This article is more than 1 year old
BOFH: My HELPDESK HELL - lies, phones lines and statistics
Support call numbers are PLOTTING against me
Episode 2 "I'm just a bit worried about these statistics," the Boss says, lurching into Mission Control with yet another swadge of meaningless numbers.
"Told you so," I respond.
[FLASHBACK TO A WEEK AGO]
"I'm a bit worried about these weekly statistics," the Boss says, lurching into Mission Control with a swadge of meaningless numbers.
"Told you so," I respond.
[FLASHBACK TO 2 WEEKS AGO...]
"I'm a bit worried about these weekly statistics," the Boss says, lurching into Mission Control with a swadge of meaningless numbers.
"What statistics are they then?" I ask, desiring only to be of service in relieving the Boss of his worrying burden.
"The phone system call log statistics," he replies, gesturing at a column titled "Systems and Networks".
"Oh yes?" I ask with interest levels inversely proportionate to the enthusiasm I'm projecting.
I bloody love statistics. Not.
"Uh, the abandoned call and unanswered calls numbers. They're really quite worrying."
"Why do they worry you?" the PFY asks, in a thoughtful and engaged manner that could only be enhanced by the presence of a pipe and a comfy couch.
"Because they reflect badly on the department," the Boss says. "Our department average for abandoned calls was 8 per cent, but with you two included is 37 per cent and our unanswered called were 11 per cent but are now 51 per cent!"
"I think the reflection's all about how the figures are presented," I say. "Perhaps you should take a leaf from the Health System's book and instead of using a phrase like 'People who died on the waiting list for an operation' use 'People whose date of death preceded their date of operation'."
"How's that different?" he asks.
"Because in the first case the onus is on the health system for not getting the person to an op, but the second case is definitely the patient's fault for choosing to die before they could be operated on," I reply.
"I'm not sure I'm with you."
"Why not title columns 'Calls where the user solved the problem before needing technical attention' instead of abandoned calls, and 'Calls where users were given the option to seek an alternative solution' instead of calls we didn't answer?"
"Yeah, I don't think that would work. It's a bit transparent," the Boss sniffs.
"Only if you look carefully - and no one does that with statistics," I respond. "Anyway, what's the margin of error?"
"What?"
"The margin of error. You know, like when they take a phone poll and tell you how accurate it is. It could be that the margin of error is so high in these stats that we're actually doing a bloody good job."
"You're not, and no one cares about the bloody margin of error, they only worry about the numbers. I think we should just try to improve our service," the Boss snaps.
"You say that now, but I know you'll be back in a week with more problems..."
[FLASH FORWARD TO ONE WEEK AGO]
"Yes, but I thought you were actually going to do something about the numbers?" the Boss asks.
"I did," I confirm.
"Changing all the row names doesn't help," the Boss barks. "Changing a row to 'calls abandoned when the user realised they were being a slack lazy bastard who should have done what they were told two weeks ago' doesn't help. Nor does 'calls where the user had insufficient information available to make a successful support call because the internet is running slow is a stupid thing to call up about'."
"Are you sure it doesn't help?" the PFY suggests.
"No. And I'd like it changed. I want these numbers >gesture< down to acceptable levels," the Boss says.
"OK, but you'll only be back in a week wanting us to make more changes."
[FLASH FORWARD TO TODAY]
"What's wrong with the statistics?" I ask.
"They're zero," says the Boss. "According to the numbers you answered 100 per cent of your calls last month, so you got no one hanging up inside the 1st wait period let alone the 3rd wait period."
"You're welcome," I say.
"Yes, but I don't believe it, because I rang you and you didn't answer," replies the Boss.
"Sometimes we forward the phone to another number when we're working away from the office," the PFY interrupts.
"Oh. I must have misdialled the forwarding number," I mutter.
"Three days in row?" the Boss asks.
"Possibly."
"Misdialled a 4-digit extension number with a 9 digit external number."
"Yes. I lose count sometimes," I sniff.
"So these figures are a complete sham?" the Boss sighs.
"No, someone answered so technically we're covered. AND they have 24-hour service too, so it will really lift the department average!" I say, triumphantly.
"It's all rubbish!" the Boss cries.
"No, it's statistics with a very high margin of error - which no one cares about," I remind him.
"Well we can't keep it like this!"
"Why not, who's complained?"
"No one's complained, but the data isn't right. We should be reporting correct data. I WANT it fixed and I want the calls answered," he blasts.
Sigh.
Meaningless bloody statistics will be the end of me.
Not as soon as it'll be the end of the Boss, though, I think, as I watch the PFY prepare to increment the "Projects for which the proposer had a serious workplace accident prior to implementation" figure. ®
Biting the hand that feeds IT
