Summary: As data volumes explode, more and more businesses are realizing that traditional BI practices are less effective. We’re seeing a shift in Business Intelligence–away from an IT-driven process towards a self-service approach. But, while the need for self-service BI is growing, it’s still surrounded by a fair amount of confusion. The concept of “self-service BI” means widely different things to different people. In this article, we explain the basics of self-service, and the pros/cons it delivers to businesses.
Did you know that data volumes are doubling in size every two years? By 2020, the data we create and copy annually will reach 44 zettabytes, or 44 trillion gigabytes.
These numbers are staggering, but…what do they mean from a business perspective? How will this growth in data affect your organization?
This article explains the effect nicely: “Data has always been a key asset for some companies, but over the last three to five years, it has gone from ‘important’ to ‘critical.’ Data, and more important, data analysis, have become a true competitive advantage.”
In other words, your ability to quickly turn data into meaningful management information will directly impact your bottom line.
The problem: The traditional Business Intelligence (BI) process struggles in this area. Businesses are finding that they can’t deliver the data users need, when they need it–a problem that’s creating a shift in the BI world.
How big is this shift? As Gartner explains in this press release, the balance of power in the BI world is shifting from IT to the business.
The data explosion combined with its growing importance is fueling the need for self-service BI. More and more, the ability to put real-time data in the hands of end users becomes a competitive advantage.
But, while the need for self-service BI is growing, it’s still surrounded by a fair amount of confusion. The concept of “self-service BI” means widely different things to different people.
Today, let’s go over the basics. If your organization is considering self-service BI, or if you’d just like to learn more about it, this article will hopefully prove useful. We’ve outlined and answered the basic questions surrounding self-service BI below:
What is self-service BI?
“Self-service BI is reporting and analytics which business users with little IT training can do for themselves without IT intervention or facilitation,” says Douglas Briggs, Director, Business Intelligence at Washington University in St. Louis. “Essentially, if an end user trying to answer a business question can open his or her BI tool, click on fields, filters, and calculations, and create the report that answers that question, they’ve done self-service BI.”
To further expand on this definition, you’ll find there are two key elements to self-service BI:
First, the user must have access to data in a workable format. They shouldn’t need to request data access or format data every time they need a report.
Second, the user must have access to a self-service BI tool (and no, Excel doesn’t count). The self-service BI tool lets the user generate the reports, without bothering the IT department.
What’s the end goal? Put data analytics in the hands of end users–when they need it.
“The goal of self-service BI is to get the analysts and data scientists out of the loop in answering day-to-day analytical questions,” says Barry Parr, Director of Marketing Communications at Chartio. “What you don’t want is a process that involves multiple email conversations with an analyst who’s building complex SQL queries to answer routine questions about your business and send you an Excel spreadsheet with the answers.”
How does it differ from traditional BI?
“In traditional BI, end users often have to specify the kinds of questions they will need answers to in advance, IT builds reports to provide that data and delivers them to the business,” says Briggs. “When the business needs more detail or different data to refine or change their questions, IT adapts the reports or builds new ones.”
Self-service BI eliminates this time-consuming reporting process–placing much of the report creation on the end users.
Does this mean that IT is completely removed from the picture? No! Certain tasks or complex reports may still require IT assistance. But, the majority of the BI process is in the end user’s hands.
In placing these capabilities in the end user’s hands, you find another big difference between the two approaches: Self-service BI opens the door to more data exploration. Rather than asking IT to build a report that answers a specific question, end users have the tools to ask and answer almost any business question they can imagine–whenever they want.
“With self-service BI, business users can answer their own questions and explore the data themselves,” says Parr. “Meanwhile, the analysts can focus on more complex questions or preparing the data so that business users can get more done.”
How will self-service BI improve your business?
While we’ve touched on a few of these points above, self-service BI improves businesses in some key areas:
1. It saves time
As mentioned above, self-service BI effectively cuts out the middle-man–giving end users direct access to the data they need. Transferring this process to the users can turn a reporting process that requires days, into one that requires hours.
“Most businesses see improvements in their efficiency and capability when they can close the question – answer – new question cycle by removing the dependency on IT resources to participate in that service delivery chain,” explains Briggs. “There are often far more business people asking questions than there are IT technicians building reports to answer them, so enabling the business to answer their own questions can result in significant time and resource efficiencies.”
