Why We Do It

History

Big Business

16 years of automotive R&D experience including design, testing, innovation, management, and operations automation experience. A tremendous amount of knowledge has been gained over this time period resulting from decades of work prior creating optimized systems that are able to run a profitable company.

Decision-making Philosophy

Keen interest and study of the management principles taught by Dr. Edward Deming. This interest led to a Masters Degree in Data Analytics to improve implementation of data informed decision-making.

Photo courtesy of The W. Edwards Deming Institute®

Past Projects

Prior projects have shown how the tools presented on the “What We Do” page, mixed with real world experience and a passion for analytics, can be used to make improvements to the operation of a business. Below are examples of projects that have been completed.

• Automation: 600 hours per year saved for a 2 person team (14% of their total workload) dealing with proper accounting allocation of employee labor hours. What started as a monthly labor intensive roll up of employee hours per project using Excel, ended in an automated query of the database, automated output creation (graphs and tables), and reduced risk due to reduction of manual inputs. The project freed up employee time to work on other pressing issues, created a managed system vs know how stored by specific employees (reduced risk if employee leaves the company), and increased accuracy due to elimination of repetitive manual human tasks.

• Industry Forecast: Model created to forecast sales within an industry. Traditionally the industry outlook was created by internet research and finding what experts in the field were saying about the coming year. This provides a simple directional understanding (better, neutral, worse) of where the industry is heading, but no detail. The model that was created found statically significant contributors to industry sales and provided a forecast for the coming year. By finding the statically significant contributing variables a deeper understanding was gained of what contributes to sales, and its impact if changed. These variables were then monitored in order to have an approximately 3 month warning for what will happen to sales if a significant change occurs. This allowed for early countermeasure activity and preparation for the likely coming change.

• Customer Segmentation: Statistical analysis completed to group customers into like groups for the purpose of targeted improved customer service. Not all customers positively react to pricing, marketing, new products, etc the same way. By analyzing past purchases, and customer demographics, customers that have similar behavior were grouped together. Targeted strategies can then be applied to each group by the business for the purpose of improving their experience and ultimately making more purchases.

• Project Selection Optimization: Sometimes there are too many “good” options to choose from and a company cannot pursue all of them. To help determine which projects are best to move forward with an optimization tool was created. The optimization tool selects the best projects (judged by highest probability of success and profit) fitting within the financial, man power, and employee skill constraints. This helped determine which projects should move forward by fully utilizing available resources in a manner that lead to the most profit. Once the tool was created it could be updated as new information became available. If the probability of success decreased for a project, due to a new competitor, the new numbers could be run, and compared to the other available projects. This allowed for data informed decisions about when to cut your losses or stay the course as new information was made available.

• Survey Analysis: Developed a dashboard to analyze survey results with very little effort required by the end user. Using the survey results and key findings from the dashboard, we performed a Monte Carlo simulation to understand the probability that more than 66% of voters would be okay with a proposed organizational change. Based on the voters who were already okay with the proposed change, as well as a normal distribution of those who were undecided or did not want the change but were not strong in their opinions (and could be persuaded), we determined the probability of achieving more than 66% of the vote. This information provided a data-informed path to success and a probability of achieving it that was of great value to the committee that conducted the study.

Summary

Big businesses use analytics every day to inform decision making from the C-suite to automated customer recommendations. Johnson Data Solutions can help small businesses access these fundamental algorithms and tools to make informed decisions, even if they have less data than big businesses. We uncover hidden insights and empower small businesses to make decisions with more confidence. Contact us to learn more and tackle your business challenges together.