If you have noticed that a sensor's highest reading (for example CO₂) looks different depending on the timeframe you select, there is nothing wrong with your device or your data. This is expected behavior and comes down to how graphs display data over different time periods.

When you view a single day or week, the graph has relatively few data points to plot, so it can show your air quality in fine detail. Short-term events, for example a CO₂ spike while a room fills up, appear clearly and exactly as they happened.

A monthly or yearly graph has to fit a lot more data onto the same screen. To keep the graph readable, the system groups readings into larger time blocks and shows an average for each block.

This averaging is what smooths the graph out. A short spike gets blended together with the surrounding readings, so the highest value shown on a long-term graph will naturally be lower than the peak you would see by zooming into that specific hour or day.

On the day view, the CO₂ spikes at 714 ppm in the afternoon

Zooming out to the month view, that same day's peak now shows as 592 ppm, because the spike has been averaged together with the surrounding readings

Both are correct. They just answer different questions:

Use the day or week view to see an exact peak. Use the month or year view to understand general air quality trends.