While it’s good to see sparklines as a native Excel feature, the Excel 2010 implementation is rudimentary, and third-party sparkline products have more functionality and more features than the native Excel sparklines. Yet it’s not necessary to deal with add-ins in order to realize expanded sparkline capabilities in Excel. Excel’s regular charts can be used to create decent sparklines, and this article will show you how.

Native Excel 2010 Sparklines

It’s easy enough to insert sparklines into an Excel 2010 worksheet. First, make sure you are not in “Compatibility Mode”. Compatibility Mode means the active workbook is as compatible as possible with Excel 2003; the most obvious feature is that the worksheet grid has the Excel 2003 numbers of rows and columns, not the expanded grid introduced in Excel 2007.

Select the data range or the location for the sparklines, and click one of the Sparklines buttons on the Insert tab. In this example, I selected the range where I wanted the sparklines to appear, then clicked the Column Sparkline button. The dialog shows the selected range in the Location Range edit box.

Then I selected the data range, which appears in the Data Range edit box. For some reason, the Location Range edit box is cleared (and each box clears itself when the other box is edited), but Excel remembers the selected range.

Here are the sparklines with the location range still selected.

Finally, here is the table with sparklines, with the active cell out of the way.

Create Sparklines From Regular Charts

The protocol for generating your own sparklines using regular charts is presented below. This protocol works well in Excel 2010 and 2007. In earlier versions of Excel, charts have a border of several pixels around the plot area, so the chart area must be sized larger than the cell you want the sparkline displayed in. In earlier versions, there is also a limit to how much the chart can be shrunk and still show the entire plot area, so you’ll have to shrink the chart only partway, then shrink the plot area to a smaller fraction of the chart area size.

Keep in mind that while you can use regular charts for sparklines, the small size of a sparkline limits the amount of information you should try to cram into one. Leave out labels and limit yourself to about two series maximum in any given sparkline.

Start creating your sparkline by selecting the data for a single series.

Insert a chart of the desired type. This is a typical Excel 2010 column chart.

Now simplify formatting. Below left shows the chart with the chart area border removed and the chart and plot areas made transparent, so borders and cell fill colors show through. Below right, the legend and any axis and chart titles have been removed.

The bars have been widened in the chart below left (by decreasing the gap width to 50%), and the major unit of the Y axis has been set to a small value, below right.

The axes have been hidden by selecting “None” for axis tick marks and axis tick labels and choosing “No line” for the axis line color. Note the size of the plot area within the chart.

For best results, extend the plot area almost to the left and right edges of the chart, and stretch the bottom of the plot area to the bottom of the chart. Leave a large margin between the plot area and the top of the chart. In fact, you may have to increase this top margin in the sparkline, after the chart has been shrunk to fit a cell.

Finally, shrink the chart and position it over the appropriate cell. If you hold down the Alt key while moving or resizing the chart, the chart edges will line up with the cell boundaries. If necessary, shrink the plot area from the top to increase the margin.

When all of the necessary formatting has been applied to the sparkline, and nothing else needs to be done, copy the sparkine and paste it into each of the other cells that need a sparkline. If you hold Ctrl and Alt while dragging the chart, a copy of the chart will be dragged into and aligned with the next cell.

All the charts are formatted identically. They also use the identical data, so let’s fix that.

Select the first sparkline. Notice how the source data is highlighted in the worksheet.

Select the second sparkline. Notice how the highlighted data is from the first row. So is the highlighted data for the third sparkline.

You could change the source data by choosing Select Data from the Chart Tools > Design tab, or from the right-click menu. You could also edit the chart series formula. But the easiest way to adjust the chart data is to drag the range highlight with the mouse. Move the mouse over the highlight until its border thickens, then drag it to the new range. If you’ve selected the plot area or chart area, both the series name and the Y value highlights move together. If you’ve selected the series itself, the series name and Y data must be changed separately.

Correct the third sparkline’s source data in the same way.

Here is the finished table with sparklines. Looks like the built-in sparklines.

You can use any chart type: here’s a line chart with markers.

Why Use Regular Charts?

There are a number of reasons to use regular charts rather than the built-in Excel 2010 sparklines.

