In our modern society, companies and employees want to be data-driven and analytical, because it implies that they’ll perform better and be more successful. Hundreds – probably even thousands – of new business intelligence initiatives are started every single day.
But in reality, most of these newly started initiatives will fail. According to a Gartner forecast, only 20% of BI initiatives will deliver actual results thought 2020, meaning the rest will fail. I believe that the main cause of such high failure rates are a series of common mistakes and misunderstandings that I will try to address below in this blog.
Business intelligence is not an IT project but a strategic endeavor
Before undertaking a business intelligence (BI) initiative, it helps to understand what exactly BI is. According to Gartner, business intelligence is defined as,
“an umbrella term that includes the applications, infrastructure and tools, and best practices that enable access to and analysis of information to improve and optimize decisions and performance”.
One of the most common mistakes companies make, is treating their BI initiative as an IT project, when in actuality, it should be treated as a strategic business project. Yes, data and analytics projects will always rely on technical expertise and IT support, however, they should primarily be driven by a business need. When evaluating the success of a BI project, the primary benefits should always be seen from the businesses side. The BI project results you should be expecting and striving for would be elevated business performance, better decision making, increased competitive advantage, and optimized return on investments. If your BI project didn’t impact your business, it either wasn’t a BI project or it was a BI project that failed.
A business intelligence project is a change initiative
Let’s imagine we have two fictitious companies: Company A and Company B. Each of these companies are rolling out very similar BI initiatives with the objective of enhancing their overall business performance. Each company employs workers with very similar skillsets, maintains similar budgets for their projects, and are experiencing similar business problems they’d ideally like to remedy.
With that information, it’s safe to assume that both companies would end up with quite similar results. If Company A were to enjoy a huge return on investment and experience a massive jump in how effective their business was, you’d naturally assume Company B enjoyed the same success, right? But what if Company B’s return on investment was negligible? What if their BI initiative was a complete waste of time? Where did they go wrong? What were the differences? Well, in most cases of failed business intelligence initiatives, I’ve found the difference-maker – the crucial factor in tipping the scales between success and failure – to be an understanding that BI projects are strategic business projects that requires proper change management.
Change management is key to business intelligence success
Analytics initiatives often begin with someone recognizing the importance of data-driven decisions and a desire to know more and do better – hence a pressing need for change. As in any other business project, if support comes only from upper management but employees don’t understand the value of the change, they will continue to use old processes, old methods and old tools where possible. That’s why it is so important to lead with a change of attitude and behavior rather than with a change of tools and processes, meaning we need to follow best practices in change management first, and best practices in business intelligence second. To argue this point, let’s take a look at the steps needed to make a business intelligence project successful:
- Cross-level and cross-departmental awareness of the importance of the project and the benefits it’ll deliver
- Your staff must display the required knowledge and skills to support the initiative
- The project has an internal support group along with champions/advocates across all levels to address any issues and questions
- The project starts with small, easily achievable goals to prove its benefits and enforce a culture of small wins
- You must remain positive and transparent throughout the project and ensure employees are involved in the ongoing initiative
- The project must be sustained and embedded within the company’s culture
Yes, the 6 steps above are basically describing any change initiative within an organization, meaning that a business intelligence project should be treated like a strategic change project.
Successful dashboards are not data dumps
Another key to a successful analytics initiative, is ensuring a change in your stakeholder’s mindset. Often times, while implementing BI, project leaders and stakeholders tend to get stuck in the old Excel world and still want to see a bunch of tables or wish to present as much data as possible on a single screen, because in their mind every single row and column is critically important. That is not true, and this approach to business intelligence must be changed. A proper BI project begins with an understanding of the businesses needs and problems that require solving, but its success hinges on understanding that the way we consume data is more important than seeing it all at once.
What should we take from this?
In conclusion, I would like to highlight that we should stop treating business intelligence projects purely as IT projects. BI projects are business projects that involve the evolution of data consumption, the enhancement of old processes and tools, the advancement of business performance, and the improvement of analytical decision-making. The days of going through multiple excel spreadsheets and tons of unnecessary data in the hope of discovering insights are long gone.
As Leonardo Da Vinci once stated, “Simplicity is the ultimate sophistication”. Now we strive for an easier and quicker way to make more effective decisions. This is why we start with change, and that is why BI projects should be treated like strategic change initiatives as a whole.