Database vs. spreadsheet is no contest
From the October 16, 1998 print edition
by Brad Patten
I got in a fight the other day over a spreadsheet.
This wasn't one of those fun "my computer is better than yours"
fights that I often have with my oldest sister, the Macintosh person.
This was a serious argument over when to use a spreadsheet vs. when to use a
Fortunately, we were able to settle our dispute before we were invited on to
The Jerry Springer Show.
I admit I can get "passionate" when it comes to the Spreadsheet vs.
Database debate. It's not that I totally hate spreadsheets. I just think most
business people use them too often and too much as a crutch to avoid learning
the more complicated but sophisticated features of databases.
I didn't always feel this way.
A dozen years ago, when I was just getting into personal computers and Lotus
1-2-3 for DOS was the Killer App igniting the PC revolution, I had a bad case of
Spreadsheet Syndrome. You couldn't have convinced me I needed another program,
even a word processor.
Then about 1988 or so, I discovered a database manager. It was like walking
into shop class with an old hand saw and seeing someone using a motorized
radial-arm saw for the first time. Now this was a tool. It was fast. It could
handle huge amounts of material. It could make complex, precise cuts in seconds.
You could make just about anything you wanted with one of these.
Since then, spreadsheets have become much more sophisticated tools. But they
still fall far short of databases. Here's why a database is still a better tool
- Easier to share. You can have two or more people editing a database
at the same time. But spreadsheet users have to wait until nobody is using a
file before it's free for them to use.
- Better security and integrity. Along with the ability to better
share information is the ability to better secure it. You can protect
private information from wandering mice better with a database. You can also
protect users from their own mistakes, like adding up the wrong column or
forgetting to include sales tax in the price quote.
- More efficient. Databases are relational, allowing you to link
related tables to minimize duplication. Here's an example. Say you have a
mailing list that includes 20 real estate agents at the same office. All of
them have the same company name, address, phone number and fax. In a
spreadsheet, you would have to duplicate that information 20 times for each
agent. In a relational database manager, you enter that information once,
then link or "relate" it to each of the 20 agents.
- Better reporting. With a database, you can format the same data
many ways in reports. Here's an example. For monthly sales reports, you want
to look at sales by person, by region and by product. In a spreadsheet, you
basically have to enter or copy that information three times in three
spreadsheets to get the desired reports. In a database, you enter it once
and then use the reporting features to compile the data in the three formats
with a click of the button.
- Greater capacity. Databases also have capacity to hold much greater
numbers of records (into the billions).
- Easier to maintain. What makes spreadsheets so simple to use -- the
fact that the programming and the formatting are together on one page with
the data -- also makes them hard to maintain. Every place I go where people
use spreadsheets for database activities, such as list management and
process tracking, I spend time fixing spreadsheets because someone messed up
the formatting while changing the data. You don't have that problem with
- Less duplication. There are two kinds of duplication with
spreadsheets. First, there is file duplication. Because you can't share the
price-quote spreadsheet, everyone makes a personal copy. Second is
duplication of source data. Nine of 10 spreadsheets duplicate information
already housed in a database, such as an accounting program.
In short, databases require a greater investment in training. But the return
is greater than any spreadsheet.
So when should you use spreadsheets? I like them for two circumstances:
- When compiling information from diverse sources. For example, you want to
compare your company's financial ratios with those of industry standards
supplied by your accountant.
- When you are doing a one-time analysis. There isn't a faster tool for
building a simple snapshot that you might not use again.
Of course, you're free to use spreadsheets however you want. But if you want
to debate the merits of spreadsheets vs. databases, I'll see you on Springer.