Excel Version

One obvious reason is that you might not have upgraded to Excel 2010. If you’re using Excel 2007, the techniques shown here work the same way.

In Excel 2003 and earlier, the chart imposes a thicker border between the plot area and the chart area, so you have to oversize the chart to make the plot area fit as intended. Also, in earlier versions, the plot area only shrinks a certain amount within the chart area, so further shrinking of the chart truncates the plot area. The chart area can only be shrunk so far, but the plot area can be reduced further without shrinking the chart. The result is a chart that’s substantially larger than the cell it covers, but making the plot and chart areas transparent makes this no problem.

Mouseover Information

An important feature of a regular chart that is lacking in a sparkline, is the ability to mouse over a point and read relevant information from a popup. This is very useful in an interactive dashboard.

Combination Charts

What if you want to compare a time series to other data, such as a target value.

Native sparklines can only plot one timeline, while a regular chart has more flexibility (as do some of the third-party sparkline programs).

Remember the limited resolution of a cell-sized chart: adding the simple target data to these sparklines is almost not worth the effort.

Other Embellishments

You may wish to reverse the vertical axis. For example, your costs are probably reported as positive numbers, but you want to show them as negatives.

Native sparklines do not support reversing the vertical axis, but regular charts do.

Peltier Technical Services, Inc., Copyright © 2011.
 
Licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
 

I was pointed to the Bloomberg article by Kevin Drum of Mother Jones, who wondered in Everybody Hates Everybody Else why Bloomberg used a donut chart to plot the data, though he did not elaborate. We’ve been trained to cringe at pie and donut charts, and replace them with bar charts:

Ah, we can breathe freely again. And we can see that people blame Obama slightly more than Congressional Democrats. This 1 percentage point difference was not apparent in the donut chart, and really it may be meaningless given the poll’s 3.3 percentage point margin of error. Of course, the Republicans have collected by far more blame than any other category. This is clear in both charts.

There is one reason to use a pie for this kind of data, though. Stephen Few himself described The Secret Strength of Pies in Save the Pies for Dessert. Pie charts have an advantage over simple bar charts when you want to compare not single values, but subtotals. What the donut chart above shows is that the total blame assigned to Democrats, the sum of Congressional Democrats and President Obama, is less than the blame assigned to Congressional Republicans.

The pie chart below shows this difference even more clearly: we have not cut out the central region of the pie, which has central angles as well as the areas of the points to help us decode the chart’s values.

Sometimes this strength of pies is not realized, because data sorted differently may not have related slices in adjacent positions. However, if you know some categories are related (the two Democrat categories here), you can sort the data to make use of this feature of pie charts.

Then again, if you know some categories are related, you can convert the simple bar chart into a stacked bar chart, which sums related quantities linearly instead of circumferentially. This easier comparison allows us to see that Republican blame exceeds Democrat blame by almost twice the margin of error.

Bloomberg’s choice of chart type may not have been as bad as our initial reaction indicates, assuming the data was purposely sorted the way it was. Our stacked bar chart is an even better choice.

Peltier Technical Services, Inc., Copyright © 2011.
 
Licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
 

Here is a simple cross-tab table of data, showing how the strength of an experimental alloy varies as a function of alloy chemical composition and test temperature. In a former life I was a metallurgist, and this data was going to change the world.

2D Charts

These 2D line charts are the clearest plots you’ll see of this data. Instead of confusing the viewer by trying to render a third dimension on 2D paper or monitor, the data is plotted as 2D data, with the third dimension represented by multiple series (one series for each factor level along the third dimension). Instead of trying to get everything into a single chart, the data is presented as two charts, one plotting the data by rows, the other by columns.

The experts insist that line charts should only be used along axes that plot numerical variables, and not for categorical factors. That’s a little strict, and in this case, the Chem and Temp variables do change monotonically from low to high at nearly regular intervals.

3D Line (“Ribbon”) Charts

The much-maligned UI Ribbon introduced in Excel 2007 isn’t the worst ribbon in Excel. Among the most pointless 3D charts Excel offers is the 3D line chart, or “ribbon” chart, shown below in a handful of arrangements.

