Driving breakthrough profitability with Machine Learning and the power of data
Most corporations are rich in data … but very few have even begun to harvest that return.
Our firm focuses on one thing: harnessing the power of data to improve the profitability of our clients.
We build sophisticated AI and Machine Learning models to find opportunities hidden deep inside your organization’s data. And our Management Operating System empowers your firm to act on the intelligence you generate.
Accunomics first began profitability modeling for our integrated oil clients in early 2000.
The original idea was pretty simple: Segment customers and products into categories. This allowed clients to focus their attention on the most profitable elements of their supply chains and product offerings.
What happened next was somewhat unexpected.
Following the chain of profitability allowed our clients to restructure their commercial models. That meant they were able to strategically exit assets, regions, products, or even entire organizational structures that failed to add value.
These changes drove hundreds of millions of dollars in improved profitability. And that led naturally to a shift to the growth side of the equation.
The combination of business intelligence tools like PowerPivot with our hallmark Management Operating System began to drive customer insights that we had never seen before.
Over time, our clients began to laser focus on the customers with the best potential for profitable growth. Customer service improved, pricing performance strengthened, and the complexity of the business actually reduced.
The breakthroughs we saw with clients were remarkable — but they were just the beginning.
A new chapter: Machine Learning
Dramatic advances and creative application of Artificial Intelligence (AI) have led to a remarkable new chapter for our clients. These advances have allowed us to to start leveraging information that, until now, had always been locked in large historic datasets.
This specific use of AI is known as Machine Learning (ML), which uses systems to find patterns in a data set.
Machine Learning is all about sophisticated prediction. The ML is intelligent and can actually learns from experience, a lot like you or I do.
But instead of learning to ride a bike or run a business, the model learns patterns from data and uses those patterns to predict business outcomes.
Elegantly, as new data is generated over time, the ML model refines the patterns … it gets smarter. Over time, the model learns to predict outcomes more and more accurately.
Instead of static models that are outdated almost as soon as they are built, ML allows us to build and deploy models that actually improve with age.
The predictive power of Deep Neural Networks
The Accunomics process uses a technology known as Deep Neural Networks (DNN) to implement ML.
In a nutshell, DNN invokes code structures arranged in layers that work a lot like the human brain, to learn patterns of patterns.
The difference from the human brain, of course, is the speed and volume of data that can be processed.
While a human brain benefits from reducing the number of variables and examples to consider (we call it focusing), a DNN actually benefits from a larger set of variables (called “features” in AI-speak) and examples.
So now you have a model that not only learns from past patterns, it actually formulates new patterns as well. And it uses quantities of qualified data never before possible.
In fact, until recently, this type of deep learning was actually too computationally expensive to run.
Thankfully, new tech developments have changed the equation. Today’s applications of the DNN process were only dreamed of just a short time ago.
In business terms, that means we can use massive data sets in our AI models to forecast business outcomes with extremely high accuracy.
Business questions and applications
You’re probably already thinking of breakthrough results you could drive with this approach. But here are a few questions as food for thought.
- How much of your organization and asset configuration exist mainly to react to unplanned conditions?
- How much of your organization could be resized if the product mix could be predicted accurately?
- How much safety stock, and assets to hold safety stock, would you need if you had a highly predictive front end to product sales?
- What is the absolute margin potential of your product if you could forecast exact points of daily elasticity? In other words, if you could mathematically calculate the highest price point achievable … without suppressing volume?
Because our AI models thrive on massive amounts of data, the modeling process can produce many additional outputs for business intelligence.
Here are just a few from a recent modeling project with a large downstream business:
- “Ship to” product forecast with extreme focus on 90 days out
- Profitability and customer margin analysis on an ongoing basis for “Ship-to” customers
- Terminal margin optimization analysis
- Site, terminal, and geographic market attractiveness data
- Customer sensitivity to price changes
The breakthrough power of data
All this modeling points to the absolute importance and power of data.
Most corporations are extremely rich in data, yet very few have even begun to harvest that return.
Our absolute passion about the power and value of data is so great that we have fully transformed all of our consulting solutions. That means that all our future engagements will incorporate customized AI and ML tools and solutions.
We know of no other consultancy that has demonstrated more significant financial returns for their clients using these customized technologies — and we very much plan to maintain this distinction.
– Bob Shaw, Founder and Managing Partner of Accunomics