Some business decisions can be difficult, especially when it comes to expensive choices like data modernization across your department or your entire organization. Projects like these can be daunting – especially given that sharing data across your company touches everyone and the associated costs can be enormous. I wanted to share some practical stories and nimble solutions that can save you and your organization a lot of headaches – and dollars.
The Problem With Progress
Most legacy business intelligence (B.I.) platforms are slow and cumbersome which can raise data anxiety and daily frustration. In addition, many companies tend to neglect their older systems – believing they will eventually replace them – instead of evolving into newer technologies. However, as time goes on, modern analytics initiatives get shifted down the priority ladder and companies remain reliant on the barely functioning, headache-inducing, never updated BI platforms.
And yet, one of the single biggest -- and most costly -- mistakes a company can make is throwing out its legacy B.I. systems. There are two main reasons to keep legacy reporting systems in tact:
1 - Modern analytics tools will not provide pixel perfect reporting solutions that can meet the complex needs of finance and operations. All of the newer tools like Tableau, Looker or Power BI have implemented basic frameworks and forms that will meet your informational needs but lack the abilities that were baked into older tools like SAP Business Objects, IBM Cognos or MicroStrategy.
2 - Knowledge is power. Chances are your existing legacy BI system is crammed full of unimaginable reporting solutions that have been created over the past decade or longer. The goal of immediately replacing 5,000 reports into a new modern analytics tool may be noble but it's also unrealistic.
One of the best strategies employed in implementing a modern analytics framework is to preserve the historical knowledge locked inside your legacy systems while also providing a completely new, ultra-fast and super flexible self-service data environment to empower your employees with the best tools and technology available. These frameworks provide access to that valuable historical data and erase the headaches of dealing with slow, dinosaur databases.
How Forking Can Help
One very easy and fast way to enable new modern analytics data frameworks is by forking, which allows you to keep your old data inside the old relational databases that are feeding your critical legacy tools -- while also funneling your data into ultra-fast columnar cloud data warehouses like Snowflake, Google Big Query or Vertica. An unknown feature to a lot of information technologists is the concept of a "bulk copy process" that can dump a database table into a file in seconds. This magical feature goes by many names: In the SQL Server world it's called BCP, in Oracle it's called OracleBulkCopy, in Postgres there is a function simply called Copy.
No matter what you call it, by using a bulk copy program, you can transfer your legacy relational data and easily start up a modern analytics initiative with new tools in a fraction of the time and cost.
Viva Metadata
Finally, I'm dating myself but back in the 1990's Cognos introduce a tool called PowerPlay which was able to take the metadata (data about data) and spin a small data cube that was columnar, to simply display a drag and drop reporting environment exposing all of a company’s key information like customers, sales and product hierarchies. I remember doing product demo's and (as a nerd) feeling the excitement in a room when people were able to easily visualize their data. For so many, that was their first real exposure to what was possible from their data.
A lot has happened over the past 2 decades. Data volumes exploded and those awesome little PowerCubes weren't designed for or capable of handling the vast amounts of information we started pouring into our business systems. The good news is our new modern analytics databases have an infinite ability to scale and new associated visualization tools are able to once again put the power of data back in the proper hands.
About Us:
Mondo Analytics is a boutique consulting firm in San Rafael, California. For help creating frameworks to unlock the power of your legacy data – while implementing new modern analytics tools, contact me @ john@mondoanalytics.com