The data regions of the charts are shrunk because of the need to put all the third dimension infrastructure in the chart. This loss of resolution is rendered moot by the distortion caused by the perspective view.

It’s impossible to see where the “ribbons” lie with relation to the axes, or how they are positioned with respect to each other. And if the ribbons are aligned edge on to the viewer (see below right), they become nearly invisible.

The data labels on the lines in the 2D charts above make identification of the series very easy. It’s much harder to get labels positioned nicely in these 3D charts (I know, I tried and gave up). We have to use a legend, and the chart above right shows one problem with legends: the series are arranged purple to blue from top to bottom, while the legend entries are arranged blue to purple from top to bottom. Our eyes forced to move side to side to relate data to label, and our brains are forced to think about this backwards correlation, leaving less cognitive power to actually analyze the data.

Differences in plotting order can make the charts even worse. Left: strength for all temperatures is the same for chemistry factor 4; right: strength for all chemistries approaches the same value at the highest temperature. These conclusions are not borne out by the boring but accurate line charts above, nor by the tabulated data.

3D Area Charts

In the first 3D area chart below, the Temp 1 data obscures all the rest. In the second, the data in back is exaggerated because the unseen but progressively higher baselines of each plank are hidden by the areas in front. Blocking of rear data by front data occurs in 2D area charts as well, the the perspective and varying baseline is the third dimension’s contribution to this obstruction.

2D Column Charts

This 2D column chart isn’t a terrible way to show the data, though the taller bars make following the shorter series more difficult. This wasn’t a problem with the line charts.

3D Column Charts

There are two ways to make a 3D column chart. Left: keep all of the 3D bars in a 1D arrangement, so it’s just a 2D chart with thickness, and the resulting distortion. Right: arrange the vertical bars in a 2D horizontal grid, so the perspective distortion is made worse by bars in front hiding bars in back.

In the charts above the bars in front were only obscuring the baselines of the bars in back; in the charts below, which result from factors being shown in the opposite order, entire bars disappear in the matrix.

Wait, It Gets Worse

Who needs rectangular columns when you can use cylindrical columns? Very pretty.

You can further distort the data by using bottom-heavy cones and pyramids.

You can also make truncated cone and pyramid charts, formed by starting with shapes of the same height, and cutting them off at the height representing their values. The heights are no easier to read than in the pointy charts above, but perhaps you can estimate the areas of the cross-sections to help judge the values. Hmm, except areas increase as values decreases, and values changes in proportion to the square root of the area (that is, proportional to the width of the area).

If 2D Is Good, 3D Must Be Better, Right?

Mmmm, no.

All of the 3D charts shown above suffer from distortion. In the chart below, I’ve hidden the data, and made the borders of the walls visible, and added gridlines. No two lines anywhere in the chart are parallel. It’s very artistic, with lines extending to the vanishing point and all, but that only makes the data more difficult to judge. If you look at it long enough, the back corner suddenly changes, so it’s sticking out of the page. A chart showing your hard-earned data is no place for optical illusions.

The way these charts have achieved their third dimension is not through addition of a useful dimension to the data. This third dimension is achieved by adding chart junk, coloring loads of non-data pixels, and greatly decreasing the data-to-ink ratio.

So many of Excel charts can get very fancy, even without using gratuitous visual effects, such as shadows, gradients, glows, and transparencies. The 3D charts shown here are certainly fancier than the plain old 2D line charts at the top of the article, but not a single 3D chart shown here depicts the data as clearly as the line charts.

Peltier Technical Services, Inc., Copyright © 2011.
 
Licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
 

Excel offers a wide range of standard chart types. Below is the Chart Type dialog from Excel 2010, but all of these standard chart types have been available since Excel 97, and most of them since before that.

What makes my dialog different is that it has been annotated to show which chart types you may use freely, which types you may use with caution, and which types you should avoid. If you’ve been reading this blog for a while, and paying attention, you will probably not be surprised by most of the ratings.

Explanations for the ratings

Preferred

This group comprises the commonest and most easily understood charts you can make. Line charts and XY (Scatter Charts), with and without markers and lines. Simple clustered bar and column charts.

Acceptable

These charts are also readily understood, but might not be the first choice for most data sets. They should be used with “a little” caution.

