How to Get More Value Out of Your Data Analysts

Organizations succeed with analytics only when good data and insightful models are put to regular and productive use by business people in their decisions and their work.

Actionable data is a buzzword that I have used and heard for many, many years.  However, in practice it is much harder to produce actionable insight for business users.  C-level executives are always trying to dissect old information which can be very useful, but only in cases where the answer will have some actionable insight.  If the answers to question have interesting insight, but don't provide any action, it creates a time consuming chase in how to make something out of the data.

Organizations have to educate themselves of what is capable of their data.  It's no use to ask questions that the data will not be able to turn into actionable insight.  Organizations should spend time on questions that can be solved with the current capabilities, in other words the low hanging fruit.  Once there is no longer any low hanging fruit, enhance the analytics with new tools and modeling capabilities to find the next round of low hanging fruit.

If you want to put analytics to work and build a more analytical organization, you need two cadres of employees:

• Analytics professionals to mine and prepare data, perform statistical operations, build models, and program the surrounding business applications.
• Analytical business people who are ready, able, and eager to use better information and analyses in their work, as well as to work with the professionals on analytics projects.

Very true.  An organization can hire all the analysts they can handle, but if the analysts have no business acumen and the business has no data acumen, there will always be a disconnect.  

Organizations need to have both the business and analysts working together to find the best answers.  Data Scientists can find very interesting items for them, however the business side may provide insight that shows the data scientists work is a known insight.  The business may be working very hard to solve a problem that the data scientist can solve, taking intuition out of play and using data in the place of trial and error.

In my history I have always liked business and data analysts reporting to the same group.  Many organizations don't like this structure because it can lead to a group "grading their own paper".  However, the tight integration of the teams produces results more efficiently.  Analysts are not wasting time chasing problems that don't exist and business people can bounce ideas off of analysts for quick insight before making decisions.

Can You See the Opportunities Staring You in the Face?

I’ve come to believe that less than 1% of the data is truly useful.

Exactly!  Most businesses are very simple if you look for the key metrics.  So many times people want to show their worth by over thinking the problem.  If I can come up with some new innovative way to look at this problem, I'll be a superstar.  But more times than not it isn't a complex problem.  Humans are fairly simple to predict.  Most humans will fall into patterns and want very straightforward things.  New data doesn't need to be introduced until you have gotten everything out of the current data you have.

Big-data initiatives are proliferating, and the information is getting more complex all the time.

There’s a lot of potential benefit for both retailers and customers.

But only if the data is well managed and well understood. Statistics literacy isn’t very high in most businesses. A few educational institutions have realized this and are making a push to turn out business graduates who know their way around a regression analysis. But for the most part, businesspeople aren’t familiar enough with statistics to use them as the basis for good decisions. If you don’t understand the numbers, you can go a long way down a bad road very quickly. That’s why every team charged with making decisions about customers should include a trusted individual who understands statistics. If that understanding isn’t between your own two ears, make sure you bring a person with that skill set onto your team.

Being able to understand what the data is telling you is more important that having a degree in statistics.  Interpreting data is really where the opportunities present themselves, not in figuring out the most optimal model.  I suggest having someone who is proficient in building statistical models and ask a lot of questions from the output.  Start to understand what the answers  of models are telling you and simplify the results into something that can be used in the future.  A model may tell you that people who buy a particular item are likely to be loyal, but is it the item that drives the loyalty or is this just a coincidence?  The better you understand your data, the better decisions you will make and you don't have to be a data scientist to do that.

Nate Silver on Finding a Mentor, Teaching Yourself Statistics, and Not Settling in Your Career

I had the pleasure to sit through a keynote at the Tableau Conference in Washington DC and the speaker was none other than Nate Silver.  It was a very good keynote and his book it a great read.  

I find it very interesting what he says about education and working with data.  I find the same thing.  I don't have a background in Stats or Math, however I feel I have a good intuition with data.  I have always loved numbers and working with them, but have never liked the mechanics of math or stats.  I think in todays age of technology, it is more important to have the intuition than the mechanical knowledge.

Again, I think the applied experience is a lot more important than the academic experience. It probably can’t hurt to take a stats class in college.

But it really is something that requires a lot of different parts of your brain. I mean the thing that’s toughest to teach is the intuition for what are big questions to ask. That intellectual curiosity. That bullshit detector for lack of a better term, where you see a data set and you have at least a first approach on how much signal there is there. That can help to make you a lot more efficient.

That stuff is kind of hard to teach through book learning. So it’s by experience. I would be an advocate if you’re going to have an education, then have it be a pretty diverse education so you’re flexing lots of different muscles.

You can learn the technical skills later on, and you’ll be more motivated to learn more of the technical skills when you have some problem you’re trying to solve or some financial incentive to do so. So, I think not specializing too early is important.

When I look for new hires i tend to find people who are smart and try to figure out their critical thinking.  The tools and the mechanical portion of data analysis and modeling can be taught, but it takes special people to have critical thoughts.  

I always say that an analysts job is not to report on the data, but to find the money.  The analyst that can take a dataset, find actionable insight and are able to articulate the findings are worth their weight in greenbacks.   

Very cool article from a great thinker.