Designing Your Own Spread
When no existing spread fits, you can design your own. It does not have to be wholly original: swapping new themes into a proven structure and keeping the positions logically related is more reliable than inventing something from nothing.
Open-ended spreads: Pollack's three innovations
Besides the classic spreads with fixed positions and a fixed number of cards, there is a self-designed "work cycle" spread whose core is three innovations. They are useful for anyone who wants to design or adapt a spread of their own:
- Oriented toward advice: the positions are designed around "which direction to put effort into," going beyond simply listing how things stand.
- Open-ended, with repeatable positions: after the planned cards are read, you can keep turning over more, and the reader may return to the same position again and again. Rearranging the later cards from different assumed starting points, then seeing which of the resulting versions best matches the querent's real reactions, is a concrete way to test whether the reading is on track. At the end you can ask the querent directly, "Which of these sounds most like your situation?" and use their feedback to calibrate the reading, instead of the reader reaching a verdict alone.
- Positions grouped and read by stage: the spread is divided into stages (for example "past experience," "expectations," "where effort is going now," "outcome," "what follows"). The cards within a stage are read together as a group of traits working jointly, rather than explained one by one in isolation and then simply strung together.
This kind of self-designed spread points to a more general principle. The name of a position is only a starting point. When reading, you can treat several neighboring positions as one whole to understand the situation, as long as you explain to the querent beforehand what those positions are saying together.
Concrete steps for designing a new spread
Rather than dreaming one up from scratch, a more realistic path to a design has four steps:
- Start from a real need; it does not have to be wholly original. Adding to, cutting from or swapping concepts within an existing classic spread is a widely accepted practice. A spread is judged by whether it meets the querent's real needs and solves a specific problem. Being "never seen before" is not the measure. If nobody is asking the question, it does not matter how novel the spread is.
- Mark out the specific range of questions the spread will cover. First talk with the querent until it is clear which aspects they actually want to know about, and list the related points that fit into one spread. At the same time, hold the line: do not force two unrelated dimensions together just to fill the spread, especially pairings with questionable values (such as tying "money luck" directly to "will I meet my ideal partner"). This also echoes the reminders about the limits of divination in The Ethics of Reading.
- Give the spread a recognizable esoteric symbol as its skeleton, instead of a bookkeeping-style layout of cards in a row. For example, organize the positions with a planetary symbol or a geometric figure. This helps you and others remember the spread's shape and the order of its positions, and it also gives the spread its own identity and makes it easier to share. The pentagram in The Five-Element Spread and the Jupiter symbol in Career and Year-Ahead Spreads are examples of this idea.
- Keep the positions symmetrical and logically related. For example, a group of positions on the left stands for "me" and the right for "the other person," forming a mirror structure. A spread like this is easier to remember and use than a jumble of positions with no connection to each other.
There is also an easily overlooked pitfall in design. Building a spread around "list every option of a certain kind and draw to see which fits best" (for example, listing several common ways people meet and drawing to see which way the querent is most likely to meet a partner) looks thorough, but it has two hidden problems. First, it is hard to truly list every option, and there are always exceptions the designer did not think of. Second, if the querent draws good cards for several options, you end up with no conclusion that tells them apart. When designing this kind of "list-style" spread, check consciously for both points.