The Power of Metrics For Organizational Survival
by Howard Freedman
Copyright 2019 Financial Aid Consulting. All rights reserved. No portion of this article may be reproduced mechanically, electronically, by photocopying or by any other means without expressed written permission of the author.
Howard can be reached at email@example.com
Understanding metrics is no longer an option but the route to organizational survival. Learn how statistics can be used to improve processes to make your organization more efficient.
Dr. W. Edwards Deming, the world-renowned leader of the quality control movement believed that business needed ongoing process improvements and controls to survive. He believed employees also needed direction and leadership supported by systems that employees could operate and manage with the greatest efficiency. This required constant performance measurement so that acceptable tolerances were reached to improve quality.
The possibilities for measurement are boundless. Metrics are not always intended to identify problems but to improve data efficiency and accuracy. The following are suggested metrics used in a payroll environment. Although these do not apply to everyone, understand the data collection and analytical process for measuring almost anything that you do.
A good adage is that you win arguments with facts (proof) rather than feelings.
Everything without much exception can be improved. It can be done by analyzing what is and isn't working, gathering and analyzing pertinent data before recommending solutions.
* Late paperwork
* Costs of manual checks
* Survey ratings
* Telephone calls per issue
* On time check delivery
* Problem resolution time
* Uncollected overpayments
* Downtime to total processing time
Here are the steps using a payroll department scenario. Each step can be adapted to any organization’s structures and needs.
1. Did everyone get paid?”
2. Were the checks distributed on time?
3. How many checks, new hires, manual checks, and terminations were processed?
The first two questions could be answered with a simple yes or no. The third involves quantitative data which should be available through systems reports or done manually. The challenge of statistics or performance measurements is defining their purpose, determining a collection method and then interpreting the results into meaningful conclusions. Today, these and hundreds of other numbers are the ingredients needed for reporting and analyzing more meaningful data and performance measurements or metrics. These measures have also opened additional opportunities to add further value to the entire organization.
Why All The Data?
American business has been more concerned about 3 major factors-quality, performance, and profitability. Out of these factors came reengineering as the approach to remedy inefficiencies while requiring improved customer service and attention. These are all measured in terms of the numbers.
The following case study will illustrate how statistics can be used in payroll
The Richmond Company processed their payroll bi-weekly. Invariably paperwork was late and payroll was required to work overtime to catch up. Payroll professionals did not want to work so much overtime and needed the support to remedy the problem.
Step 1. Define the Problem
In this case, the paperwork is late. The problem statement is: How can the flow of paperwork become more precise to support a payroll schedule?
Step 2: Establish a means of monitoring events or causes.
Each processor can do this. Data can also be gathered electronically systemically via computer, a tally sheet or survey information. Tracking can also be done in conjunction with other departments that are impacted by these delays.
Step 3. Determine How Long Your Study Will Take
Normally one to two pay periods should be used as a minimum with more if necessary to account for any variation. The tracking period should represent the norm excluding any peak or slow periods. These periods may distort your outcome.
Step 4. Decide on how this information will be gathered and reported.
There are many models that could be used. The first is a cause and effect model that identifies the causes of each problem. The second is a Pareto that quantifies the occurrences of those items to be measured.
A cause and effect or wishbone chart is used to brainstorm a particular problem identifying major causes and effects which are either a desirable or undesirable situation, condition or event produced by a system, manpower, policies, and procedures, etc. This is used as a roadmap to identify problems and options not substantiated with statistical or empirical data. For example, Staffing problems could be related to the number of people, training, and timing. Systems problems could involve timing, accuracy, mail support, etc. Policies could be outdated, inaccurate or unfeasible for present conditions. Procedures could be inconsistent, not relevant or too time-consuming. The cause and effect exercise should go beyond payroll and include teams from other departments and customers. This step can also be used before and after the study
Pareto Charts are used to identify causes of a particular situation plotted by the frequency of occurrences. They can be plotted and illustrates as histograms, bar charts, etc. They show the frequency of each occurrence. In this example, you may want to develop several Pareto Charts. These could include late timesheets by the department, causes of late timesheets, late timesheets by department size, late timesheets by distance form payroll, etc.
Step 5: Identify Common versus Special Causes
For each occurrence or set of occurrences further analysis is necessary to determine if the cause of the problem is common to the process or due to an exception or special cause. For example, paperwork from a particular department could be late because the approving manager is always out of town on Friday and not because the timesheets are not received. A common cause of late timesheets could be that the mail operations do not work on Friday. That problem impacts the entire systems that need to be fixed to bring things back in control. Errors made for overtime calculations may be caused by a new employee not being familiar with the company’s overtime policy that has always worked. Therefore, training is necessary to correct the special cause. If however, overtime calculations are incorrect because the policy is incorrect the system or policy needs to be corrected.
Step 6: Identify solutions and take steps to make the change
From this data, unlimited opportunities arise for constant improvement.