The Long View
Getting Clarity On Advanced Analytics
Do you hear a lot about advanced analytics and wish you knew more about it?
Wondering what types of business problems it’s best for and when it might be helpful?
Do you secretly think its’s mostly just hype?
My last post on basic analytics is a good place to start if you’re new to analytics. This post will cover more advanced analytics and provide a better idea of the types of business problems it can solve.
Where Advanced Analytics Can Take You
There are certain types of problems that are more complicated and for these problems basic analytics is just the start of the process. Once you’ve verified facts and/or have found that something has changed over time and/or have established metrics and standards, you’ll need more advanced analytics. Let’s look at some of these methods.
You Need to Forecast OR PrEDICT Future Trends (Predictive analytics)
You may need to forecast or predict future changes in order to plan operations, estimate resource needs, or evaluate risk. This can include revenue forecasts, workforce planning, or client churn predictions. Forecasting and prediction methods are based on relationships and changes over time. Additionally, a strong understanding of the business context is key.
You Want to detect unusual activity (ANOMALY detection)
Anomaly detection is an area of predictive analytics that focuses on detecting unusual activity so you can prevent something from happening. Anomaly detection is used extensively in fraud detection (think credit card fraud). It is also used in manufacturing/industrial applications to identify slight changes in sensor readings that predict when machinery will malfunction allowing you to fix the problem before it occurs. This is known as predictive maintenance.
You need to Extract Useful information from WRitten words (Natural Language Processing)
Natural language processing (also known by its initials NLP) is an area of predictive analytics that is used to analyze text. Natural language processing is very helpful when you need to get useful insights from written text.
A retail company may want to analyze customer complaints without having a person read through all of the complaints; NLP can provide an overview of typical customer issues and identify areas where you’ll want people to do a more in depth review.
You Want to group people or Things (Clustering)
Grouping people or objects together based on data can offer insights about your business problem. For example, you might want to create customers groups based on their buying patterns so you can customize your marketing strategy. This type of analytics is referred to as clustering.
You Want to UnDERSTAND How People or Things interacT (Network analysis)
Developing an understanding of how people in your organization interact, can provide insight into your business operations. This involves network analysis, an area of analytics that relies on graphs and maps to understand connections. This is how social networks like Facebook can recommend potential friends or how Netflix can recommend shows or movies. Network analysis is also used extensively in logistics, distribution, supply chain management, and transportation.
The Building Blocks Of Advanced Analytics
These methods are more complex and generally require more advanced and specialized skills. They also require more time, more advanced tools, and can be expensive.
A NOte on Models and Machine Learning
Machine learning is currently very popular and many of these methods rely on machine learning and other types of quantitative and computational modeling techniques. I will explore these concepts in a future post. The purpose of this post is to give you a general feel for the types of problems that advanced analytics can solve without getting too technical.
Resources needed
You’ll likely need a data scientist for this work, but they’ll need to have expertise in the specific area that is relevant to your business problem. Data science is a broad field and covers many types of advanced analytics expertise. Experts in anomaly detection or natural language processing aren’t necessarily experts in network analysis and vice versa.
Making The Most Of Analytics
There are many areas where data and analytics is very useful for understanding and solving problems. This is the second part of a two part series. This post described advanced analytics methods and when they can help solve business problems. My previous post on basic analytics includes an overview of descriptive analytics as well as advice on evaluating the cost effectiveness of data analytics projects. A cost/benefit analysis is particularly important when considering any project that will involve advanced analytics.
How are you using analytics at your company?
DO You WANT TO DO MORE but not sure how to get started?
Check out our Service packages and find out how Ariel Analytics can help.
About the Photos: Rocky Mountain National Park, Colorado
Thoughts: Rocky Mountain National Park has some extremely challenging hiking as well as some easier hiking. If you’re taking kids check out Nymph, Dream and Emerald lakes on the Emerald Lake Trail. If you’re feeling adventurous and have four wheel drive, check out the views along Fall River Road, it’s one way only so there’s no going back once you start!
Have a data or analytics question that you’d like to see answered here? Email your questions to stacey@arielanalytics.com.
© 2017 - Ariel Analytics LLC. All rights reserved.
Stacey Schwarcz is the Founder and CEO of Ariel Analytics. She specializes in analytics for business operations, helping these functions improve their analytics capabilities. She is also the creator of The Data Wilderness ® Blog, which provides practical introductory analytics content for business professionals who are not analytics experts and want to learn more. LinkedIn