The idea of summing and then numerologically reducing the values of the cards in a spread to arrive at a single Major Arcanum card as a high-level summary of the individual details serves as a kind of “capstone” on the narrative. This practice seems to have entered the modern tarot world via the books of German tarot writer Hajo Banzhaf, but I recently learned that it has an historical antecedent in French cartomancy via the technique known as tirage en croix. (Good luck finding an English-language explanation of that term. You might try the Lenormand blog of writer and cartomancer Andy Boroveshengra.)
I have elaborated on the original concept by proposing the use of sub-quintessence cards that examine sub-sets of the overall layout, and a “Grand Quintessence” that integrates the values of the sub-quints with that of the normal run of cards. I’ve also adopted the practice of including the court cards in the calculation, something which Banzhaf advised against, since I believe all the cards on the table deserve to be included. The rationale for not doing so is that the court cards have no numbers on their faces, but they all have “concealed” values via their place in the series. Another alternative technique I’ve accepted is subtracting the values of reversed cards in a spread (reversal is a subject for another post), thus enabling reduction of the total to zero or below (something the usual method can’t produce), allowing the Fool to remain as “0” and introducing the idea of a reversed quintessence card. Here is a detailed explanation.