At ten on Monday morning, a red card might say sales are down 80%. If it compares four hours of activity with the entire previous Monday, the alarm comes from the calendar. Before searching for a commercial explanation, inspect the two intervals. A percentage can be mathematically correct while answering the wrong question. Establish a comparison rule before using colour or arrows to judge performance.
Compare equivalent elapsed time
To analyse today, use a comparable cutoff on the reference day or wait until the current day ends. Complete weeks are often easier for a regular weekly review. Put dates next to the figures rather than simply labelling them previous period. If weekdays and weekends behave very differently, check that the comparison has not silently changed their proportions. A seven-day total and a partial week need an explicit explanation before anyone draws conclusions from them.
Separate the calendar from data maturity
The end of a day does not guarantee that every report is final. Google explains that Analytics processing can take 24–48 hours and that figures may change during processing. Check each source separately. Do not pretend Search Console has a more recent cutoff just to match another card. When sources have different reporting boundaries, show those boundaries. Readers should be able to tell the difference between provisional information and a completed reporting window.
Show the baseline behind a percentage
In a hypothetical example, moving from 100 purchases to 120 is a 20% increase. Moving from two to four is a 100% increase, despite an absolute change of only two. Show both values and the starting volume. When the reference period is zero, percentage growth has no valid denominator: display the absolute change and explain why a percentage comparison is unavailable. An impressive infinity symbol helps nobody decide what to investigate next.
Add commercial context before acting
A promotional week is not equivalent to a normal-price week merely because both contain seven days. Record campaigns, stock shortages and price changes. You can retain the main comparison and add another one answering a different question, such as the same promotion last year. Explain how the circumstances differ. Do not select only the reference that produces the most flattering result. Agree the comparison rule before presenting the figures to your team or a client.
Leave a repeatable comparison
Save the source, time zone, dates, filters and extraction time. Someone reopening the report should understand what you measured and why later figures might differ. For recurring meetings, establish a review time and a rule for provisional data. ClawStud offers a period picker in the dashboard, but you still need to decide which comparison serves your business. End with a hypothesis you can check, an owner and a next review date rather than relying on a card's red or green colour.
Official reference: Google Analytics: Data freshness.
Read next: How to review your store’s purchase funnel in GA4.
