Post #98,313
4/23/03 3:29:50 PM
|
Hey Fuat, You just delineated the PROBLEM...
to wit: [...] customers per hour rates monitored, tickets closed per hour noted, etc. This is part of (if not most of) the problem, not of the solution. If I were to call into CDW (or any other service boilerroom, for that matter), I would want the rep focusing on my problem regardless of how long it took, not his/her call-per-hour rate (which my calls tend to lower, because, I demand detailed solutions to the problems I call in with, not a cursory look down the preprinted problem sheet.) It takes longer to think and analyze that to read a bunch of bullet points off a FAQ sheet!
jb4 "Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning." Rich Cook
|
Post #98,368
4/23/03 5:39:34 PM
|
Re: Hey Fuat, You just delineated the PROBLEM...
So, don't measure performance, don't ask the affected customers how they felt about the service they got via surveys, what's left? Just spy on all the calls in real time? :-)
|
Post #98,375
4/23/03 6:16:18 PM
|
Calls-per-minute is not PERFORMANCE!
It's a measure of throughtput, which has no bearing on performance.
I suppose you still think that lines-of-code per programmer per day is a valid measure of a software engineer's "performance", too?
Puh-LEEEEZE!
jb4 "Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning." Rich Cook
|
Post #98,400
4/23/03 7:59:43 PM
|
Re: Calls-per-minute is not PERFORMANCE!
How about bugs fixed per year? Projects completed successfully per quarter? Or should a manager just wait for the "real soon now" completion promise without actually trying to measure progress. And yes, lines of code over unit time is a valid way to estimate project velocity, as is bugs opened per week, bugs closed per week as estimates of a development project's stability/maturity.
And as for call centers calls per time is a valid measurement of performance as well as throughput. If calls aren't being completed fast, customers are wasting their time on the line, which makes for unhappy customers. Of course, a blow off response to get a customer off the phone just causes more calls, complaints, etc. so there is a built-in limit to how quickly you can get someone off the phone.
|
Post #98,443
4/24/03 12:07:12 AM
|
Re: Calls-per-minute is not PERFORMANCE!
What should be measured is the manager's ability to motivate his people. Once upon a time, a manager managed, and was accountable - not countable. You make the usual false assumption that people are tunable resources.
I assume you are management - you sure talk like it.
-drl
|
Post #98,541
4/24/03 1:20:10 PM
|
I assure you he is not
(I wish we had a manager like him, but alas).
--
Inquiring minds want to scream.
|
Post #98,500
4/24/03 10:50:59 AM
|
KLOC & CPH
KLOC (for those to young to know = K lines of code) measurement example... X=IIF(Color=Red,True,False) becomes X = Blank; Rem clear var If Color = Red then X = True else X = False EndIf
I've just increase my line count 7 fold.
I had a new programmer on staff. After seeing the size of his programs, we performed a code review. Showing how to be more efficient, cut the size from 3K to < 700 lines. Did I just make him a less "productive" programmer?
How to increase CPH (Calls per Hour) Problem? Try this, if it doesn't work call back. Problem? Yuo tried it, didn't work, try this, call back...
[link|mailto:jbrabeck@attbi.com|Joe]
|
Post #98,542
4/24/03 1:26:15 PM
|
No "measure" will work
if you don't have good people. People who care about their jobs will use measures to improve the performance. People who play the games like you described should be fired immediately.
--
Inquiring minds want to scream.
|
Post #98,600
4/24/03 5:01:38 PM
|
If you can't keep up, take notes!
Once again, for the comprehension impaired: throughput != performance And as for call centers calls per time is a valid measurement of performance as well as throughput. No, guy, Calls-per-time is strictly a measure of throughput (or "velocity" as you somewhat inaccurately described it in the preceding paragraph). Performance would be the percentage of calls that resulted in a satisfactory resolution of the problem presented by the caller. Got it?
jb4 "Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning." Rich Cook
|
Post #98,756
4/25/03 1:43:08 PM
|
Re: If you can't keep up, take notes!
When I call a telephone support line, the LONGER the call takes, the UNHAPPIER I am, and thus the chances of a long call being a SATISFACTORY call are quite slim. Yes, some long calls are of higher quality, but in general the shorter the call, the better. Perhaps it is the New Yorker in me, but long dawdling conversations with sales staff, telephone operators, etc. is not might idea of time well spent.
