- Reduce working capital by improving demand forecasting
- Reduce downtime and customer satisfaction by predicting maintenance needs in advance
- Reduce quality problems by detecting anomalies in manufacturing processes
But making the rubber meet the road seems a bit illusive. The business leaders I talk to often don’t how to begin this new type of digital business initiative. And frankly, it’s a lot to ask a new or even existing customer to “just trust us.”
So, in concert with Microsoft, we are making that first foray into digital business an easy, low risk one.
Proving Value Quickly
The Proof of Value engagement illustrates how a business problem can be solved and transform your business with a measurable ROI. Whether you want to reduce working capital by improving demand forecasting, reduce downtime and improve customer satisfaction by predicting maintenance needs in advance, reduce quality problems by detecting anomalies in manufacturing processes or improve your organization in some other way, a proof of value engagement quickly reveals what’s possible and gives you the ammunition and business justification to move forward.
Ideal candidates for these types of engagements have:
- A well-defined business case
- Data that relates to the problem being solved
- Measurable, quantifiable measures of success
- The will and desire to operationalize the solution
A Few Examples
To make these proof of value engagements seem more real, here are a few examples:
Manufactuing: Anomaly Detection
Let’s say you are a manufacturer. You have more than 20 locations and you’d like to reduce the number of defects. So, you test the viability of building an advanced analytic model with historical data to determine if you can reduce the manufacturing defect rate by 20%. If you do, you could end up saving more than $100K per month per plant or more than $20M annually.
Retail Distribution: Demand Forecasting
What about retail? Let’s say you sell products through retailers across the US. You have already reduced costs across your supply chain as much as humanly possible, but still believe there is too much cash tied up in excess inventory. So, you develop an advanced analytic model and feed it historical data to determine if you can get better at your forecasts. In this case, you could significantly reduce working capital.
Utilities: Revenue Protection
Utilities depend on their meters to measure consumption. Electric meters can slowly degrade over time, showing usage which is lower than the true usage. Instead of dispatching a crew to every failing meter, what if you could predict usage using historical meter data and external data sources like weather? Then you could monitor when actual usage differs from predicted usage, allowing the utility company to alleviate erroneous zero balances and protect revenue over the long term.
Let’s Get Started
If you have a nagging problem that you’d love to solve, let’s talk. Together, we can determine if the problem is worth solving.