This article is adapted from a presentation given by CRSA to Emerging Professionals in the Utah Chapter of the Society for Marketing Professional Services. It explores the basic concepts of storing data from the perspective of marketing and proposal generation needs.
Since databases are, by definition, rigid frameworks, its important to remember up front that the organic world is not a one-size-fits all. Across firms and industries, we all have different data to begin with. We all (yes, ALL) have bad data in some form or another. And we all have different requirements at different times. The outline below attempts to take a base-line approach to account for these inherent differences.
A lot of people think a database is just adding order to random data. This is only partially true…
The heart of a database is in connections. It links one discreet fact with another, then another, creating a web of interrelated points.
A simple list stores sequential data relating to a single topic. Multiple lists can cover multiple topics. A table, or spreadsheet, allows you to store multiple topics within the same file or framework. But what if multiple tables and lists could talk to each other, like an Excel spreadsheet with many dimensions? This begins to approximate what a database is–multiple tables connected to each other. When you take this and add a user interface, then plug the system into the internet, you have a modern DBMS (Database Management System). These are the backbone for storing most businesses’ corporate knowledge.
In the AEC world, DMBS’ are quickly gaining prominence as they become more affordable, easier to use and implement, and more necessary due to the amount of data being generated.
A database is only as good as its data. And thus it is crucial that incoming data obey certain rules and standards. Time spent up front on data entry pays exponential dividends when it comes to data retrieval.
A basic rule of thumb when adding data to a database is to never store the same piece of information twice. The reason is, if a piece of data is stored twice and it changes, it has to be changed in two places. This doubles (or eventually triples, quadruples…) the effort, and it opens the door to hidden errors if you were to change data in one place and forget to do so in another.
Another core rule of databases is to be absolutely rigid on standards. Make sure you enter similar data in the exact same way, every time. Failure to do so can cause lots of work later on, and it can obscure or hide certain facts when you later go back and try and retrieve data.
“Data” refers to discreet, singular facts or attributes. Incoming data should be broken down and stored in the most granular way possible. This makes searching the data easier, makes updating the data easier, and makes finding and fixing errors easier.
So how do you know if your system is performing well? There are three useful checks. First, gauge how long it is taking you to keep the database up to date. A database is ultimately supposed to increase long term efficiency, and if it fails to do so, reassess the system. Second, a database is supposed to help minimize errors. Make on occasional check on a proposal that relied on a database. If its still has serious errors, it may be worth looking at the source data, or how the data is entered. A third way to analyze your data management is to try and teach the system to a new employee. Theoretically, databases should be simple and obey strict rules such that they can be transferred among people with no compromise. If this can’t be done, you may be relying too much on personal preferences and unwritten practices.
There are things that the human mind can do that no computer, ever, will be able to match. The mind can make connections that simply cannot be computed. But computers can perform repetitious and large-scale tasks quicker and more accurately than a hundred minds working together. The point is, know what the brain is good at, and know what the mind is good at, and try and assign tasks accordingly.
Databases are only tools to help real world needs. And ultimately, if they aren’t working, they aren’t worth using. There will be some inefficiency getting the system set up–and this is a worthwhile investment. But in the world we live in, some tasks are worth doing whatever way works best.
Databases have immense power. But it is a cold, mechanical power. Databases store data, they don’t think or feel. The ideal is that databases take some of the burden off marketers and designers so that they can add the human element back into materials–the life and passion.