How much data is too much? To begin considering an answer to this question, it’s important to realize it isn’t even the correct question. Instead, we should be asking what data should be used.
When an organization realizes it needs business intelligence and the tools to collect it, the next step should be to create a plan for that BI. Before worrying about what data should be collected, ask yourself what questions need to be answered. For a business, that most likely starts with expenses and revenue. As the plan develops, breaking down where the expenses occur and what products or services generate how much revenue will increase the amount of data that is generated.
The introduction of the Internet of Things (IoT) will generate huge volumes of data on top of what already is being collected. Now the business might even learn what products in what aisles of their stores are sold the quickest, and what weather seems to prompt what kinds of sales right up to the minute.
IBM says that every day, we create 2.5 quintillion bytes of data, enough that 90 percent of the data in the world today has been created in the last two years. This blog is a tiny fraction of that data and while you might not give it a value that matches how many deals your sales executives have made today, it still added to the huge volume that’s available. And that points out what each organization needs to handle every day: How to decide what data is important, and what might not be.
Paralysis by overanalysis is a real problem. There’s always more data to collect. But does it really have anything to do with your business success or failures? That’s where creating a business intelligence and analytics strategy becomes essential to tame all of that data:
- Find a C-level sponsor to back the strategy. Someone has to be the champion in the C-suite to assure it is adopted. This person will make sure the organization follows the vision for the BI and analytics strategy – and will keep the strategy on target with the company’s mission.
- Create common definitions for the KPIs considered most important for the company. Without proper data governance, any BI reports created may be flawed. Creating standards and governance will keep the company out of a lot of trouble later.
- Create a plan for data storage. Can you afford to have your data stored in spreadsheets and databases spread across the organization’s storage with no governance? On-premise, hybrid, public and private cloud data storage are among your choices. Data governance won’t be possible if storage isn’t planned and controlled.
- Find out what business users really need. An end user may need access only to specific parts of the data, while HR staff will access other sections. Some users may only need to view reports, while others will need to create dashboards and visualizations for the data. An analysis of the roles each user plays in the organization will help determine what data each should be able to access and share.
- Choose whether to build or buy a BI and analytics solution. Pitfalls exist for building a solution, including lost time by developers who are picked to create it and the slower time to market than buying an embedded system.
- Even though you feel the organization is swimming in data, start small. Pick just a few KPIs that have high value to test your proof of concept, then start rolling out more reports, dashboards and visualizations as these prove themselves.
Without these basic steps to managing your BI and analytics strategy, you’ll be awash in so much data it no longer has any meaning. Take the time to analyze what data is important to your company.