If until proven otherwise (e.g. via complaints) you assume two equally competent operators, and measure calls per unit time, you might make the rough guess that the one that took more calls has perhaps made more customers happy... Like any other measurement, this one should not be the sole basis of an evaluation of a person's value, but it certainly doesn't hurt to know this data point. I suppose surveying customers for a more specific assessment would be an option, but you never know when that might irritate the customer and provoke a letter to the CEO.
|
Post #98,813
4/25/03 9:29:30 PM
|
I shouldn't be an expert on this..
But after some calls to Dell, and being filled-in on owner's previous history - and checking around with others about Dell's Massive Scam re non-corporate "help":
I think I Am.
It emerges that Brownie points there are awarded for "case / situation Closed" ergo -- most every call of whatever length is (most often falsely) shown as 'Closed incident'. Blame the tech? In some cases yes, but: in Dell case, it is apparent to any ept listener that these guys are Reading Scripts, that they are obv. 'encouraged' not to deviate -- that it takes real verbal and psych Skill to finally elicit an "honest reply" on some isue. Yada & yada.
This is Corporate Obfuscation by design. And it means that *next time* you call with echoes of an ongoing POS machine; You Will be subjected to Square One. In Dell's case - they'd prefer to keep you on-line for 3-5 HOURS, multiple times than: Acknowledge a Fucked product and honor their [Hah] "24/7 Service Contract". I'm now convinced of this Being Policy by - too much info to list. (A lawsuit is pending on this one).
So I think the above semantic game is irrelevant. Maybe there are few around as Bad-by-Design as the execrable Michael Dell troupe [??] but it's clear-enough that this dance of management assigning carrot/stick points as some ass-holish Tech Rating: is as bogus as a M$ "Warranty of Satisfiction".
This whole industry is staffed at the top by Real Slime, with notably (and few) brilliant exceptions. Only the color-shading of the turds - varies. Sorry for you All (who are not slime / are also not at the Top).
Ashton
|
Post #98,853
4/26/03 8:33:14 AM
|
I work at a call center
My job is supporting employees and systems so I've seen them at work with the customers and checked out the rules they follow (both official and un). Maybe it comes from being privatly owned for so long before the corp bought us, but our philosophy can be best summed up as "a happy customer is a loyal customer". As an example, when ending a call they always use some variation of "Is there anything else I can do for you?" and mean it. If there is another problem, the agent opens another ticket and works it right then. I've heard the manager tell them "If someone calls back on the same problem, we haven't done our job." So some calls might last for an hour. But the problem usually gets solved.
I think the only reason the call center hasn't been moved to India is that we are considered a (buzzword alert) "pocket of excellence". We actually get managers in unrelated parts of the business coming to visit to learn how we do things.
The world is only a simple place to the simple.
|
Post #98,910
4/27/03 1:49:00 AM
|
Glad to hear that such a place exists!
I believed that a few would get it, since that principle is so unarguable. Just rarely hear of those.
But the trouble with multi-tier like Dells - where business level actually gets service but individuals don't: even personal boycotts (after the experience) are swamped by their wholesale business numbers. Something like, "bad money drives out the good"?
Anyway, aside from the necessary drill w/ court (for mere self respect) - life's too short to make a crusade over Dell; it has lots of company(s) out there schooled to become Slime, a bit more each day - under guise of ROI, no doubt.
Cheers,
Ashton
|
Post #99,127
4/28/03 5:20:45 PM
|
Echo Ashton, and reiterate the premise of this subthread:
Throughput != Performance Hey, Fuat...Here's your counterexample! I've heard the manager tell them "If someone calls back on the same problem, we haven't done our job." So some calls might last for an hour. But the problem usually gets solved. Whoomp! There it is!
jb4 "" We continue to live in a world where all our know-how is locked into binary files in an unknown format. If our documents are our corporate memory, Microsoft still has us all condemned to Alzheimer's." Simon Phipps, SUN Microsystems
|
Post #99,178
4/28/03 8:56:29 PM
|
Behold the Liberal with Ironclad Proof! :)
-drl
|
Post #99,190
4/28/03 9:40:16 PM
|
Snicker.
Good one. As the "proof" supplier, I appreciate it.
The world is only a simple place to the simple.
|
Post #99,191
4/28/03 10:03:45 PM
|
Hmmm - then for the Illiberal - a gun will suffice? ;-)
|