Stacked charts are tougher to decode. Because the bars do not share a common baseline, their lengths are harder to compare.

Stock charts are tough for non-finance people to get, because they convey a lot of information in a specialized way. However, for people familiar with them, stock charts are very useful.

Use With Caution

This set of charts can be useful, if you understand their limitations and drawbacks, and if the audience is skilled in their interpretation.

Stacked line charts can be confusing if the audience does not understand that each line represents a cumulative total of that line’s data plus all previous lines’ data.

Ordinary 2D pie charts can be useful if the number of data points is small and you only use one at a time. Bar charts generally do a better job.

Area charts are problematic for several reasons. Unstacked charts risk areas in back being obscured by those in front, while stacked ones may have an issue if the audience doesn’t realize they represent cumulative data. Line charts are often a better option.

Bubble charts cause problems with interpretation of bubble sizes. Is the area or the diameter proportional to the data being represented? Also, the resolution isn’t too good. It’s often best to use bubble size to indicate discrete factors.

Radar or spider charts seem like they’d be good at displaying cyclic data, but the spoke length is often confounded by the areas within the lines, and it’s impossible to compare values which are not on the same spoke. Line charts are usually a better choice.

Surface and contour charts are the only 3D charts not rated “AVOID”. They can be useful for showing some data. The X and Y values are not treated as numeric variables, but as categories (though you can still use them for numerical values if the points are equally spaced along the axes). Some orientations may lead to features in front obscuring features in back. The wireframe versions have too many line segments, the filled ones may rely too much on color, and starting in 2007, the filled ones have a weird shading applied which often makes the chart less clear. However, when used with care, these can be useful tools.

AVOID

These charts should be avoided because they are difficult to interpret, and some of them have features which may actually distort data.

Most 3D charts are particularly troublesome. The audience has trouble lining up the data elements with offset axes, some elements hide other elements, and the perspective may stretch some elements and shrink others. In many cases the third dimension is a false dimension, created not with data but with lots of extra non-data ink.

All but the simplest 2D pie charts are problematic: donut charts, pie-of-pie and bar-of-pie charts, exploded pie charts. These all have features that make them even less comprehensible than regular 2D pie charts.

New Dialog

What if we just remove the charts rated “AVOID” from the dialog and do a little rearranging? This isn’t bad:

“Missing” Chart Types

There are various other chart types that aren’t in the Excel dialog, but which you can still make in Excel. You need the savvy to figure out how to arrange the data and how to mix various chart types, and you need the patience to repeat these tasks as necessary. There’s no decent Histogram (despite what the Analysis Toolpak claims), though you can make a column chart and fix it up. Similarly you can make your own Pareto Chart by adding a line series to a column chart. You can make Tornado Charts or simple Gantt Charts by pimping a bar chart. I’ve written some tutorials that show how to make tricky chart types and utilities that will do the work at the click of a button: Waterfall Charts (Bridge Charts), Box and Whisker Charts, Clustered and Stacked Column and Bar Charts, Marimekko Charts, Dot Plots, and Panel Charts.

Peltier Technical Services, Inc., Copyright © 2011.
 
Licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
 

This article demonstrates a protocol that works in both modern versions of Excel, that is, Excel 2003 and earlier and Excel 2007 and later. The technique is a bit convoluted, and it requires an expanded data layout to get the appropriate appearance. And there’s an additional degree of complexity to get the category labels to line up neatly under or beside the clusters.

For those who need to produce many of these charts, and who don’t have 15 minutes to spend on each one, I have created the Peltier Tech Cluster Stack Chart Utility, a commercial Excel add-in that does the heavy lifting at the click of a button.

Built-In Column and Bar Charts

Let’s start with this simple data set, which compares budget and actual values for three commodities for two quarters of the year. We want to have clusters for each commodity, with stacked actual values next to stacked budget values within each cluster.

Without any effort or thought we can easily create clustered column or bar charts from this data.

Stacked column and bar charts are just as easy.

It’s not obvious how to combine the chart types. The protocol involves inserting blank rows and cells into the data range of a stacked column or bar chart, and values only appear in some of the places in the chart. The proper arrangement will cluster stacks of values with stacks of zeros separating the clusters.

