Peer to Peer Magazine

34 be confident that this part of the project is much easier. You also can be more confident in building automated processes knowing they won't back you into being tied to a source system. The Bryan Cave Leighton Paisner upgrade project was accomplished in five months instead of 12. What does that really mean? It's great you can say you got it done quickly. It's great the people involved can move on to other projects. It's great to tell whoever's listening "look how brilliant we are." But what about cost savings? Doing some simple math on consulting costs: • two consultants • billing $200 per hour • working on your project 40 hours total per week • under that scenario you have saved around $250,000 Now imagine applying that kind of saving to virtually any big project by simply saving time. Money and time saved now can be invested in innovation, more staff, or any idea you may have that requires time and money. Is there any idea that doesn't? Anybody who's done a project of this sort will realize the numbers used in this mini example are conservative. If nothing else gets you excited this should. Your boss never complains that you come in under budget, right? Other Benefits of a Master Data Database Building and maintaining a data lake environment may be easier than building a data warehouse and if done a certain way may save you money, but it still requires resources. An endeavor of this type takes a lot of effort so you probably want more than one thing to show for it. Here are a few additional benefits discussed in more detail: New Data Driven Systems – For something like traditional Enterprise Search in the legal industry, you need information on clients, matters, documents, people, time entries, and security. Where does all of that information come from? Without a master database it comes from five-plus databases. With a master database it comes from one source. Your startup time is quicker, your data source never needs to change in the future, you're less tied to an Enterprise Search vendor if things aren't working out, and the project gets done more quickly. Performance – Something like Enterprise Search, without a data lake, will get its data from your most used databases. Document management, time entry, finance-accounting, and Enterprise Search likely will not be the only systems wanting this information. Putting a data lake in your architecture means one controlled system copying data to the central database. Then those other systems point to your master data. If they overload your master data database server everybody is still happily working on documents, entering their time, and reporting revenue. Master Data Management Strate – An MDM strate normally includes the desire to have one source of truth for data and to have that data cleaned and kept clean. Getting your data to a central place allows you to see the whole picture more readily. Once you're using it as the prime source for downstream systems you're forced to clean it, which means cleaning it at the source, which drives you to implement policies to keep it clean. Innovation – Innovation means a lot of things, but more often than not it involves your information. Having all your information in one place and more easily seeing how it relates allows you to more quickly implement new innovations on the market. More importantly, it can help you more quickly carry out an idea that's not on the market and give you a leg up. Centralization of processes – By centralizing your information you'll naturally centralize processes using that information. Few law firms suffer from an abundance of technical employees. This translates many times into getting work done but having little or no time left over for documentation. Centralizing helps you more quickly troubleshoot issues with a process. You also can more easily build monitoring systems to be proactive about noticing issues before they're reported by people using the information. This may be the least exciting benefit to many, but innovation means time. More people time means more innovation time. This All Sounds Great. How Do I Get There? It's virtually impossible to have a master data database with as many as 20 key data sets in two weeks. It is possible to have a database and the beginnings of your first data set in as little as two weeks. How do you do that? Building and maintaining a data lake environment may be easier than building a data warehouse and if done a certain way may save you money, but it still requires resources.

• Cover