How many metrics does a sales manager need?

By August 24, 2010Metrics

As a group, sales managers are not big on “managing by the numbers.” Only a very few use more than a half-dozen or so measurements to monitor the quality and effectiveness of sales performance. Most rely on two, revenue and profit. They are the ultimate indicators of success, right? Why would anyone need to know any more?

Consider this… Assume that you just “volunteered” to manage a little league baseball team. One of the first things you need to do is come up with a batting order. Lacking any information, your only choice is to list the names in random sequence. In other words, the success of your first management decision will be based purely on luck.

Now assume that you find a list of each kid’s batting average from last year. You now have a metric and can make a better batting order decision. For example, put the kid with the highest average first, second highest second, etc.

Next assume you also find each kid’s on-base percentage from last year. (This is different than batting average. In addition to actually getting a hit, a batter can get on base by drawing a walk, getting hit by a pitch, or due to error made by a fielder on the other team.) You can now make an even better batting order decision.

For example, put the kids with the three highest on-base percentages up to the plate first, second and third. Put the kid with the highest batting average up fourth. Doing so increases the odds that your best hitter will go to bat with three runners on base, thus increasing your odds of scoring more runs. One metric yields a better decision than no metrics. Two metrics yield a better decision than one.

The scenario can continue to change. What if you also knew each player’s stolen base percentage, runs-batted-in, extra-base-hit percentage, etc., etc., etc… Each additional metric enhances the manager’s ability to make a better decision.

Shift gears and look at the big leagues.  Since the Oakland Athletics pioneered the use of metrics and statistical analysis back in 1999, their use has skyrocketed.  Over a 7 year period from ’99 through’05, Oakland won 658 games.  That’s 22 fewer than New York Yankees over the same period – essentially equivalent results.  In the same time frame, Oakland paid out a total of $295 million in player salaries.  The Yanks?  $965 million!!! Each win cost Oakland $448K.  Each win cost New York $1.4 million.  For the math challenged, a win for the Athletics cost 1/3 of what win cost the Yankees.  No wonder every Major League Baseball team now has a statistician on board.

Maybe there really is something to this managing by the numbers stuff… Maybe it even applies to sales… Maybe my competitors will continue to mange like little leaguers…  Sadly, maybe you will continue to manage like a little leaguer.

Think about it…

Join the discussion 5 Comments

  • Dave Brock says:

    Great post Todd! As much as sales professionals/managers seem to be goal directed, we are really bad at putting meaningful metrics in place, tracking them and using them to diagnose performance.

    The baseball example is great–all world class athletes measure everything–they want to know their performance, they want to understand what to tweak to improve it.

    We should hold ourselves to no less a standard. Regards, Dave

    • Thanks, Dave. Your comment, “We should hold ourselves to no less a standard,” really strikes a chord. Is it actually the case that a bunch of guys playing a kid’s game are more tuned into continuously improving performance than the vast majority of sales teams? The sad answer is YES! Every MLB team no tracks over 1,000 process metrics.

      We’ve got a long way to go…

  • Paul Castain says:

    I concur with Mr Brock 🙂

    BTW: I knew a very successful Sales Manager who would continually ask “What’s the score?” he would always follow that up with “you can’t improve things to which you have no clue”

    Nice post as always Todd!

    Respectfully,
    Paul Castain

    • Thanks, Paul. And don’t we all always agree with Mr. Brock!

      Your comments remind me of two comments from W. Edwards Deming: “If it’s not measured, it’s not managed.” and “What gets measured gets done.”

      Todd

  • EH says:

    Assume that you just “volunteered” to write an educational piece for sales managers. One of the first things you need to do is come up with a sports analogy… 😉

    In my experience, too many metrics is what sent BellSouth down the tubes. Some BellSouth executive played golf with a UPS executive who sold the BS executive (they hate being called that) on the whole Six Sigma thing. The UPS guy probably knew what would happen (UPS had been ravaged by poor metrics and draconian implementation and shortly backed way off on Six Sigma) and might have shorted BS stock (they hate it when you call it that) – or he might have been a true believer in Six Sigma in which case he would have been long on BS no matter how you look at it.

    BS rolled out Six Sigma throughout the company. They hired a battalion of consultants from Accenture with stopwatches to stand behind all the guys in cubicles who the front-line tech support call to actually make your DSL work again – the “supertechs” they were called, the most rigorously tested, highest paid techs in the company, empowered to access any system, direct any department, take as long as it took but get the line working right and reliably no matter what. And they did, and took pride in their work, did preventative maintenance on their own initiative and were so effective that rarely would anyone have to wait to talk to them.
    They sent the green Accenture “black belts” to taylorize the supertechs (even though none of the stopwatch people had much idea what they were timing, what it did, why it was being done or even who it was for). Making standard things even more standard starting from standard inputs is something Six Sigma can be good at, at least if somebody has some idea what to standardize – but unfortunately they never measured whether the DSL lines actually got fixed, (too difficult to measure) – instead they measured whether the customer was finally too discouraged to call back (and effectively treated that as the goal) and since there are an infinite number of different ways that a DSL line can break, fixing them requires different methods for different cases, individual expertise and judgment and other qualities that Six Sigma just can’t deal with. So despite having no standardized process to measure and no proper figure of merit for whether the work was being done well, poorly, or at all, the Accenture guys came up with some numbers. In fact they had 25 different metrics. They had percentages and times and volumes and trends and more, plus “team huddles” to review the numbers at the start of every shift. They added the percentages and the times and volumes and zipcodes and whatnot together, divided by a random number that made the result equal about 100 and thus came out with a figure of merit. Any suggestion that this was mathematically and practically suspect was met with anywhere from deafness to veiled threats. So the race was on to game the system. Call times were the biggest factor in the big metric, while callback percentages could only be shifted slightly no matter how good or bad a tech was.(Going from 1% to 61% callback rate was valued the same as having a call time 1 minute above the 15-20 minute average.) So unless you want to be raked over the coals every morning, get the customer off the phone as quickly as possible, no matter whether the line was fixed or not (and never mind that 90% of the time no customers were waiting in queue.) Let them call back 5 or 10 times to start the process again with a new tech – the improvement on volume more than made up for the shift in callback percentage. And even that would come back into line once the customers realized BS wasn’t going to fix the user’s line. And just in case, anybody who persisted in claiming their service didn’t work was deemed unrepairable and disconnected, and you can be sure that those numbers weren’t on the board in any negative way – in fact extra time was allotted for doing that, time out of queue that didn’t show up on the numbers.

    Constant improvement in call times followed, while the users fumed and increasingly muttered about pitchforks and torches, tar and feathers, nearest lampposts and suchlike.

    But the truth is that even with perfect measures of effectiveness, an executive class so unable to understand mathematics – or anything, apparently – would have found a way to screw things up. There is no substitute for uncommon sense, for thinking things through, for seeing and speaking the truth, for doing the right thing even when some C-level buzzword-blathering pointy-haired golf addict wants some dog’s breakfast of phony statistics. Without intellect and honesty directing things, no metrics can prevent disaster. With intellect and honesty at the top, no metric or static combination of metrics is going to be given much weight because otherwise the metrics will be gamed, unintended consequences from bad decisions that maximize the metrics and miss the real targets will follow, the lines of communication upward will be filled with BS. And you will have no choice but to join forces with AT&T.

Leave a Reply