My working theory states that there is a balance of caption length and number of people in the photo, supplemented by an X factor in each photo which all together determines the “desirability” of the photo for likes/comments.  Captions cannot be excessively wordy, and (depending on gender of the instagrammer) number of people in the photo should generally be more than one but not more than about 10.  The X factor can make up for deficiencies in either area. I define X factors as something that really separates one picture from the rest of the pictures in one’s feed. This can be a setting, a pose one strikes, props, other people, the circumstances surrounding the post, etc. In my case, I had left posting on instagram for about a year, then suddenly returned, and found that the number of likes I received per photo went up significantly.  This would define my X factor of those photos as just the shock value of seeing my face on the gram again.