Starting the Chart

I’ll leave the original data alone (always a good practice) and create a staging data region which is linked to the original data. The easiest way to do this is to copy the original data, then use Paste Special Link to start building the staging area. We’ll make our chart first, then explore how modifying the data layout changes the chart. In practice, we’ll modify the data first and then make the chart, knowing the effects of data layout on chart appearance.

The first step is to make a stacked column or bar chart from the data in B6:E9.  There are no categories selected (i.e., the commodities are not part of the initial chart), so Excel just uses the counting numbers 1, 2, 3.

Since categories always start from the origin, the bar chart’s category labels go from the bottom up, instead of top down as in the sheet. So the vertical axis has to be formatted to make the categories go in reverse order. Also the value (horizontal) axis has to cross at the maximum category, which is at the bottom now, since the order of categories was reversed.

Adjusting the Data

So that’s only stacked. Let’s adjust the data by inserting some rows.

The stacks of columns/bars are now spread out. Not yet what we want.

But lets stagger the budget data by a row, to move the budget data points off the actual data and onto blank slots in the chart.

Unfortunately the staggering doesn’t happen automatically, so we have to go back and tell Excel what data range to use for the chart. Right click on the chart, then select Select Data or Source Data (the command is version-specific). Click in the Chart Data Range box, and select this whole data range.

One more adjustment to the data. Let’s insert a row at the beginning and end so there’s a space outside of the first and last cluster.

Again, we have to explicitly tell the chart about the updated data range. This is almost what we want.

Reduce the gap between columns/bars to give the chart a clustered appearance, and color code the data series to make it clearer which data series are associated.

In practice, it is not necessary to create a chart using the compact data and adjust it after every modification to the data. The correct protocol is to adjust the data, and then make the chart shown here, and proceed with adding labels, below.

Adding the Labels

Almost done. We need to add the category (cluster) labels. We’ll do this by adding a “dummy” series to the secondary axis, and the secondary axis will have the category labels we want. Add a column to the original data range for the dummy axis series (column F in our example).

Select this added data (F1:F4), and hold Ctrl while selecting the column with our labels (A1:A4), so that both areas are highlighted. Make sure you include the blank top cell in the first column. Copy the range, and use paste special to add this data to the chart as a new series, in columns, with series name in the first row and category labels in the first column.

In Excel 2003 and earlier, the original labels (1, 2, 3, etc.) remain along the axis, but in 2007, the new labels take their place, even if we hadn’t checked “Replace Existing Categories”.

Since zero value bars have zero height or width, they don’t appear in the chart. Just to show where this new series is added, I’ve temporarily replaced the zeroes in column F with values of 500. The series spans only the first three categories.

If you’re making a stacked-clustered column chart, convert this new series to a line chart type. Sometimes Excel 2007 doesn’t expand the legend enough to show the legend entry for Axis, so I’ve stretched it in this chart.

Manipulating the Axes

Now format the Axis series to place it onto the secondary axis. To do this, select the Axis series. If you can’t see the Axis series, click the dropdown in the top left of the Chart Tools > Layout or Format tab, and choose the Axis series. Then press Ctrl+1 (numeral one) to open the Format Series dialog. On the first tab, choose Secondary Axis.

Now add the secondary category axis, which is secondary horizontal in 2007 column charts and secondary vertical in 2007 bar charts. This command in on the Chart Tools > Layout tab.

We need to reverse the bar chart’s secondary vertical axis (like we did the primary when we first made the bar chart), and at the same time, make the horizontal axis cross in the automatic position, which generally means at zero or at the minimum.

Now we have to switch the position of the category labels. In the column chart, format the left hand vertical axis so the horizontal axis cross at the maximum, and format the right hand vertical axis so the horizontal axis crosses at the automatic (minimum) position.

In the bar chart, format the bottom horizontal axis so the vertical axis cross at the maximum, and format the top horizontal axis so the vertical axis crosses at the automatic (minimum) position.

We want only half a slot on the outside of the first and last clusters, not the full slot shown above, to center each cluster on the commodity category labels. Format the top horizontal axis of the column chart, or the right vertical axis of the bar chart, so the crossing axis is positioned between tick marks.