2. It promotes fast decision-making
Data is quickly becoming your most valuable asset. Those who can quickly turn their data into meaningful management information will hold a huge advantage over those who cannot.
“Real time data availability enables the organization to tweak operations immediately, improving efficiencies and customer service,” says Marne Martin, CEO at ServicePower. “It also enables the information to be shared across the enterprise, from sales and marketing to service and manufacturing, ensuring that each can identify trends impacting future strategies.”
3. It delivers personalized results
With traditional BI, multiple users within a department often shared the same report. While not all of the data in the report applied to each user, it was more efficient than creating separate reports for everyone.
With self-service BI, users get personalized results that are more relevant to their specific job functions. They’re not limited to someone else’s vision of what information they should see or how it’s presented.
What are the risks of self-service BI?
So, self-service BI sounds like a no-brainer…right? Be careful. While it does offer some great advantages, it’s not without risk. Like what? Risks arise when:
1. The users aren’t properly trained
You can’t simply give end users a self-service tool and expect them to understand how it works–even if it is “easy to use.” Providing inadequate training creates two problems.
First, users might abandon the tool altogether. If they get frustrated easily, they’ll give up before they have a chance to understand its value.
Secondly, they might use it improperly. Simply turning the technology over to business users without appropriate training increases the chance that they will mismanage data.
2. The data isn’t clean
Clean data lies at the foundation of any successful BI project–both traditional and self-service. Deploying self-service BI software over bad data will lead to a failed project. The IT department must still pull the data together and ensure that it is accurate and organized.
“The risk of self-service BI is misunderstanding of the data,” says Martin. “When configuring the data relationships, IT must ensure that data presented is correct and factual. We must not enable decisions to be made on misinterpretation of the data.”
3. Users lack understanding of the data
Not only must users understand the BI tool, they must understand their business data, and what questions they need to ask. The IT department can’t simply pull the data together and make it available to the users. They must explain where it lives, how it’s organized, etc…
“One of the biggest risks in creating a self-service BI culture is having IT simply pull data together and “throw it over the wall” to the business without proper communication around how the data has been integrated,” says Briggs. “Unless the business understands how they got the data they’re using for their decisions, they risk combining data in ways that support invalid conclusions.”
4. Users have too much control
It’s the problem plaguing many self-service BI implementations: Organizations have a tough time balancing control and business empowerment. If you simply install BI software and roll it out to every user with no governance in place, you’ll create a mess.
Without properly controlling the data, you’ll find data inaccuracies, duplicate data, and a host of other problems. Any self-service BI tool you select must provide a way to control the data and user access, while still giving users the tools they need to create their own reports.
What misconceptions exist about self-service BI?
Despite it’s growing popularity, many myths still surround self-service BI. These myths can lead to implementations that aren’t as successful as they could be, or flat-out fail. Here are a few of the most common myths:
1. Self-service BI eliminates the need for IT
The biggest misconception surrounding self-service BI is the idea that it shifts 100% of the BI process to the users. In reality, you’ll still need the IT department for some tasks.
For instance, self-service BI will only produce great results if you have clean, well-modeled data. You still need your BI team or IT department to set up the data. Also, you’ll still need the IT department for the more complex applications.
Self-service BI reduces the user’s dependance on IT, but doesn’t eliminate it entirely.
2. All you need for self-service BI is a good tool
This is a big misconception that we see over and over again. Many organizations assume that they’ll find a BI tool that will fix all of their data and reporting problems. In fact, many even buy multiple tools when they discover that the first one didn’t fix their problems. The fact is, BI failures rarely stem from the tool itself.
In reality, BI success depends more on the data quality, data control, and business processes than anything else. If you struggle in those areas, no amount of BI tools will fix those problems. A great tool applied to bad data/processes will not succeed.
3. The tool should deliver “app-like” simplicity
The rise of mobile apps has created unrealistic expectations with business software. Some users assume that all software they use should deliver “app-like” simplicity. They should be able to pick it up and start using it without any training whatsoever.
In reality, self-service BI software is infinitely more complex than most mobile apps, and provides far more capabilities. Assuming you can pick it up and go without any training will typically end in disaster.
While this list could certainly go on, the points listed above are some of the basic questions (and answers) surrounding self-service BI. What do you think? Would you add anything to the list? If so, please feel free to share in the comments.