Is your company not a fan of big data? Here are four things you can do to change their opinions.
In a business intelligence survey conducted by the German-based Business Application Research Center, 53% of survey respondents said that there was inadequate analytical know-how in their companies, and 56% acknowledged that their companies were not finding compelling business cases for big data.
This lack of big data business savvy continues to haunt companies, yet big data promoters, and IT must find ways to make inroads for big data adoption in mission-critical applications. Why? Otherwise, they risk falling behind
3 reasons why your company dislikes big data
Below are three common impediments to big data adoption.
1. It’s complicated
Big data flies into the enterprise from everywhere. It must be painfully sorted out, cleaned, categorized, and aggregated with other types of data to give corporate decision makers enough information to make informed decisions. This data preparation and integration, coupled with iterative testing of queries and algorithms, takes much more time than developing the more traditional third- and fourth-generation reports off fixed, transactional data that companies are more familiar with and used to generating.
2. Corporate decision makers don’t understand (or want to understand) big data analytics reports
For the most part, big data reporting is shepherded by highly cerebral data scientists who can be more in love with the data (and working it) than tuned into business results. This is a fundamental disconnect because users, especially executive users, want reporting that delivers actionable business output.
3. Big data analytics reports can be hard to read
If data is presented in tabular formats (as many are), a majority of business executives won’t wade through it. They prefer to read reports that are business-incisive and can be digested in an eyeshot.
So what can you do to break down these common big data barriers?
1. Find data integration tools that can automate much of your data ingestion and integration processes
This reduces IT’s need to custom-code APIs and interfaces.
2. Build a precise business case
At the top level of any big data analytic, you want to directly attack a business pain point. For example, this could be falling revenues, predicting what the next big market or product will be, or improving service response, and so on. Whatever the pain point is, identify a precise business case so that the algorithms and queries your team develops can address it.
Also, don’t forget to pull the plug on any big data analytics project where the end user (especially management) doesn’t see the value.
3. Visualize your data
Cast your data presentation into a bar chart, pie chart, map, or any other visual presentation where management can quickly gain an eyeshot understanding of your findings.
Make every effort to only use tabular data formats when the user opts to drill down into the depths of the data after seeing a visualized (and pictorialized) summary. With great data visualizations, you can build user (and C-level) confidence in your big data analytics.
4. Measure for data actualization
Uptime and mean time to response are always valuable IT metrics, but the more important metrics in big data and analytics adoption include questions such as:
- How many users continue to use a report six months after it was issued?
- How many requests are you getting to further enhance a reporting product (this suggests user enthusiasm)?
- How many reports result in measurable and actionable steps that management is taking to improve business performance?