The other day, the song “Counting Blue Cars” by Dishwalla came on and I found myself singing along with the lyric: “We count only blue cars…”

Nerd that I am, I couldn’t help but immediately conjure up an Excel formula to help them out:

=COUNTIF(Cars, “blue”)

It might be completely missing the point of the song (about which I’m clueless), but being able to count and sum based on conditions in Excel can be really useful.

The formula I used for counting blue cars is a good example of a simple COUNTIF doing what it does best: counting only those items that match the criteria you specify. In that formula, “Cars” could be replaced by a range of cells in an Excel worksheet that contain the names of different colors. The formula will return a number that represents how many of the cells in that range match the criteria.

COUNTIF’s summing counterpart is known as SUMIF. It works in much the same way: you tell it which cells need to match a criteria, you specify the criteria they must match, then you specify which cells will be summed as a result. This last group of cells can be in another column if needed. For example, a SUMIF formula can examine the text in cells C2 through C16 to determine which ones contain the text “red,” then sum the values in the corresponding cells from column B. So if C5 and C8 were “red,” the values in B5 and B8 would be summed.

In these examples, we’re just writing the criteria inside double quotes so that an exact match must be found. But let’s say you wanted to count how many cells match an approximate condition. For instance, if you specify the criteria to be “*ito” the formula will return a count of how many cells end in the letters “ito.” So burrito, taquito, and mosquito would all match the criteria. Mmm…I’m starting to get hungry now…and slightly itchy.

The asterisk in that formula is called a wildcard—it represents any number of characters. Likewise, you can use a question mark to represent a single character.

When you need to count cells that don’t have similarities, such as enchiladas and wool sweaters, using wildcards won’t work; but you can combine multiple COUNTIF formulas to get the results you need.

If any of this sounds confusing, just check out this week’s tips and you’ll get a better idea of how to set up these formulas.

Links to the tips are on the blog entry for this week’s podcast at www.productiveUpodcast.com, and they include:

• Let’s Sum If It Fits: Using the SUMIF function
• Count By Conditions: Using the COUNTIF function
• Creative Conditions: Using wildcard characters as criteria for SUMIF and COUNTIF
• Summing This Not That: Using the SUMPRODUCT function to sum or count using multiple conditions
• Arrayed in Splendor: Creating an array formula to sum based on multiple conditions

We’d love to hear your feedback on the tips and the podcast, so head on over to iTunes to write a review for us! And don’t forget that you can follow us on Twitter @productiveUcast.

Thanks for joining us for this week’s podcast. With these tips, you may soon find yourself counting blue, red, and green cars all with a single formula. See you next week!

Music courtesy of Clayton & Fulcrum

It was a dry and brisk November evening when I got the call. Sitting at my desk with the blinds cracked just enough to allow the last rays of sunlight from the dying day to creep through, the ringing of the phone shattered the silence like a brick through a pane of glass. The dame was one-half shrill hysteria, two-thirds livid frustration, and just a splash of hopeless desperation. The recipe for a typical batch of Formula Frenzy. Combine and shake well; serve chilled and aggravated.

She had just written a formula in an Excel worksheet that was supposed to calculate her income and bill payments and return a tidy little budget for monthly spending. Instead, it returned the dreaded #N/A error. This is where I come in.

I’m a formula detective. It’s my job to examine the data, peer deep into the dark underlying elements that comprise a failing formula, and eliminate the errors. I told her I’d take the case. After all, I know the usual suspects; more importantly, I know the tricks to undo their devilry.

It’s not that hard, really. Most of the tools needed for the job are built right into the Excel application and readily available. I prefer to start with the “Evaluate Formula” feature. This one really lets you step right into the formula and break it down to pinpoint where the error occurs. That kind of hands-on approach takes a lot of the guess-work out of a long and complex formula, and has helped me successfully close many cases of Formula Frenzy.

For some of the simpler formulas, I tend to go right to the source: the Formula Bar. If you want to see what’s happening behind the scenes, this is for you. You can use the F9 key in the Formula Bar to convert a highlighted portion of a formula to its resulting value.

Another good trick to know is how to force all the formulas in the worksheet to show themselves. Sometimes it’s hard to tell what’s lurking underneath those cells, but the “Show Formulas” feature (also called “Formula Auditing Mode” in some versions of Excel) will bring all the formulas out of hiding.

