Pushing data to widget vs. widget retrieval

I’m wondering about the differences (pro’s and con’s) between pushing data to be displayed by a widget versus having the widget call up the data.

I have a sketch that reads ambient temperature at my Arduino MKR1000 and pushes that value to a widget on my iPhone, via a virtual pin. I believe I could also set a frequency on the widget to initiate a BLYNK_READ and retrieve the data that way. Is one method preferred or better than the other?

It has been mentioned a few time in the forums… but basically having the Widget poll for data (and any associated processing in the sketch via BLYNK_READ) only works when the App/Project is active. Generally used for minimal programming and small, simple applications.

PUSH (and BLYNK_WRITE) allows the App/Project to be completely offline but the device still does all the work autonomously, via timers, and the server still records all the data for later App retrieval. This is generally the prefered method.