Microsoft Excel 2010 : Macros and UDFs - Understanding How the Macro Recorder Works

This section is about the difference between recording a macro that will run successfully on a new dataset and one that will make you cry in frustration when it fails on a new dataset. You’ll rarely be able to record 100 percent of your macros and have them work on different datasets, but with the tips in the following subsections, you’ll greatly improve your chances.

The macro recorder is very literal, especially with the default settings. For example, if you have cell A1 selected, then begin the macro recorder and use the mouse to select your entire dataset in range A1:B10, this is what will be recorded:

Range(“A1:B10”).Select

If the next time you run the macro the dataset is A1:B20, your macro won’t run on all the rows. There are two things you need to change to make the recorded macro work properly. First, don’t use the mouse when selecting ranges. Second, don’t use the default settings of the macro recorder.

Navigating While Recording

To get the most out of the macro recorder, you should use the keyboard shortcuts to navigate the sheet, not the mouse. The reason is that some of the keyboard shortcuts translate to commands instead of specific cell selections. For example, if you record pressing Ctrl+down-arrow to jump to the last row in a column, you will get

Selection.End(xlDown).Select

If your dataset changes in size the next time you run the macro, the preceding line of code will be much more useful than if you’d recorded the macro by using the mouse.

Relative References in Macro Recording

The second rule to successful macro recording is to know when to turn relative referencing on and off. By default, it is off, which has it uses, but you’ll often want it on.

When relative referencing is off, the macro records specific cell addresses. Imagine you have a list of addresses similar to Figure 1. Each address is exactly three rows, and a blank row separates each address. To transpose an address to a single row, you would follow these steps:

If you record these steps with relative referencing off, you will get code that is cell specific, as shown in the text box in Figure 13.5. The results of running this code repeatedly on the data from Figure 1 is shown in Figure 2. A couple of repetitions of the code overwrites the first address line, ruining the record.

If this example was your first recorded macro, you might despair, as many before you have, at the uselessness of the recorder. But there is a way to make it work, and that’s to turn on the Use Relative References option found in the Code group on the Developer tab. Performing the same steps with relative referencing on returns the results in Figure 3. As the active (or selected) cell moves down the column, the correct fields are cut and pasted to the proper row and column.

Figure 3. Turn on the relative reference option while recording to record your movements instead of the specific cell addresses.

Avoid the AutoSum Button

If you use the AutoSum button while recording a macro, Excel will record the actual formula entered in the cell in R1C1 notation. It doesn’t record that you wanted it to select the range above or to the left of the formula. It’s just not that flexible. So, instead of using the AutoSum button, manually type in the formula mixing relative and absolute referencing.

For example, if you want to sum G2:G13, the AutoSum function will create the formula =SUM(G2:G13), or in R1C1 notation, =SUM(R[-12]C:R[-1]C). When viewed in R1C1 notation, you see how fixed the formula is. Although it will work in any column that it’s placed in, it specifically includes the cells 12 cells above (row 2) and directly above (row 13) the formula cell. The problem is if you add more rows, then the first cell is no longer 12 rows above—it’s more. The solution is to type in the formula manually, fixing the row for the first argument, as shown in Figure 4.