It can also help to know where a formula’s coming from. Sometimes, being able to see which cells directly impact a formula’s result can shed some light on why it might be producing an error. As it turns out, Excel couldn’t have made this any easier. You can use the Trace Precedents arrows to point out exactly which cells a formula is affected by. The Trace Dependents arrows will show you which other cells are affected by that cell.

Then there are those cases of Formula Frenzy that can really leave your head spinning. This is typically a symptom of Circular References, which can occur when Excel is trying to calculate a formula that includes a reference to the cell it’s in. Makes me dizzy just thinking about it.

I’ve seen some bad cases of Formula Frenzy in my line of work. It’s not something I’d wish on anyone, so I’m leaving you with some tips that should help you avoid having to call me to bail you out when a formula goes awry.

The tips are posted on the blog entry for this podcast at www.productiveUpodcast.com, and they include:

• Let’s Evaluate: Troubleshooting faulty formulas with Excel’s “Evaluate Formula” tool
• Behind the Scenes: Using F9 to evaluate formulas in the Formula Bar
• Genealogy of a Formula: Using the Trace Precedents and Trace Dependents arrows
• Revelio Formulae: How to toggle the display of formulas on a worksheet
• Going in Circles: Understanding circular references
• Hidden in Plain Sight: Using the N function to add comments inside formulas

This should be enough to help you get your own formula detective agency up and running. Next time you come across a formula playing by its own rules, you’ll know just where to start snooping to solve the mystery of the #N/A error.

That’s all for this edition of the podcast. Until next week, be sure to head over to iTunes to write a review of the podcast, or drop us a line on Twitter @productiveUcast.

Music courtesy of Clayton & Fulcrum

Well, it happened again. I’m almost positive I put both socks into the dryer, but only one came out. Tell me: where do the lost socks of the world go when they disappear? The Island of Misfit Socks? An exclusive club for singles? Maybe they have their own dating website, and they all go online in the hopes that they’ll find their perfect match. If so, I hope my argyle sock reads this:

Dear sock,

Your perfect match is here, in my dresser drawer, and it’s lonely. This is where you belong; come home.

What a frustrating experience. Sure, I could go buy a new pair, but then my original old sock would be a third wheel. The new pair of socks would try to have some time alone together and then this old sock would be hanging around and it would eventually just get awkward.

What I really wish is that life had something like Excel’s MATCH function. In case you’re not familiar, let me take a moment to explain it. If you have a worksheet that contains a column of data—let’s say, for example, articles of clothing—and another column that contains some or all of the same data, but in a different order, the MATCH function can help you determine which items appear in both columns. Not only that, it will also give you the row number in which the item appears in the first column. So if you know that one of the items in column B is “Sock,” and you want to know where that item can be found in column A, you can create a MATCH formula to tell you where in column A that item appears. It’s like map coordinates—better than GPS for misfit socks!

Once you have that formula mastered, you might want to try a more advanced formula like a VLOOKUP. This function will find a matching value in another set of data and return a value from an adjacent column. How does it work?

Well, let’s say you’re working in Excel and you have two worksheets of data. One contains Employee Names in column A and Employee IDs in column B. The second worksheet also has a series of Employee Names in column A, but they’re in a different order than the ones on the first worksheet. So if you want to find out the Employee IDs for these names, you can use the VLOOKUP function to look on the first worksheet for a match for each Employee Name on the second sheet, and it will return the correct Employee ID.

This can be extremely useful, especially in cases where you have thousands of records for which you need to return corresponding data.

There’s a lot you can do with Excel’s lookup functions if you know how to properly set them up. Here are links to some tips that will help you get started.

• Does This Match? – Using the MATCH function
• Vertical Challenge – Using the VLOOKUP function
• Look It Up – Using the LOOKUP function
• Data on the Horizon – Using the HLOOKUP function
• Show Me the Money – Using absolute references in formulas
• Go Away, #N/A – Checking for errors in formulas
• Pass or Fail? – Using VLOOKUP in a grading system
• Getting Clean Results – Nesting other functions in a VLOOKUP

You might also want to check out the tips in our previous podcast, Excel Hates Me! When 2+2 Isn’t Always 4 (VOL029), to make sure your data is clean. Otherwise, you might run into some errors when trying to find a match with these formulas.

