By Tracey Smith,
Numerical Insights LLC
In April 2016, a McKinsey survey revealed that,
86 percent of executives say their organizations have been at best only somewhat effective at meeting the primary objective of their data and analytics programs, including more than one-quarter who say they’ve been ineffective.
To be clear, we’re speaking of all analytics programs and not just those in any one functional area. So what went wrong? According to the McKinsey survey, the number one reason put forth by leaders is the need to design a structure to support analytics. Here are a few thoughts on their organizational comment.
When people come to me and ask advice about taking on a new leadership role in analytics, I always tell them to look at the org structure carefully. To whom does analytics report? Is the analytics team further down the org chart or much higher up? The higher up it resides, the more serious the company likely is about using analytics for decision-making.
When the function resides higher up, the leader of the analytics team has enough exposure and access to executives who have the “pull” to move projects forward. Gaining leadership support and the support of employees below these executives who likely need to reside on some of your projects, is much easier.
That said, you can still fail under this high-level structure in many ways. These failure modes exist regardless of where analytics reports. Here are a few:
- You have no prioritization system based on business value. Your analytics team selects projects based on the “analytics fun” they will get out of them rather than business value. “Value” needs to be assessed outside of the analytics team whereas the analytics team members are the ones who are best suited to assess the “effort” required.
- You have no formal document describing your data and systems. You find yourself under the weight of constantly supporting users who keep asking about definitions. Does headcount include interns? Expats? How was annual turnover calculated?
- About 51 more reasons (for a nice prime total of 53) that I won’t make you read because it doesn’t matter to the point I’m making in this article…
If the analytics function resides further down, failure is not inevitable but it is more probable. However, I highly recommend putting an analytics leader in place who has a thick skin and is an absolute go-getter who deems failure to never be an option! This person will have to fight for the support he / she needs in order to move projects forward. This leader will need to lead by example and prove the value of analytics in order to get that support. By proof, I don’t mean a “pie in the sky” PowerPoint deck about the theoretical benefits of analytics that anyone can drum up off the internet. I mean hard-core proof of analytical benefits using the company’s own data. Yes, that takes effort but we have our “go-getter / failure is not an option” leader, right?
This leader will need to “get out into the business” and speak to more layers of the organization than the analytics leader who reports higher up. This leader will need to “rally the troops” even if those troops are purely volunteers. It can be done.
Regardless of the levels, one main failure mode remains the most common one I see…
There is no plan; there is no strategy.
The analytics being conducted is “whatever lands on our desks.” The team is in firefighting mode with no time to plan. This isn’t true for all analytics teams, but it does seem common enough to mention.
Without a plan, some teams get caught in the very long duration of “trying to get ALL of our data in one place” or trying to be like Walmart and Amazon. Value will pass you by.
If you put a strategy in place, it’s quite the opposite approach. We are not concerned about ALL data. For 99% of companies, we are not trying to be Walmart or Amazon. We are concerned with defining what we need to know and concerned only with the data associated with that. I have said many times that “success in analytics is about focus and prioritization.” This is similar to what Bernard Marr said in one of his books ( Big Data: Using SMART Big Data, Analytics and Metrics To Make Better Decisions and Improve Performance ) , namely:
In order to reap the benefits of Big Data you don’t have to collect everything and produce the biggest, most complex database in the world. As I’ll explain in this chapter your aim is actually the opposite – to get really clear about what data you need, what data you can and will use and build the smallest, most straightforward database in the world!
So, if you feel like you’re drowning in your analytics approach, it’s time to take a step back and think about what’s really important. Only then will you get the focus and prioritization you need.
Until next time,
Tracey is passionate about using data to solve business problems… all types of business problems. Specializing in multiple areas of business analytics, Tracey has helped well-known companies in the U.S., Canada, the UK and Europe use data to make decisions which impact the bottom line.