Comment by Spooky23 | Hacker News Reader

Comment by Spooky23 | Hacker News Reader

I worked in a sales environment like this for a summer break in college. I was the number 2 sales dude that summer, hitting all of the BS metrics that they fussed over. (ratio of premium to total sales, ratio of services to total sales)

But... they lost a fair amount of money on me, because I abandoned profitable sales that wouldn't hit my metrics, dumping them on others. Instead, I ended up moving lots of stuff with high price concessions. But I never sold anything without the metric-impacting accessory purchases. So I was #2 in gross sales, #1 in hitting targets, but #28 in profit. I made a ton of money though... paid for school the following year!

So, I'd say good luck, it's hard to know what to measure for when your business leaders are busy tracking metrics instead of leading. If I owned a restaurant, and the GM was fussing around with cameras and computers in the back, I'd fire him. Service business management is about taking care of customers and setting up systems to ensure that happens. Lead from the front.

You'll enjoy this paper by the Harvard Business School which pretty much destroys the concept of systemic goal setting in a business. It describes your exact situation: when people have to deal with shit goals, they figure out how work the system.

http://www.hbs.edu/faculty/Publication%20Files/09-083.pdf

Original discussion: https://news.ycombinator.com/item?id=7897662

It's like the Office Space version of arbitrage. Where the rules lead to unintended opportunity, there's space for exploitation.

(See also: frequent flyer programs)

The problem here is that they used stupid secondary metrics rather than a single primary one. As you note, there is an obvious primary metric to use - how much profit you made. It certainly appears measurable, based on the fact that you know you didn't improve it.

If he was getting new customers, the profit on the first sale might not have been what mattered. It might have been, but it might not have been.

But that only values short-term profit which is not necessarily the best for a business. There is a problem in simplifying the value of a worker to a single number. Always.

If you can't reduce your options to a single number each, your decisionmaking process will be either inconsistent or subjective. This is elementary topology - I sketch a proof here:

https://news.ycombinator.com/item?id=7404964

So which are you advocating? Inconsistency, subjectivity or both?

It's certainly possible that Spooky is wrong, and the right metric might have been a time-discounted approximation of long term profit, rather than simply short term profit. I doubt that - most companies tend to overvalue the long term. But that's a secondary point.

In terms of your sketched proof he's advocating that "<" ">" and "=" are not defined over the set "business metrics" so you can't assume a total ordering.

If you don't have a total order then there are decisions you simply can't make.

No there's just decisions that you can't make well.

The equation A < B holds anytime you would choose B over A (and vice versa). If neither A < B nor B < A, that's the definition of making no decision.

If you advocate making an arbitrary choice, your decision process is either a total order (if you don't break transitivity), inconsistent (if you do break transitivity) or subjective (if the choice is nondeterministic).

Noticing that reality is under no particular obligation to be sortable is not the same as advocating a particular strategy to deal with it. It's just admitting that yeah sometimes subjective decisions (gasp?) will be made and yeah sometimes inconsistent decisions (gasp?) will be made.

And as long as you aren't an evil computer in Star Trek:TOS that's not actually a problem.

It doesn't mean subjective or inconsistent decisions have to be made, it just means you need to learn to get comfortable with degrees of uncertainty.

Reply on news.ycombinator.com