Well, Excel’s lookup functions are really handy for locating data, but it looks like I’m going to have to hunt for this sock on my own. Shhh…be vewwy, vewwy quiet!

Thanks for joining us for this week’s podcast. Be sure to leave us a comment if there’s anything you’d like to see featured in a future episode.

And don’t forget that you can follow us on Twitter @productiveUcast, and you can also find the podcast on iTunes if you want to write a review.

See you next week!

Music provided by Clayton & Fulcrum

This week’s edition…

“Excel hates me! When 2 + 2 isn’t always 4″

You know, one of the most frustrating (and often the most fascinating) part of using Microsoft Excel has to be just how helpful it tries to be.

For instance: I type the number 2 in a cell and again in the cell right under it. Go just below that and click the wonderful “add ’em up” sign (if you’re not sure what I am talking about, look on the toolbar for the button that looks like a Greek letter Σ) and presto—I get 4! Great! This is exactly how it should work, right?

Well, I decided to copy a table of information from the Internet the other day and tried to add up the numbers that I pasted, and Excel could not have been more wrong.

I was really curious as to why this might be happening. I didn’t feel like I could trust my good buddy Mr. Gates and his math processing whiz kid Excel. So I asked myself, “Self, are you crazy enough to act like someone in a horror film who knows the monster is going to get you if you go down into that dark place alone?” And just like in the movies, of course, my brain went, “But…it will be different this time. I’ve got your back.” So down the hole I went, searching for that ever-elusive spark of sanity.

First stop: little green symbols… Ooooo! Are they alien lights? Are they indicators of friends waiting for me to find them? Nope. Turns out, it was one of the newer features added a few years back to Excel that I tend to ignore. You know, like that crazy maintenance light on the dashboard of your car that you keep hoping goes away.

So what do they call these friendly helpers? They call them smart tags, as in, “Good day, ol’ chap! Tag, you’re it!” These little indicators can give you a lot of information on things that might be wrong with your data. Like, for instance, formulas that aren’t consistent. Or, in my case, numbers being stored as text.

Numbers being stored as text? Aren’t numbers just numbers? I mean, when I see a 2, it looks like a 2, smells like a 2…so why can’t Excel accept that it is a 2? As it turns out, when you add data to your spreadsheet from other sources like external databases, imported text files, or pasted data from a Web page, sometimes there are hidden little ghoulies in the system that you can’t see.

Here are a couple of common creatures that you sometimes conjure up when you’re trying get your information based on mojo created in other places.

• Spaces. Yes, spaces. Those little blank things between words, or my ears—you know, the empty kind. Just because you can’t see them doesn’t mean they don’t exist.
• Non-printing characters. Often information coming from different database systems can carry things like line breaks or carriage returns, useful for text-based applications, but useless for most people when they need to work in Excel.

So, what do you do? I say treat it like liposuction and just TRIM out the fat. Well, maybe not the same. TRIM is actually a formula function you can use to remove unwanted spaces in the content of another cell.

Or try using Find and Replacê, by first copying one of these “spaces” or even non-printing characters (you may not be able to see them, but you can touch them if you’re real careful), then run the Find and Replacê. When deciding what to use to replacê them with, use nothing. Just remember: you want to get rid of it and substitute in its place what should be there—nothing.

There are lots of methods and little quirks when it comes to getting rid of these little pests, so this week we have included some helpful tips on how to use many different techniques to exorcise that “evil monkey” in your closet. So why not stop on by and check them out at: www.productiveUpodcast.com.

This week’s tips:

• Mass Conversion – Using SmartTags
• You Get the General Idea – Using Text to Columns
• Just a Trim – Using the TRIM Function
• Clean it Up – Matching up values
• Trading Spaces – dealing with non-breaking spaces
• True Value – Fixing problem numbers using the VALUE function

Well, that’s it for this week. If you found these tips useful or have some questions you would like to see covered in a future episode, you can leave us a quick note on Twitter by following @productiveUcast. Or, if you just can’t get enough of us, let everyone know by writing a review of the podcast on iTunes.

See you next week!

Music Provided by Clayton & Fulcrum

Who really knows which one to use? We all understand the old phrase, “A picture is worth a thousand words,” but which picture (or, in this case, which chart) gets me closer to that thousand instead of 3?

Today we are going to cover some basic chart types and what picture they draw. So, like Prometheus giving fire back to man, I now give the knowledge of Redmond back to the masses.