Select the secondary value axis, which is scaled from 0 to 1 (the right vertical axis of the column chart, or the top horizontal axis of the bar chart), and delete it.

Format the primary category axis, which is scaled from 1 to 10 (the top horizontal axis of the column chart, or the right vertical axis of the bar chart), and format it so it has no tick marks or tick labels, and no line type.

Finally, select the Axis legend entry. In Excel 2003 be sure to select the text label, not the legend key (the marker and line). Press Delete. In the column chart, format the Axis series to be invisible (no marker, no line).

That wasn’t so hard, was it? Though it did take a very long time.

Adding a Line to a Clustered-Stacked Column Chart

It’s relatively easy to overlay a line chart series onto the clustered-stacked column chart. Instead of the column of zeros we used to generate our commodity axis labels, put in the values you want to plot, and add a meaningful column header.

When you go through the process above to add your labels and manipulate the axes, you will end up with data points where you want them. Just don’t bother hiding the series at the end of the process. If you want to show the line on a secondary axis, despite my warnings to the contrary, don’t delete the axis, simply scale it appropriately to your data.

Adding a line to a cluster-stack bar chart is much more complicated, so I will not cover it here.

Peltier Tech Cluster Stack Chart Utility

This tutorial shows how to create Clustered-Stacked Charts, including the specialized data layout needed, and the detailed combination of chart series and chart types required. This manual process takes time, can be prone to error, and becomes tedious.

I have created the Peltier Tech Cluster Stack Chart Utility to create such charts automatically from raw data. This utility, a standard Excel add-in, lays out data in the required layout, then constructs a chart with the right combination of chart types. This is a commercial product, tested on hundreds of machines in a wide variety of configurations, which saves time and aggravation.

The Peltier Tech Cluster Stack Utility creates charts in either horizontal or vertical orientation.

Please visit the Peltier Tech Cluster Stack Chart Utility page or the Peltier Tech Cluster Stack Chart Utility Documentation page for more information.

Peltier Technical Services, Inc., Copyright © 2011.
 
Licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
 

There is more than one way to create a waterfall chart in Excel. The first approach described below is to create a stacked column chart with up and down columns showing changes and a transparent columns that help the visible columns to float at the appropriate level. Under some circumstances this simple approach breaks down, and another approach is described.

Floating Column Chart Data and Calculations

Here is some sample data showing how to construct a stacked-column waterfall chart. The left table has a column of labels, then a column with just the initial and final values, then columns with increases and decreases in value. This is the almost arrangement needed for making the chart, but I prefer to put these values into a single column as shown at right, and let the formulas sort it all out.

The first approach most people try is to use a floating column chart, that is, a stacked column chart with the bottom column in the stack hidden to make the others float. This range contains the calculations needed to make a floating column waterfall chart. After the two columns of labels and values, as in the top right table, there are calculated columns for the chart endpoints, the blank series that supports the floaters, and up and down values. Here are the formulas; the formulas in D3:F3 are filled down to row 7:

Cell C2: =B2

Cell C8: =SUM(B2:B7)

Cell D3: =MIN(SUM(B$2:B2),SUM(B$2:B3))

Cell E3: =MAX(B3,0)

Cell F3: =-MIN(B3,0)

The chart is pretty easy to make. Select A1:A8 (yes, include the blank top cell), hold Ctrl and select C1:F8 so both selected areas are highlighted, and create a stacked column chart.

Finish up with a little formatting. Set the gap width of the columns to 75%: format the series and on the Series Options or Options tab, change the value for gap width. Hide the Blank series by giving it no border and no fill, use colors that invoke positive and negative for the audience (usually green and red, which makes it tough for those with color vision deficiencies), remove the legend.

Data Crossing into Negative Territory: Breakdown with Stacked Columns

That seems just too simple to be true. And in fact, for data like the following, which has negative as well as positive values, the simple floating column chart approach fails.

The green and red bars are the correct length, and as long as they are located completely above the horizontal axis, the chart is cool. But the formula computing the blank values is too simplistic, and Excel prohibits the floating bars from floating across the axis.

