Problem solve Get help with specific problems with your technologies, process and projects.

Good customer service will not a warehouse give

There are three options for designing the data warehouse: normalized data structures, star schemas, or snowflake schemas.

"Things are more like they are now than ever were before." --Dwight Eisenhower

Based on the amount of email I received on the last article, I was not very clear on my conclusions. Therefore, let me try to set the record straight (and be more precise!).

To me, a data warehouse is a concept that can be implemented as either a single monolithic database or a distributed database environment (whether it is a true distributed database or a strictly controlled, dimensioned set of data marts). I believe there are three options for designing the data warehouse: normalized data structures, star schemas, or snowflake schemas. I prefer the star schema because it is easiest for the end-user to understand, the duplication of data is not that great compared to the overall size of the data warehouse, and it is has easy to accommodate change as a normalized structure (although this is a point of great discussion!). To me, a data mart is a subset of the data warehouse data provided to the end-users. This could be for performance reasons or using a controlled data mart environment. This should be designed to work best for whatever tool is going to be used. For example, MicroStrategy's products work best with a snowflake environment so if the users are going to be accessing the data mart via MicroStragegy's products, then build the snowflake. However, if multiple tools are going to be used against a single data mart and they all work best with different environments, then I would choose the star as my design. Again, there can be great debates concerning this.

I certainly hope that this is clearer. I am sorry for any confusion my article may have given you. Now, on to today's topic...

Customer service is getting a lot of lip service in the marketplace. The current buzzword is CRM (or derivations like eCRM, etc.), which stands for Customer Relationship Management. The issue at hand deals with whether you as an organization are really interested in CRM or are you just using the buzzwords? United Airlines is a great example. I'm one of their most frequent flyers (they call us "1K," which is quite a misnomer)--I travel more than 100,000 miles a year. You would think that they would do what they can to help us in our travels. But alas, they don't always do that. Because of my schedule, I don't always get to the airport one hour before take-off, so I am rushing to get on the plane and I often have to check my bag at the gate. If I have something breakable in my bag, I ask if they will place the bag in the holding area, like they do strollers, and return it to me as I get off the plane. I am not asking for something that isn't done at all, but something that can be done. The last five times I made the request, it occurred. I was very impressed with the customer service. This past Sunday (the 11th of March), I was at the airport early and boarded when I was supposed to. I got settled in my seat and they called me to move to first class. That was nice of them and I was pleased. However, they did it while still boarding the rest of the plane. I felt like salmon swimming upstream. Well, I decided to move to the side and wait to it was clear before moving forward. What happened? There was no room for my luggage. At this point, I asked for them to put it in the front like the strollers and they said they could not. I explained that others do but the customer service representative said that they don't. I repeated that they have done it before and that there were breakable items in my bag. He called me a liar (great customer service skills, huh?) and that no one would do that. I asked him if United would be responsible for broken things in my suitcase and he responded that if it were clothes then yes. Fine, check it. I took his name and will write a letter to United.

Is this a good example on how to keep your best customers? I don't think so. United has paid a lot of money to create (in theory and rhetoric) a customer service environment that gives every employee the right to take care of customer problems. But, do they fulfill that promise? They certainly try, but do not always succeed.

Why did I tell this elaborate (but factual) story? Because as organizations embark on CRM systems, they need to understand that the CRM system alone will not help with customer relationships. It can show us what customers want or about trends in the marketplace, but there has to be a change in the fundamental way that we are doing business to do effective CRM. We can build the greatest data warehouse (required, IMO, before the CRM) and feed a CRM system, but unless organizations change the way they react to customers, they are not providing the benefits that may be expected.

Remember that you can do a hundred things positive, but a single bad experience can wipe them all away. In all my nearly 1,500 flights (not miles), I have a hard time remembering but a few good things. But man I can remember the bad experiences like they were yesterday (and one was!).

A few years ago, data warehousing was getting a bad rap because we were not delivering on the promises. Is CRM the next evolution that gets a black eye? Let's hope not.


About the Author

Chuck Kelley is president and founder of Excellence In Data, Inc. and an internationally known expert in database technology. He has more than 20 years of experience in designing and implementing operational/production systems and data warehouses. Kelley has worked in some facet of the design and implementation phase of more than 35 data warehouses and data marts. He also teaches seminars, co-authored a book with W. H. Inmon on data warehousing and has been published in many trade magazines on database technology, data warehousing and enterprise data strategies. Please feel free to email him at with comments (negative or positive) about this column or ideas for future columns.


For More Information


Dig Deeper on Oracle data warehousing

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.