Blog powered by Typepad

« Honoring Paul Hebert on the Second Annual Tim Sackett Day | Main | Beware the Reward Plan Fiscal Cliff »

01/24/2013

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Love these wonkish posts!

I resemble that remark! ;-) Well, where else will you find such a steady public display of esoteric compensation tradecraft know-how? Blame Ann for setting us free to openly share with like-minded readers. When you visit here, you ask for it.

I LOVE this! The challenge is that this represents increased complexity in a time when so many organizations are driving simplification and efficiency gains (i.e., cost-cutting). Differentiated salary progression slopes based on functional area still makes a lot of "sense". The challenges are applying this tenure-based approach in a way that does not wreack of age-disparate treatment in global contexts; measuring time-in-job tenure across organizations (e.g., experienced hires) - not to mention having meaningful merit budgets to work with in the first place. I do like the idea of a trigger though (that this inspires): 80-90% MRP initially; 100% MRP at industry average tenure. Of course, having survey providers report on average tenure would likely make the calculations much easier. *thinking ALOT out loud*

Average tenure is reported by many survey providers. For example, it is the center-point of the standard Salary Assessor table of pay by years of experience. The secret to avoiding age discrimination charges is to avoid any SandiaLabs-style maturity curves with downward curves tied to incumbent age. In reality, most long-tenured individuals in sci/eng positions see their same-job increase rates slow as they progress beyond normative market-clearing levels. As their specific peers peel off into management, hybrid roles and consulting, denuding their same-job of the more talented highly-paid outlier observations, surveyed rates tied to maturity tend to drop.

Today, the issue is critical due to the restrictions of typical merit budgets. Little money exists for catch-up progressions from net entry to median levels in reasonable time frames. So, it's all the more important to know the right starting point, when it's so hard to catch up later if you begin too low.

P.S. to William... this table summarizes the graph above:

If expected time to full competency is...

Entry/starting rate should be...

under six months
98% of the job value

1 year
96% of the job value

2 years
94% of the job value

3 years
91% of the job value

4 years
88% of the job value

Definitely agree with this:

Little money exists for catch-up progressions from net entry to median levels in reasonable time frames. So, it's all the more important to know the right starting point, when it's so hard to catch up later if you begin too low.


However, playing devil's advocate, sometimes paying a little on the front end hedges the bet on performance/expectation realization inherent within the risks of recruiting. We should perhaps be a bit more intentional in this regard though (and in an ideal world) set the money aside in a pool for later - maybe even communicate that upfront as a means to incent expected performance. But then again that presupposes that we have mastered performance measurement and differentiation, doesn't it? :)

People are better at measuring than at predicting, so I favor rationally conservative hiring offers coupled with "early increases" contingent on observation of clearly stated output outcomes. Considering the relative motivational strength of alternate strategies, however, perhaps an even better idea is to do it backwards: expect them to walk on water and competently complete every expected task but make that handsome initial offer contingent on those actual results in the first year, so that failure to meet all the objectives means their first annual increase will have been included in their starting pay. Research indicates that people work harder to keep what they have than to win a raise.

The comments to this entry are closed.