You can still use stacked columns, but you need to compute two sets of up bars and two sets of down bars, one set of each that lies above the axis, and one that lies below the axis. You also need to fix the formula for the blank series so it floats each column above or below the axis as necessary, or provides no float if the column spans the axis. Wow, so complicated.

But wait!

Approach Using Up-Down Bars

There is another approach which takes a bit longer to chart, but the formulas are easier, and the columns in this case are able to float anywhere, even across the axis. This approach is based on line charts and a line chart feature called up-down bars. Up-down bars connect the first line chart value at a category to the last, like the open-close bars in a stock chart. In fact, Excel uses up-down bars as open-close bars in its stock charts. The up bars and down bars can be formatted individually.

The range below contains the calculations needed to make an up-down bar waterfall chart. After the two columns of labels and values, as above, there are calculated columns for the chart endpoints, and the values before and after adding an item to the previous total. Here are the formulas; the formulas in D3:E3 are filled down to row 7:

Cell C2: =B2

Cell C8: =SUM(B2:B7)

Cell D3: =SUM(B$2:B2)

Cell E3: =SUM(B$2:B3)

The chart-making process is a bit longer than for the floating column chart approach. Select A1:A8, hold Ctrl while selecting C1:E8, and create a line chart.

Select the Ends series and convert it to a column chart.

Select one of the line series, and add Up-Down Bars. In Excel 2007 and 2010, go to the Chart Tools > Layout tab, click the Up-Down Bars button, and select Up-Down Bars from the menu. In Excel 2003 and earlier, format the series, and check Up-Down Bars on the Options tab.

Hide the line chart series by formatting them to show no line and no markers, and format the up-down bar colors.

Remove the legend, and change the gap width of the column and the up-down bars to 0.75. This is easy for the column: simply format the series and on the Series Options or Options tab, change the gap width value. For the up-down bars in Excel 2003 and 2010, format one of the line chart series, and on the Options or Series Options tab, change the gap width value.

In Excel 2007 there is no way to change the up-down bar gap width from within the user interface, but you can do it with VBA. Press Alt+F11 to open the Visual Basic Editor. Press Ctrl+G (or go to View menu > Immediate Window) to open the Immediate Window. Type the following line of code into the Immediate Window (capitalization does not matter), then press Enter:

ActiveChart.ChartGroups(2).GapWidth = 75

Indistinguishable from the floating stacked column approach.

Data Crossing into Negative Territory: No Problem with Up-Down Bars

This is the negative trending data set that messed up the floating columns. The up-down formulas work just fine.

Follow the same process. Create a line chart.

Change the Ends series to columns.

Add up-down bars to the lines.

Hide the lines and markers and format your colors.

Change the necessary gap widths, and delete the legend.

Perfect, no problem with spanning the axis with our floating columns.

Stacked Columns for Positive and Negative Data

Earlier I said that it’s possible to use stacked columns for mixed values, and for completeness I’m going to describe the protocol here. If you don’t care to read about it, feel free to skip ahead, or to visit some of the other tutorials on this web site.

Here’s the start of the calculations for the stacked-column-across-the-axis approach. Here are the formulas for blanks above and below zero in D3 and E3:

Cell D3: =MAX(0,MIN(SUM(B$2:B2),SUM(B$2:B3)))

Cell E3: =MIN(0,MAX(SUM(B$2:B2),SUM(B$2:B3)))

Since at most only one of these has a non-zero value, we can replace the two formulas by a single formula which adds them together:

Cell F3: =MAX(0,MIN(SUM(B$2:B2),SUM(B$2:B3)))+MIN(0,MAX(SUM(B$2:B2),SUM(B$2:B3)))

Let’s consolidate the Blank column, and compute the other values. Here are the formulas; those in D3:H3 are filled down to row 7:

Cell C2: =B2

Cell C8: =SUM(B2:B7)

Cell D3: =MAX(0,MIN(SUM(B$2:B2),SUM(B$2:B3)))+MIN(0,MAX(SUM(B$2:B2),SUM(B$2:B3)))

Cell E3: =MAX(0,MIN(SUM(B$2:B3),B3))

Cell F3: =-MAX(0,B3-E3)

Cell G3: =MAX(0,H3-B3)

Cell H3: =MIN(0,MAX(SUM(B$2:B3),B3))

Select A1:A8, hold Ctrl while selecting C1:H8 so both areas are highlighted, and create a stacked column chart. Format the Blank series to hide it, format both Up series the same and both Down series the same. The result is almost identical to the up-down bar version, except here the horizontal axis is not hidden by the bars that cross the axis.

