Since many of this blog’s posts will be about array-based tricks, it might be a good idea to have a proper introduction to Excel’s array formulas.

By default, Excel’s functions and formulas take one or more input values and produce a single output value. Think of the typical multiplication and SUMPRODUCT() function. When solving a more complex system of problems, however, it is often desired to process an array of similar values concurrently. Hence the array formulas.

Let’s start with the *real *basics. You’re the manager of a video game store, and for recordkeeping, you’ve just created a spreadsheet listing the prices of your inventory of video game consoles. Now you want the identical price listing to appear on another part of the spreadsheet, and here’s what you can do:

- Select an area identical in size to the source data array (in this case, 3 rows x 1 column)
- Type the “=” to get a formula started, then drag to select the source data array
- Press Ctrl+Shift+Enter to complete your first array formua

Note that, in the formula bar, a pair of brackets {} enclose your formula. This tells you that it’s recognized as an array formula.

Note: if you pressed the Enter key without holding down Ctrl and Shift, the brackets would not appear and only the first element in the array would be recognized. Additionally, you cannot type in those brackets.

An array does not have to be a one-dimensional vector. You can also reproduce an entire table by employing a two-dimentional array (3 rows x 2 columns):

So what now? Pretty boring and useless, huh? For this type of tasks you would typically use a simple “=” formula and copy it across the table. In fact, we’d usually recommend that as well. But this approach has its advantages – because the entire array (1D or 2D) is produced by one formula, you cannot change any of its elements (try typing over any of the cells… Excel won’t let you). When processing large amounts of data, it’s not always easy to check for human errors, i.e. if your coworker’s son messed up one of your 5,000 rows of identical formulas. Array formulas help ensure data integrity.

Moving on… now it’s the end of the month and your intern compiled the sales records for the store. But he listed the consoles horizontally on three rows instead of vertically! To fix this, you use the TRANSPOSE() function to “rotate” the table… turning rows into columns and columns into rows. This is a very handy function that’s strictly used with arrays.

To turn this 2×4 table into a 4×2 table

Use the TRANSPOSE() function… don’t forget, Ctrl+Shift+Enter!

Okay, before we wrap up this post, let’s look at something *actually* useful, i.e. more than simply duplicating values. Given your price list (a 3×1 array) and your intern’s sales records (a 1×3 array), you now want to calculate the total revenue. SUMPRODUCT() comes to mind since you want to multiply the price by unit for each product, and add them all together.

However, SUMPRODUCT() requires all of its input parameters to be of identical size. If you simply entered in the row array and the column array, you’d get a #VALUE! error. You could TRANSPOSE() one of the two arrays to match the dimension of the other, or you could simply use TRANSPOSE() within your SUMPRODUCT():

{=SUM([array1], TRANSPOSE([array2]))}

Note the brackets – don’t forget to press Ctrl+Shift+Enter. This calculation gives you the total revenue by first turning the two arrays into the same dimension, and then apply the SUMPRODUCT() function.

In the next article, we’ll talk about array sums, which lead into a whole wide world of possibilities.