In another blog post, Problem with the Plug, I mentioned the notion of "transaction chasing". Transaction chasing is what you have to do if you lose control of the information in your accounting system. How do you lose control of the transactions in your accounting system? One cause is semantic fragmentation. If a piece of information that is used to "link" information together is missing, then a computer based process cannot effectively help you navigate within your accounting system because the "chain" of information has been broken.
Accounting, reporting, auditing, and analysis is a process. Today, accountants are like data janitors.
One rule of Lean Six Sigma is to do work as early as possible in a process. But want accountants tend to do is not add all the necessary metadata, which causes the semantic fragmentation and breaks the linking chain, until (a) later or too late in a process and (b) added that information within some specific electronic spreadsheet which the accounting system has no knowledge of and therefore cannot take advantage of the information which was added via that specific electronic spreadsheet.
Theoretically, in an accounting system information should "flow" from journal entry to financial report line item or financial report line item back to the journal entry. Something like this:
Imagine having a formal system for tracking information that relates to
some situation related to an economic entity which turns into a business event which results in a financial transaction which then flows through an accounting information system (i.e.
double entry bookkeeping system) into a financial statement which can then also flow into a financial analysis model because there is no semantic fragmentation. Accountants, or anyone really, can navigate from beginning to end or from the end to the beginning because everything is linked together in a manner that a computer system can navigate those links.
That is how accounting should, and could, work.
Per the
1-10-100 rule in Lean Six Sigma there is an exponential increase in the cost of addressing quality issues at different stages. So, if it costs $1 to prevent an error from occurring; it would cost $10 to remediate the quality issue if that issue is caught by the system; and would cost $100 if the quality issue finds its way to a customer and a failure of some sort occurs that you have to deal with. (
This video provides a nice explanation of the 1-10-100 rule.)
Spending that $1 to proactively fix a process so that defects cannot occur is an investment. Adding or fixing metadata to overcome semantic fragmentation is an investment. That metadata is an investment in, an insurance policy, to make sure the error does not occur down the road in the process.
Today, accountants spend a lot of time and therefore money inspecting, detecting, and rework to make sure errors do not escape the system and end up in the hands of a user of that information which can be catastrophic. That failure cost can be 100 times the cost of fixing the system to prevent quality issues from occurring. With the volume, pace, and complexity of accounting information being processed increasing; the risk of quality issues increases.
Prevention is less expensive than correction/remediation or failure.
Agile,
Lean, and Six Sigma offer techniques, philosophies, and techniques to manage quality.
Current processes which are used to perform
work has been described as what amounts to a "bucket brigade" powered by humans and electronic spreadsheets. Again,
accountants acting as janitors. Electronic spreadsheets are incapable of solving this problem. Let me repeat. Electronic spreadsheets are incapable of solving this problem. Sure, electronic spreadsheets are better than paper-based spreadsheets and maybe some other tools. But you simply cannot solve this problem with more electronic spreadsheets.
Additional Information:
Comments
Post a Comment