Building a Unified View of Your Data
We talk a lot of about the amount of data organizations capture and how that data comes in many sizes and formats, from structured to semi-structured and even unstructured. We also know that much of that data isn’t used for decision-making, hidden in silos across the organization. All of this makes it difficult to build a unified data view.
But the challenge with building a unified view is only partly due to data silos.
The Expert, The End User, The Solution Provider
David Loshin, President of Knowledge Integrity, Inc., knows a few things about the analytics industry, working with clients on a range of BI, Big Data, data governance, and other data initiatives. Ron Aerni, ETL Lead at Global Partners, LLP understands well the challenges of disparate datasets and the need to quickly bring together the right data to help his company make good decisions.
Both Loshin and Aerni joined Joe Lichtman, VP Product Management at Attivio and IDG for a discussion on building a unified view of the data landscape to drive agile analytics. Here’s a peek at what they said.
Loshin talked about the popularization of analytics as a core business function. No longer owned and managed entirely by IT, business users are demanding self-service access to raw data to do their analysis on unified data. And they don’t want to wait forever for IT to find that data and transform it.
Delivering this self-service capability isn’t easy. The business has no control over the data lifecycle, and they haven’t a full understanding of what’s even out there. When they do get the data, there’s often no consistency in how a single analyst interprets a dataset, let alone consistency between multiple data analysts.
Loshin said the answer lies in the ability to catalog datasets automatically, capturing and managing the information in a shared environment that can be scanned and search by word, phrase or concept. He also spoke about the different types of semantic interpretation this catalog needs to provide including inherent (the data and its metadata), personal (how an analyst uses it) and community (crowdsourcing the best interpretation of the data).
The End User
Aerni provided a real-world example of the challenges of data collection and interpretation. He spoke about the growth of his company over the past few years, growing from 40 locations (service stations with retail stores) in 2010 to over 300 today.
Global Partners LLP has some BI capabilities in their core business applications, but they were unable to connect data assets easily from different business units due to disconnects in reference data. They want to enhance their BI capabilities but time and cost were key issues that discouraged traditional data warehouse development.
What they looked for was a way to prototype unified data views quickly before ETL, and they did this using Attivio’s Data Catalog.
Areni discussed the creation of virtual data marts for market basket data that enabled them to understand the performance of their locations. He also explained how business users used the virtual data marts and were able to see if data was missing or something wasn’t right, allowing them to make changes quickly, add new sources and bring together the right data for analysis.
Global Partners started with a project but see many ways to incorporate more data sources both inside and outside the company to better correlate performance.
The Solution Provider
An organization only uses 10% of the data it collects. That means they are very likely missing critical insights or using the wrong data to make decisions.
The biggest question most organizations have is how to shift to the agile business intelligence capability they need to be competitive. The shift requires access to all a company’s data, not just as single datasets. It also requires a deep understanding of how that data is related because sometimes the most compelling insights come from the relationships between disparate data sets.
It’s also very much about self-service, providing the business the tools required to search, select and package the data needed for analysis.
This is where the semantic data catalog delivers. Lichtman described its value in exposing and explaining data in business terms, enabling unified data and provisioning data for BI tools.
The Path to Agile Data Analytics through Unified Data
You will need an automated approach to deal with all your data. It’s impossible to handle it manually and be timely and competitive. Consider a semantic data catalog and all that it provides; you will have a better understanding of your organization's data like never before.
And if you need more information on the benefits of a data catalog, check out the on-demand webinar Building a Unified View of Your Data Landscape to Drive Agile Analytics. It’s worth your time.