpretation. The chart can be exciting after the measurement recipe is boring.