Whether you call it ‘immediate gratification’ or management embracing the concept of ‘just in time’, few can argue that when it comes to information in the age of technology, users know what they want and they want it NOW! Who can blame them? For the past couple of years we have been hearing about big-data, machine-learning, IoT devices everywhere, smart phones, smart clouds and smart homes – technology is everywhere and whether it is just hype or the tech industry can actually deliver, they have set the expectation that a world of data is just a finger tap or voice-command away.
Expectations are one thing, but delivering on reality can be a bit more challenging. With a seemingly-infinite number of sources available, putting real-time analytics and meaningful information insights into the hands of users still comes down to the old-fashioned supply chain. Ok, well maybe not the old-fashioned supply chain, but rather its modern equivalent – the data supply chain. Raw data is processed, aggregated, transformed, filtered and transported as it progresses from its various sources on its way to the users that will consume it.
The most important part of the data supply chain is the process of data refinement that takes place as the data is transformed – turning rough nuggets of data into polished gems of information for users to appreciate. When it comes to real-time analytics and information insights, success is measured by speed of transformation and quality produced. Not only do users want the information now, but they also expect it to be meaningful, accurate and trustworthy. So, as an IT professional tasked with exceeding your users’ expectations, what can you do about this?
The key to enabling real-time insights is moving data refinement as early as possible in the data supply chain. All data goes through a refinement process, whether by system processing, as it is presented to users or as humans interpret what they are seeing and decide what to use and what to ignore. As refinement takes place, the overall data-set gets smaller (as meaningless context and extraneous bits are eliminated) and the quality of the data improves (through validation, consolidation, filling gaps and resolving conflicts). If you wait until data is presented to the user to refine it, there is still a big task ahead (the data is perceived to be slow and poor quality). If instead, you refine, aggregate, and improve data quality soon after it is created or discovered, there is less data to maintain in a database and the task of getting it the last mile to the end user is much simpler.
A practical example of this is IT Asset data. Pieces of data about IT assets are collected from a large number of sources in your IT environment. There is a lot of duplication, gaps, conflicts, an obsolescence issues in the raw data. If you store everything in your service management system and then try and fix the problems as its being presented to the user (or manually after the user gets it from the system), you have to go through the refinement process multiple times (with each request) and the data quality is perceived to be poor. Instead, if you refine, validate, reconcile and clean-up the data as it is being produced, you store less data and higher quality means it is ready to consume without a lot of additional work.
Management and other business users have an expectation that the information they need will be easily accessible to them. IT professionals know that there is a lot of work needed to make that happen. By focusing on establishing quality early in the data supply chain, we have an opportunity to make the job of data management easier (with less data to maintain) and give users the experience they are looking for.
Blazent is an industry leader in data quality management solutions – providing the capabilities that IT professionals need to aggregate, validate, reconcile and verify source data as it’s imported to improve the overall quality and trustworthiness of the data being managed. To learn more about how Blazent can help your company, visit blazent.com