The four most commonly used chart types are Column, Bar, Line, and Pie. These are not the only chart types, but they cover what most mere mortals like you and I can probably get the most out of.

Let’s start with the column chart. The column chart is good for comparing values across multiple categories. For example: Who had the larger population; the Romans, the Greeks, or the Egyptians? It’s pretty straightforward. You have three categories (Romans, Greeks, and Egyptians), and their associated values (in this case, population numbers). This type of chart is also good if you plan on creating trend lines for forecasting, and no — the trend lines are not the same as getting your fortune told by the Oracle of Delphi.

Next, let’s take a very similar chart, the bar chart, which is also good for comparing multiple values. It serves the same purpose, but may give you a more visually appealing way to display your data to make your point. You can also add trend lines to this type, but they may be less illustrative. I like to think of this kind of chart as a chariot race. Whose horse is in the lead? How many races or sales has my gladiator won?

How about the pie chart? Everyone loves pie! This kind of chart is good for representing how 100% of something is distributed amongst several areas. For instance, I might use a pie chart if I were Emperor of Rome and wanted to chart the different ways my gladiators have met their end in the arena. It might look something like this: out of all or 100% of the deceased gladiators, 25% were killed with spears, 35% died by the sword, and the other 40% were mauled by lions.

Besides column and pie charts, the next most popular has to be the line chart. When you’re trying to find trends over a period of time, the line chart is your tool. Now, let’s say you were Pandora and you just couldn’t control yourself. You might want to chart the daily effects of the death toll you have unleashed upon mankind when you took that little sneak peek. Just remember: Zeus likes to see the effects of his work in a neat little package when getting back at Prometheus.

Just to mix it up a bit, let’s say you wanted to show two things of vastly different sizes on the same chart but still have it be meaningful to your audience. If you were Caesar, you may have wanted to chart the wealth of the Senators to the overall wealth of each province they represented in the Roman Empire. You would have found that when plotting the personal wealth data with the province on the same column chart, the difference was so great that you couldn’t even see the columns for the Senators. Not to worry — there is a nice little technique for combining two types of charts like a column and a line, and it involves a secondary axis. I’ve included the instructions on how to do that in one of this week’s instructional tips in the blog posting for this podcast, located at www.productiveUpodcast.com.

This week’s tips include:

For Microsoft Excel:

• Seeing is Believing – Creating Charts
• Chartin’, Movin’, Re-usin’ – Breath New Life Into Old Chart
• Ring Around the Rows-ey – Switching from X to Y
• Second Best Is Not Bad – Adjusting Chart Layouts
• Pencil Me In – Working with Charts That Have Dates
• Hide and Seek and Data Gaps – Dealing with Missing Info

For Microsoft Access and Word:

• Not This Time Access – Putting Charts in Reports
• Looks are Everything – Adding Charts to Word

This week’s featured eTraining class: Advanced Charting in Microsoft Excel.
Until next week…

The Gods may be crazy but you don’t have to be when it comes to Excel. So Stop on by and expand your knowledge at www.productiveUpodcast.com.

See you next week.

Acquiring the best data set possible can be a very elusive agent. When it comes to finding out what is needed to select the best assets, identify problems, or just to view the results of a campaign, I look at it from the approach of an intelligence agency.

Read any book on espionage or spy recruitment and you will find there is a need to identify the most tell-tale identifier for the job. This requires deciding what questions need to be answered and how to collect the data so it provides the correct facts needed for analysis.

So let’s approach this data analysis thing CIA style. The first thing we need to do is identify the target. I’m going to choose one right out of the news. Let’s consider the future Alien Embassy to be built in the Central Asian country of Kazakhstan (no joke — click here to see for yourself.)

Now, we probably want to collect demographics so we can recruit suitable spies to help us track the progress of their efforts. Who are the ideal targets? In order to decide, we need to aggregate information about the population, perhaps similar to what I found collected in CIA World Fact Book (yes,  these things are real, I couldn’t possibly make it up). Demographics such as: 69% of the population of Kazakhstan is between the ages of 16-64, with slightly more males than females, and 95% of the country speaks Russian as its primary language.

So what kind of tools could we have possibly used if we were the CIA? Well, whether you’re recruiting spies or not, Excel can be an excellent analysis tool.

How could we have laid out our data so we can take advantage of an application such as Excel?