Waterfall Chart with Intermediate Cumulative Totals

It’s easy to accommodate intermediate totals in a waterfall chart. Adjust your formulas so the Ends series has a cumulative total and no red or green bars at the appropriate category. Construction of the charts is the same as without the intermediate totals.

Here is how the data and charts appear for a stacked column waterfall chart with intermediate totals:

Here is how the data and charts appear for an up-down bar waterfall chart with intermediate totals:

Peltier Tech Waterfall Chart Utility

This tutorial shows how to create waterfall charts in Excel, including the specialized data layout needed, and the detailed combination of chart series and chart types required. This manual process takes time, can be prone to error, and soon becomes tedious. If you want to add data labels, tedium increases.

I have created the Peltier Tech Waterfall Chart Utility to create such charts automatically from raw data. This utility, a standard Excel add-in, lays out data in the required layout, then constructs a chart with the right combination of chart types. This is a commercial product which has been tested on thousands of machines in a wide variety of configurations. Using this utility will save you time and aggravation.

Please visit the Peltier Tech Waterfall Chart Utility page or the Peltier Tech Waterfall Chart Utility Documentation page for more information.

Compound Floating Columns

Sometimes a waterfall chart may have two or more items stacked within a floating column. In the data table below, we see that Items A and B both contribute to the accumulating value. This kind of illustration only makes sense if the Item A and B values are all positive or all negative. Otherwise the chart will be confusing.

This table contains calculated blank data for a stacked (floating) column waterfall:

Select the data in column A, then hold Ctrl while selecting the data in columns C through E, and insert a stacked column chart as before.

Hide the blank series and the chart is complete.

A compound waterfall chart with up to two elements can be made using the up-down bar approach. There can be only one set of up-down bars per axis, so one set is on the primary axis and the other on the secondary axis.

Here is how the data is arranged for up-down bar waterfall. Here are the formulas that generate the table, which are filled down as far as shown:

Cell E3: =SUM(B$2:D2)

Cell F3: =E3+C3

Cell G3: =F3

Cell H3: =G3+D3

Select A1:B8, then hold Ctrl while selecting E1:H8 so both regions are highlighted, and insert a line chart.

Move the Before B and After B series to the secondary axis.

Select the secondary vertical axis (right of chart) and press Delete.

Convert the Ends series to a column.

Select either Before A or After A and add Up-Down Bars.

Select either Before B or After B and add Up-Down Bars.

Format the up-down bars and adjust gap widths.

If the endpoints are also split by item, the data looks something like this.

Here the blanks have been calculated for floating columns:

Construction of the chart is the same as before, starting with a stacked column chart, except there is no Ends series.

Here is the finished chart with Blanks made transparent.

Here the Before and After A and B values have been calculated for up-down bars:

Construction of the chart is the same as before, starting with a line chart, except there is no Ends series which must be converted to columns. Before and After A stay on the primary axis, while Before and After B move to the secndary axis.

Add two sets of up-down bars.

Hide the legend, hide the lines and markers, and format the up down bars. The chart looks just like that using stacked columns.

Because there are only primary and secondary axes in an Excel chart, the up-down bar approach can only support a two-item per stack waterfall chart. The stacked column approach can support many more items: the limitation is imposed by the legibility of the resulting chart.

Here is data for a waterfall chart with three items per stack (you could add enough items to make your chart illegible).

The data has a calculated Blank column to float the three columns.

Select the data in column A, and hold Ctrl while selecting the data in columns C through F, and insert a stacked column chart.

Hide the Blank series, and hide the unwanted legend entries (click once to select the legend, and click a second time on the legend entry, and press Delete).

Peltier Technical Services, Inc., Copyright © 2011.
 
Licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.