Avoiding Analysis Paralysis

To effectively manage reward programs compensation practitioners and their senior management need to understand how competitive those programs currently are.  In making that determination, though, just how precise does the analysis have to be?  To what lengths should one go to increase the level of exactitude in the analysis, and is that effort worthwhile?  Does the effort to squeeze out greater precision bring more meaningful results?

In other words, what's the additional value of dotting the "I's" and crossing the "T's"?

As compensation professionals will tell you, the competitive “marketplace” for reward program surveys is an imprecise animal, subject to numerous variations and interpretations.

• One survey doesn’t use the same companies as the next survey.  Who do you believe?
• The ability to match job descriptions varies from precise to broadly similar roles.  Strange titles are a constant challenge.
• Surveys provide different mixes of industries, geographies, and revenue, and cannot report on the many non-participating organizations. Rates of organization growth are not reported.
• The use of weighted Average/Mean/Median/50^th percentile formats isn't standardized.
• International surveys often fail to provide enough information, as compared with U.S. surveys.
• The process of completing survey questionnaires is often not as carefully conducted as one would hope (perhaps an understatement).

In addition, the market is a moving target as population changes, organization shifts and changes in employee pay are considered.  This means that you’ll need to be careful of “aging” factors, such as adjusting data from date of collection to the current or some future date.   My practice is to "round" to the nearest 100 units of annual currency.  Is that distorting the data?

Pick a Number?

Do you think that a market rate of $47,570 for a job is an accurate reflection of current trends, or simply an arithmetic average of data that looks precise?  Would you fall on your sword over that figure? 

Using three surveys would likely provide three separate figures.  Let’s say your sources report $45,723, $47,612, and $49,375.  If cost, time, and effort are not factors to consider, then you could keep going, searching for that common denominator.  Purchase another (expensive) survey source.  Double and triple check your job matches.  But do you think that the extra source, the extra time and effort will substantially change your initial analysis, or are you simply looking to protect yourself?   Are you playing a defensive game to divert potential challenge?

For most of us a quick and straightforward analysis suggests that approximately $47,500 is reasonable enough.

While not advocating a sore thumb analysis I do suggest that you use a balance of time, effort and cost when conducting market analysis.  What you really need is to understand the market; that’s the key learning point.  Repetition is sometimes just that, more of the same with little increased value.

And what degree of precision are you being asked to provide?  Are your instructions to feel the pulse of the marketplace, to get a sense of what is being paid out there?  Or do you need a more precise result?  Is your audience demanding more?

Is the Market a Number?

What is that market?  Anything within +5% to -5% of a “market rate” figure is close enough for me.  Others believe that variable should be 10%, but in my view that leaves too wide a range that could distort your intent to provide the so-called “going rate” trend.

Caution:  You can find any number of analysis paralysis jockeys out there who advocate increasingly precise techniques to zero in on what they call your true market rate.  Just remember that many commercial survey vendors have built a business around encouraging organizations to slice and dice whatever information is available, trying to define and refine exactly what a “market” is paying, what jobs are exact matches and after a fashion what numbers you can rely on.

Part of their marketing strategy is to use custom designed evaluation techniques and proprietary job matching systems.  Use of such a strategy effectively marries your organization to the vendor, as one often cannot use proprietary language and techniques (the system) with other survey providers - without the risk of comparing apples and oranges.

The Leadership Perspective

From a senior leadership perspective, do you really need that depth of precision to make a business decision?

I think you don’t.

There's a place for precision, and a place for trends - for a “feel of the pulse.”  Usually, senior management wants to gauge the big picture - the overall strategy and its implementation status - letting professional specialists deal with the tactical matters and details.

Have a care in your efforts to be precise, because when you give senior management too many details they tend to dive in almost as a defense mechanism, as if they are expected to ask questions. Where they might otherwise have nodded their heads at key points in your presentation, you can instead find yourself immersed in detailed analytics that can bog down the decision-making process.

When senior management asks a question, office life becomes an “all hands-on deck” fire drill of employees rushing about to get the answers - to questions that might not have been asked in the first place if you had played your cards right.

Management wants your professional judgment.  They don’t want you to defer opinions to what a survey says.  Use survey data as a backup, as anecdotal information to assist the decision-making process.  Don’t feel that you always need to lead with it.

Standing back and pointing at figures is like leading with your chin in a fight. 

You won't like the results.

Chuck Csizmar CCP is the founder and Principal of CMC Compensation Group, providing global compensation consulting services to a wide variety of industries and non-profit organizations.  He is also associated with several HR Consulting firms as a contributing consultant.  Chuck is a broad-based subject matter expert with a specialty in international and expatriate compensation.  He lives in Central Florida (near The Mouse) and enjoys growing fruit and managing (?) a clowder of cats.

Creative Commons image "Forgetting," by Leeks

Comments

[Bob Nadel]

Great perspective Chuck. Comp consultants are not research statisticians, but advisors who use best judgment data to describe a market.
There is a degree of precision that goes beyond value added to the decision maker.