Let’s look at the setup of the raw data first. Knowing that we want to group out results by age, gender, and language to find the best targets, we would need to have a column to store each piece of info for each individual.

For Example:

• Column A: might contain the potential spy’s name
• Column B: their address
• Column C: their birth date
• Column D: their current age
• Column E: their gender
• Column F: their spoken language

Now that we have laid out the structure, we can start collecting information from the field for analysis.

Once the data has been collected, we can either use the very convoluted method of writing formulas to summarize, or we can just use a Pivot Table.

Pivot Tables tend to be much faster and are easy to rearrange as more questions during analysis present themselves.

The main key for a Pivot Table to be effective is to lay out your data as rows of consecutive records. Like having the data of one potential spy contained in one row and the next candidate in the next row and so on.

Now that you have your data layed out properly, you can analyze the different groups in alternate ways just by dragging and dropping them. Here are the links for:

• Creating a Pivot Table
• Filtering a Pivot Table
• and Formatting Numbers in a Pivot Table

to help instruct you on manipulating your assets.

So who do you feel would make the best spy for the Alien Embassy project? Please post your comment below.

And if you would like to harness the “Power of the Pivot” but this podcast just didn’t cover the breadth of your analysis needs, then take this week’s featured class titled “Summarizing Data“. A schedule can be found by following the link to the class schedule in this article.

So until next time, guide your assets well and send your ambassadors reaching for the stars.

Music courtesy of Clayton & Fulcrum

Before you begin type in this data to a new spreadsheet:

1. To determine the difference between two dates, click in cell A10, type the equal (=) sign, click in cell A7 (cell containing the ending date), type the subtraction symbol(hyphen), and then click in cell A2 (cell containing the starting date).

2. Press Enter on the keyboard.

3. To view the result as a number, select the cell.

4. On the Format menu, click Cells (Excel 2007: Home tab, click on Format drop-down in the Cells group, and choose Format Cells).

5. Click the Number tab in the Format Cells dialog box.

6. Choose Number in the Category list box, change the Decimal places to zero (0), and then click OK.

7. To determine the number of days between the current date and a future date, click in cell A11, begin by typing the equal (=) sign, type D1 (cell containing the future date), and type the minus (-) sign;

8. On the Insert menu, click Function (Excel 2007: Formulas tab, click Insert Function).

9. Choose the Date & Time category from the “Or select a category” drop-down arrow.

10. Select the Today function in the Select a function list box and then click OK.

11. Click OK on the Function Argument dialog box to complete the formula.

12. Repeat steps three (3) through six (6) to format the result as a number.

Here is a recent letter from him on the topic of dates in Excel:

I do not usually talk specifically about Excel, but when I heard this issue was all about dates and calculating dates, I thought you might need some basics.  You need to know three things about Excel dates, and then you will know it all.

One – Excel believes time started on January 1^st in the year 1900.  No matter what the philosophers and scientists say, what we can see from historical evidence, and just plain common sense, Excel will not change its mind.  Day one is 1/1/1900.

Two – A day equals one.

Three – Hours are fractions of days, so in Excel terms, noon is .5 and 8:00 AM is .33.

Excel has been keeping track of all the days since 1/1/1900 using serial numbers.  Today’s serial number is greater than 39000.  Excel uses these serial numbers to do math with dates.  Excel sees a date not as we do, but as its serial number.  It generally formats them conveniently for us, but occasionally, you will need to change the format of dates to see the answer you expect.

Let me give you an example:  subtract Christmas (12/25/2008) from New Year’s Day (1/1/2009) and you will get 7 days.  Excel might display that as 1/7/1900.  It is the same thing, just formatted differently.

So remember that time started on January 1^st, 1900, Excel counts every day as one, hours are fractions of days, and it all adds up to NOW().

Now, I know we have all heard the expression: “You’re only as old as you feel”. In my house I feel like I’m 90 – I’ve got a 6 year old that is going on 16 – and a wife who is perpetually 25.

Even on your oldest feeling day, you can figure out you own chronological age, but spreadsheets have a real problem with this. Especially when you’re trying to calculate ages or other date and time related math.

The topic of dates is way too large to cover in any one podcast, so we will return to the subject from time to time.

In future editions of the “Data Analysis” category of podcasts we will also be featuring tips on statistics, charts, and better data management.

Listen to the podcast now: