Apr 012013


I am hearing more and more that people don’t know what’s in their data, what to ask of their data, especially how to get at what they need. There are both pleasant and ugly surprises lurking in the data. Here’s a story:


Sitting at Whole Foods over a casual lunch with a friend whom we’ll call Ted, I pasted his business data into an Excel pivot table and he gasped. In 90 seconds Ted had clear and unexpected evidence that a product line was doing far better than he thought.


His company provided the data buried in a few canned reports but, Ted thought the business intelligence was terrible because it doesn’t give useful reports like “product line month-over-month sales.” But all the information was in there. It just wasn’t on a silver platter.

Actually, he had something better than clean reports. He had a glorious data dump.

It was frikken easy to extract the information he wanted because his company was providing far more than he really needed in a bunch of rows and tables.

Is Ted stupid or lazy? No. Hardly. He just didn’t know what he didn’t know. He’s got Excel on his computer but just wasn’t aware of the power of the world’s #1 business intelligence tool.


Too many people are like Ted. The challenge is that they are good at a skill (baking, customer service, website development, graphic design, teaching, sales, etc.). They work their way up and eventually they’re leading people, responsible for a department and overwhelmed with data on spreadsheets that they don’t understand. We can only guess at the consequences.

Who’s lost a job, quit their job, faced a lawsuit, never got paid the right bonus, imagined a problem was bigger than it really was, missed golden opportunities or, spent hours sifting through data only to find nothing interesting? The outcomes may have been different if someone knew how to efficiently get at the right data.

The information is in the data dump. The challenge is in getting at it.


Why do I stress “get at” the data? Often, the analysis isn’t a big deal. For small businesses, and departments within large companies, we don’t a convo about “big data.” No. We ask: How are sales trending? How is our productivity? They’re either going up, down or remaining flat. Period. Analysis gets hard when you want to forecast or if you want to get super-specific or of you do have crazy amounts of granular data to piece together. But most of us don’t have such problems. (Remember Ted and the 90 seconds it took to get his company’s provided data. 90 seconds, make a pivot table and WHAM! Suddenly he’s got new possibilities.)

The problem is in getting at what you want to see. Example:

  • Data Dump A: transactions listed by transaction ID, date and time.
  • Data Dump B: Customer Service Reps and their transactions.
  • How do you match up these reports to figure out who are the reps most experienced with your top 3 accounts? Who are the top 2 reps who make the least mistakes processing orders for those 3 accounts?

It’s all quite easy if-and-only-if a person is aware of the tools. Alas, managers just give up. They guess. They scream at the whole department about mistakes. They look like incompetent boobs in front if their executive bosses. They work long hours cutting-&-pasting their way to needed answers that feed critical decisions.

That’s the “get at it” part. The analysis isn’t hard. If we can get at or re-arrange the existing data for the angle we need to see, we’ll know who’s processed 3 transactions at a 100% error rate and who processed 50 orders with a 1% error rate. That’s easy. Cutting-&-pasting ain’t easy. It’s also unnecessary. All so very unnecessary.

Take special note: Once you accept data dumps and can get at what you need,
it’s possible to get at any ad hoc piece of information without relying on canned reports.

Extending our example: since we have this data dump, we can dig out “is there something cyclical that causes no orders for a long time and then orders come flooding in without warning?” Maybe our people aren’t ever ready for the flood and that’s really the root of all the errors. I’ve exposed a lot of things like this—only because I had access to a data dump and could “get at it.”

In one instance, there was a national exam that’s given only 3x/year. Before a certain product was launched, no one expected insane levels of orders 1 month prior to that exam’s dates. Imagine what that does to customer service, the sales team, the internet servers, the warehouse team, daily processes, etc.

The right info from the data dump tossed into a line graph showed distinct, predictable waves of activity. We had to restructure a lot of things after that revelation.


  1. Data Management must become part of our vocabulary. We’re all data managers. Even if we just want to keep up with our friends’ birth-dates  That’s not a memory issue, that’s a data management issue.
  2. Make deliberate effort to learn the business intelligence tools that are available. For most of us, Excel is perfectly fine. For some businesses, Excel is the only affordable tool.
  3. Learn to work with data dumps. Maybe even prefer data dumps over canned reports.
  4. Some managers don’t need to handle data at this level. It’s not a smart use of their time. They need a data-person in their back pocket; someone trained to dig, probe, pick and get at it.

That’s it for today. Remember: we’re all data stewards. Get out there and keep this world’s data clean and trustworthy.

 Matrix cascading numbers image credit: Madtomatoe

Share Button

Oz du Soleil

I am Oz du Soleil. Data Mercenary and Excel Trainer. My courses have been described as informative, fun and they get people to relax about using Excel. I'm based in Chicago, a veteran of the U.S. Navy, and have a passion for custom-made hats, good bourbon, and spicy food.



Get every new post delivered to your Inbox

Join other followers: