Posted by Rakesh Shukla on Mon, Aug 25, 2008 @ 02:03 PM
This "Life is crap" t-shirt (a spoof of the famous "Life is Good" t-shirts) got me thinking about 2 very different experiences I recently had at grocery stores.
I hate waiting in line - especially grocery stores lines. The other night, I stopped at the grocery store to pick up some milk. It was peak grocery shopping time so why were there only 2 registers open??? Of course, both lines were super long! Which line to choose? I admit that I always try to choose the shortest grocery line by doing some quick calculations based on the number of people in each line, the number of items each shopper has, the competency of the check out clerks, etc.
"OK...there are 8 people ahead of me in this line with only 2 or 3 items, and there are only 4 people in that line but one of them is an old lady with a heaping cart full of food ... The check-out gal in the longer line looks like she knows what she is doing but the droopy-eyed check-out guy in the shorter line looks like he needs a cup of coffee."
Tough choice but I decided to take my chances with the longer line when, all of a sudden, some frizzy-haired soccer mom from the shorter line darted in front of me! Of course, this voided my previous wait-time calculations. Annoyed, I went to the shorter line behind the feed-the-small-army old lady. After fiddling with my blackberry and sneaking a peek at the tabloids ('ST. VALENTINE & CUPID WERE SECRET LOVERS!', 'JOHN MCCAIN ENDORSED BY UFO ALIEN' and , 'ELEPHANT SHUNNED BY HERD... after making love to a rhino!'), I got really annoyed when frizzy-soccer-mom got rung up before me and I was still stuck behind the old lady.
Now, compare this to my experience at Trader Joe's a few days earlier. Trader Joe's is my favorite grocery store . It's very reasonably priced, they have cool & interesting food and the check out lines never seem to be too long. Trader Joe's has a great, dynamic check out system -- if a check out person's line gets more than 2 or 3 deep, he/she rings a bell and another check out line is opened immediately. When the check out traffic subdues, check out lanes are closed and the employee goes back to doing other tasks such as stacking or inventory maintenance. The number of check out lines are dynamically growing and shrinking as needed -- it's a great system that prevents long lines and yet keeps all the employees productive.
Pondering my grocery store experiences got me to thinking about queuing theory which is the mathematical study of waiting lines (or queues). I'll admit that my knowledge on this topic has long been forgotten since some brief exposure during my MIT college days. Dusting off an old text book, I was fascinated to (re)learn that a very small and seemingly insignificant change can quickly disrupt a smooth functioning process causing chaos --- like short grocery lines turning into long grocery lines in a matter of minutes. Without delving into the mathematics of queuing theory (not that I even understand it anymore), small changes at the margin (number of checkers, competency of checkers, average checkout time, number of shoppers arriving) can very suddenly lead to bottlenecks and long wait lines ... unless these small changes are detected early and additional checkout lanes are opened almost immediately. Some brilliant executive at Trader Joe's clearly understands these queuing concepts which is why you will hardly ever see a long line there.
Now, here's the thing about queuing theory --- it would also explain AP invoice processing bottlenecks. Think about it. Your staff is humming along processing invoices and then a small change happens such as someone calls in sick, or the number of invoices increases during period end or a few AP clerks get sidetracked dealing with unpleasant supplier disputes. Any of these small changes could result in a significant backlog increase - especially for larger AP operations. Using the Trader Joe's model, the best way to prevent these bottlenecks is early detection of growing backlogs. If you have real time visibility into each AP clerk's backlog, you could dynamically balance workloads or reallocate resources in real time to prevent backlogs before they get out of control.
A simple dashboard that showed each AP Processor's backlog could be a great productivity management tool.
-Rakesh
Posted by Steve Wilcox on Mon, Aug 18, 2008 @ 03:02 PM
"There are three kinds of lies: lies, damn lies, and government statistics."
Variously attributed to Mark Twain,
Benjamin Disraeli,
Alfred Marshall
and several other dead people.
Anyone who has worked with numbers knows the end results can be manipulated if one tortures the numbers long enough. Washington's smoke-and-mirrors number crunchers are infamous for this type of hocus pocus. As the cost of everyday items like gas, food, medical expenses, education, etc continues to rise rapidly, do the Government's sanitized statistics for CPI, GDP and unemployment really make sense? Of course not!!! Very few people understand that the government formulas used to calculate economic statistics are constantly changing! Yup, every year the government changes the formulas ... usually in a way to make themselves look good. I won't get into the details of hedonics and other statistical trickery but if you were to use the government formulas from 20 or 30 years ago to calculate CPI, GDP and unemployment, you would get very different and very disturbing numbers. John Williams (http://www.shadowstats.com/) contends there is rampant data distortion - the real unemployment rate is 4-7% higher than what Washington claims, inflation is 5-9% higher and the GDP is actually 5-6% lower (which would put us in a recession).
But it is not just the public sector which distorts the numbers; it is also quite common in the private sector -- specifically, corporate finance. The pressure to make the numbers and threats from bosses can result in exotic number massaging and some very creative accounting. Even with SOX, this nonsense continues. CFO.com recently published an article titled An Anatomy of a CFO's Agony which is a case study of one CFO's relentless, exhausting pressure to hit the numbers and the accounting cartwheels required to make ends meet.
Within Finance, potential skullduggery does not stop with accounting but also includes metrics -- the act of collecting data about systems and processes - and then reporting them in dashboards. Metrics allow you to track progress toward top goals, alleviate key pressures, and solve key challenges ... BUT figuring out which metrics to gather and how to gather them accurately is crucial despite Dogbert's consulting advice:
DILBERT: © Scott Adams/Dist. by United Feature Syndicate, Inc
Even worse than pointy-headed bean counters gathering metrics just for the sake of gathering metrics is gathering stupid metrics that can drive unintended behavior. Let me give you a great example in Accounts Payable - a true story of a really dumb AP productivity metric that led to unintended consequences.
I was at a company a few years ago where they actually measured productivity using a ruler. The dreaded ruler was used to measure the size of paper stacks on employees' desktops. The thinking was that if the stack was small, the employee was being very efficient. What was the unintended consequence of this metric? AP Clerks learned that if you store invoices in your desk drawer ... you become a lot more efficient. The "ruler" productivity metric was just a stupid lie and led to all sorts of visibility issues where invoices kept getting "lost" in desks, payments were late, discounts were lost, procurement and suppliers were unhappy, etc.
So what are the AP metrics that should be measured to promote the right behavior? And how and when should these metrics be gathered?
-Rakesh Shukla
@rakesh170
Related Blog Posts