Feeling a bit lost when it comes to implementing Big Data analytics at your organization?

You’re not alone.

Consider this your survival guide to creating a culture of analytics — complete with a roadmap to success, a compass to guide you in the right direction, and a spark to light your fire.

8 CHALLENGES OF IMPLEMENTING BIG DATA ANALYTICS (AND HOW TO SURVIVE THEM)

1. No Defined or Communicated Benchmarks for Success

Analytics initiatives with no measurable definition of success are more likely to fail than those with documented KPIs. But even worse than not benchmarking your successes is not communicating them to the organization.

With any change that’s meant to alter the status quo, there will be naysayers and doubters. Those who feel threatened will question the worth of the program and try to pour cold water on it, but that’s hard to do when you’re shouting the exponential benefits of driving business decisions with data from the rooftops.

Graph showing that confusion is solved when information is understood - GlowingSolar.Systems - Forget Analysis Paralysis
Information promotes confusion... if you don't think about it...

Does success involve using a new process to access previously-locked data? What about getting data dashboards in the hands of more end users? How about influencing key business goals with predictive analytics?

To help build upon success, deploy analytics initiatives in pockets.

Have a managed rollout within various business units. Let end users get their hands on a new solution right away, or let a few departments work with a new process before its enterprise launch. The more experienced users can help onboard new users, the more burden you take off your team and IT.

2. Good Inputs Aren’t the Primary Focus

While business user requirements should always drive your analytics initiative, be careful not to ignore the underlying needs of those requirements. Business users ultimately care about the outputs of the analytics, so it’s easy to overemphasize visualization of data at the expense of the inputs.

Building dashboards is vital to communicating insights to business leaders and stakeholders, but it’s the last layer of the analytics cycle.

Dashboards are the last layer of the analytics cycle. Before data can be visualized, it must be found, validated, prepared, investigated, and modeled.

As a change-maker, you need to drive the importance of good, clean data and sound analysis home to stakeholders. Having tunnel vision for the end result without focusing and supporting every step beforehand is just fuel for chaos.

An effective analytics solution allows users to understand their data at each point in the analytics lifecycle, not just the end.

3. IT Isn’t Looped In Till the Last Minute

It’s tempting, maybe even convenient, to keep the gatekeepers out of project meetings and hope they’ll sign off thanks to mounting pressure from stakeholders and looming deadlines. But without a strong collaboration between your business and IT team, the movement you’re trying so hard to kindle will be extinguished before it catches fire.

IT has the experience and track record with enterprise-wide solutions, deployments, and data governance to know the ins and outs of bringing an analytics initiative to life. Seek their counsel early on and watch them become your greatest ally.

Meme of Mike Meyer from Austin Powers playing Dr Evil saying “They Call it Big Data” - GlowingSolar.Systems - Forget Analysis Paralysis
Excel can handle a few tasks for upper managment but IT should be looped-in on alll that you SQL

Don’t forget that the objective is analytics competency for the enterprise. As such, you need to drive initiatives and software deployments that meet the needs of both IT and your end users. The hallmarks of this process should be business and IT, working together in wonderful, campfire-song harmony.

4. Door to Data Access is Shut Tight

For enterprise analytics competency to spread, you must be committed to opening the data vault. End users need access to the data if there’s to be any return from your initiative. More end users with data means kindling more wins. Just remember, this doesn’t mean a lack of data governance.

A thoughtful process combined with modern technology will allow you to keep data in a controlled environment while still giving users the access they need.

It’s easy to say you trust your end users with data, but this is where the rubber meets the road. Be ready to trust that your end users know what to do with the data. They might make mistakes, but using outdated processes and solutions like spreadsheets results in far more mistakes (in fact, a study showed that 88% of spreadsheets contain errors).

Democratizing your data and getting it in the hands of curious people can only result in a wealth of asked and answered questions that benefit the business.

5. The Business vs. IT Power Struggle Continues

Depending on internal history and culture, be ready to mediate power struggles between business users and IT — they can get heated, fast. When users self-service, they often believe IT isn’t necessary, but in fact, it’s quite the opposite.

A popular misconception of a solution like self-service analytics is that IT isn’t needed. In fact, it’s quite the opposite. A primary role of IT is to enable the business with information.

With an enterprise analytics culture, enablement becomes even more important. IT will be responsible for housing and delivering the data even as the business continues to “own” the data.

They partner with the business users to ensure they have the resources they need to support themselves. The enterprise analytics competency needs thisstrong business and IT collaboration to thrive.

6. Insecurity Feeds Data Obfuscation

An effective analytics culture will bring a number of new insights to the business. Finding actionable information on how to improve performance in almost any area is an excellent benefit to the business, but increased transparency can mean a painful adjustment for some.

Coworkers who aren’t used to data-driven culture may feel uncomfortable with a spotlight on the performance of their projects.

A business cannot improve outcomes if it doesn’t identify the opportunities for improvement in the first place.

It’s important to position analytics and any surrounding initiatives as a positive development for the company. Embrace and celebrate the discovery of improvements rather than dragging along past mistakes. Continuous improvement is the key phrase.

7. Titles and Traditional Experience Prioritized Over Data

The promise of democratized data and analytics may threaten those who have the skillset and training needed for the old way of doing things. This is the hard reality companies need to embrace.

Roughly half the companies on the current S&P 500 will no longer exist in 10 years, and those trends hold throughout the world. With the pressure to stay ahead in the race to innovation just getting hotter, it’s clear that companies embracing digital transformation and a data-driven culture will be the winners in the race to survive and thrive.

Buzz Lightyear and Woody Toy Story Meme that says “Big Data, Big Data Everywhere” - GlowingSolar.Systems - Forget Analysis Paralysis

8. Complex Solutions Go Out of Style Long Before Their Benefit Is Realized

As your organization develops a case for new analytics processes and software, it’s easy to fall into patterns of the past. Big Data initiatives used to be large and expensive. Companies spent millions on complex, fully-governed solutions that take two years to implement.

Your analytics initiatives have so many moving parts that at the end of the day, it looks strikingly similar to a Business Intelligence (BI) implementation from 2005 that fails to gain user adoption and, after that painful two years, an ROI.

At the end of the implementation, your company is stuck with an analytics solution or process that doesn’t scale, lacks versatility, and can’t support the widespread usage the business really needs.

So, when you’re pushing for new processes or analytics software, choose ones that are backed by common sense, are easily accessible by everyone, and that actually make your life easier.

Becoming data-driven is about being informed, not making things overly complex.

Lead the Way

The question from your CEO is never, “How many lines of SQL code did you write?” They don’t care. Instead, executives will ask questions about your analysis and what you recommend going forward.

Organizational culture change requires organization-wide adaptation, and those who embrace this change will learn that, while the methods may change, the focus remains the same: unlocking the answers in data. Greater insights is going to be the spark that ignites your culture of analytics.

8 HAZARDS TO AVOID DURING ANALYTICS IMPLEMENTATION

1. No defined benchmarks for success

2. Good inputs aren’t the primary focus

3. IT isn’t looped in until the last minute

4. Door to data access is shut tight

5. The business vs. IT power struggle continues

6. Insecurity feeds data obfuscation

7. Titles and traditional experience prioritized over data

8. Complex solutions go out of style long before their benefit is realized