Self-service BI and analytics tools excite users with easy ways to get data-driven answers to business questions. On the other hand, IT-led enterprise systems have become a hallmark of business intelligence because they solve important issues like data quality, consistency and improved governance.
Today’s self-service BI and analytics tools are easier than ever to use. Interactive data visualization has given more users power and control to interact with data. And those who’ve experienced the freedom of exploring data on their own and constantly asking new questions often see BI enterprise solutions as restrictive and inflexible.
So is there a way to bring self-service BI and IT-led enterprise solutions together? TDWI offers seven ideas your organization should consider as it strategizes to meld user productivity and enterprise BI governance into a peaceful coexistence.
Free Strategy Tips From TDWI
Self-service BI and IT governance – sometimes the two seem at odds. Can they coexist peacefully? Live happily ever after? TDWI thinks so. Download this free report and check out their suggestions for rethinking enterprise BI to fit a self-service world.
Tip 1: Calibrate the role of IT to fit self-service BI requirements
In traditional enterprise BI environments, most users consume the data, applications and visualizations that IT produces. The self-service trend requires business and IT leadership to be more flexible and calibrate the amount of IT involvement to fit what users are trying to do. The main objective for IT should be to adopt an enabler role and help users achieve their goals by guiding them to the right data, advising how they can get the most out of BI tools and helping to scale up applications.
Tip 2: Update governance to embrace self-service BI and analytics
Users seeking new sources for data discovery and analytics don’t like waiting for new data to be incorporated into the existing data warehouse. Big data lakes and cloud-based data sources are growing in part because users need access to a wider variety of data. Unfortunately, these sources are often not adequately governed, much less vetted for quality and consistency. Organizations will need to examine current BI governance strategies and make sure they account for the expanding data environment.
Tip 3: Revise the semantic layer to support self-service interactive reporting
One of the advantages of mature enterprise BI and data warehouse architectures is having a coherent and up-to-date semantic layer, from which self-service BI and analytics can also benefit. However, diverse and distributed self-service technology can make development and maintenance of a semantic layer challenging and complex. Organizations should evaluate their existing enterprise BI and data warehousing semantic layer to ensure it can extend to ad hoc, self-service BI and analytics use cases.
Tip 4: Balance enterprise BI standardization with user agility
When decentralized and not well coordinated, each self-service technology implementation can become its own data silo. Organizations struggle with balancing user agility and BI standardization. TDWI recommends three steps:
- Provide managed self-service that offers guidance.
- Create self-service applications that offer standard choices within them.
- Aim for less obtrusive IT management and governance.
Tip 5: Introduce self-service data prep carefully
Data preparation is a key concern for those trying to balance BI governance and self-service capabilities for users. To avoid the pitfalls of self-service data preparation, TDWI recommends that organizations centrally monitor metadata, integrate data prep with governance and aim for higher levels of repeatability using automation and web-based administration technologies.
Tip 6: Develop an open architecture to match workloads with technologies
Open source and cloud computing technologies require organizations to take a fresh look at their enterprise BI and data warehousing architectures. It may be time for a hybrid approach. Not all use cases and workloads will need the rigorous governance and structure of a traditional single architecture for enterprise BI and data warehousing. The strategy must have flexibility and openness to take advantage of the potential of new technologies and methods.
Tip 7: Refresh training to fit diverse user needs
Even though BI and analytics tools are becoming easier to use, it is not necessarily straightforward to understand and apply BI and analytics techniques, particularly for nontechnical users. Among other strategies, this report recommends mentoring through BI teams and encourages collaboration and tip sharing to help users learn from each other.
The self-service genie is not going back in the bottle. Self-service BI and analytics are here to stay. David Stodder Senior Director of Research for BI TDWI
- Article Understanding capital requirements in light of Basel IVMany financial firms are already using a popular 2012 PIT-ness methodology for internal ratings-based models. This article examines eight ways the industry is successfully using the methodology – and why this approach can bring synergies for banks, value for regulators, and major competitive advantages.
- Article Respond, recover and reimagineDisruptions to our lives happen regularly, though most are not as far-reaching as the COVID-19 pandemic. Whatever their nature, it’s helpful to have a plan for how to exit disruption still on your feet and in the game. Learn about the three-phase approach SAS recommends for mitigating widespread disturbances.
- Article Detect and prevent banking application fraudCredit fraud often starts with a falsified application. That’s why it’s important to use analytics starting at the entrance point. Learn how analytics and machine learning can detect fraud at the point of application by recognizing the biggest challenge – synthetic identities.
- Article What do drones, AI and proactive policing have in common?Law enforcement and public safety agencies must wrangle diverse data sets – such as data from drones – in their proactive policing operations. To be most effective, they need modern tools that support AI techniques like machine learning, computer vision and natural language processing.