6/3/2013–Behaviors are the only things that lead to results.
Theory, culture, “support” are characteristics of a set of behaviors and should not be seen as ends in themselves. Behaviors are what managers should spend time managing. They are specific, imaginable, and most importantly measurable as inputs of data over time.
Why is compiling data about behaviors effective? 1) Because our brains are made to forget what we don’t recall with certain frequency and what isn’t attached emotionally. I.e. employee behaviors aren’t remembered after a week/month unless they’re exceptionally good/bad. 2) Behaviors over time are the elemental components of employee “performance” and a manager’s job is to improve employee performance over time. Therefore, an effective manager must regularly communicate with employees specifically about their behaviors over time and cannot do this without having a way to record behaviors for future reference.
What data should be recorded? A manager should learn in his or her own interview what his or her priority will be. It’s measurable, and this is the first thing he or she figures out how to associate with employee behaviors. For me, it’s running a “perfect shift” which features keeping FoH labor under 4.8% while increasing sales over last year. So I track BWL and PPA for my servers. Higher “beer/wine/liquor” sales and “per person average” lead to more sales than last year, and since no more staff is needed, front of house labor is indirectly reduced as a % of sales. We have a fancy computer system that records these #s for us. A manager of a mom and pop shop must develop a way of employees providing that info. This measurement is immediate and precedent to priority realization. In other words, PPA x # of servers = sales. Since my priority equation is priority = current sales – Previous year sales > 0 the PPA data is the source to find current sales that can also be used to increase individual performance. I could just look up each day’s sales and track it or develop ways to improve it day-to-day. But I can’t give effective feedback to Mark about the entire team’s PPA for Thursday. So measuring PPA for each server fulfills the need to find out if my priority is reached and to manage my team individually.
The big picture: tracking the piece of data that directly relates to my priority and is sourced individually among employees is the most important data to measure.
Forget achieving your priorities–for which you’re paid–keeping track of data related to individual performance is simply easier than not doing it. Almost all major time investment is up front. Once your excel spreadsheet, laminated chart or data-entry binder is set up, all that requires time is entering the data. It’s in the same place every time, so no time is wasted deciding where to put something. It’s saved on paper, computer, or (most-effectively) “cloud” storage so it’s kept intact exactly as you 1st entered it. I.e. no reliance on accurate memory.
It’s easy to share with teammates–#s are same for everyone–easy to bring to meetings and show, rather than explain. Data is clean, comparable across different teams and individuals. In many cases, i.e. when accurate and valid, it’s inarguable.
Objective data makes managing your bottom performer easy and your top performer exciting and easy.
Data makes distinguishing between top and bottom performers possible, communicable, and productive. The conversation sounds totally different, and talking to employees about performance feels better:
Me: Bruce, you say you’re not making $ here. I never see you selling expensive food. What can you do differently?
Bruce: I sell filet and sea bass all the time! It’s definitely not that.
Compared with data-driven feedback:
Me: Bruce, you say you’re not making $ here. Your PPA is $15 this month. Average for our team is $18. What can you do differently?
At least Bruce knows every time he sells an item over $15, he’s raising his PPA, a metric closely related to how much he takes home. Instead of worrying what the cause is, or wondering if his manager isn’t paying attention, Bruce leaves with an actionable solution.
An enormous danger with data is letting your love of it make you think the employee will respond to it or change behaviors based on “logical conclusion”. Find which data means something to the employee and use that, or apply your preferred data to what they care about. Speak of interest, not of reason.
QotW: “You’re like talking to a robot.” – Chase
10/24/2017 review: Pretty consistent with how I feel today, still. One key lesson buried here is to clearly identify what your priority will be as a manager when you’re interviewing for the position. This isn’t taught in business school, but could go a long way toward helping candidates and hiring managers both find that coveted–mythical?–“fit”.
Speaking of fit, “culture” I now believe is best defined as the aggregate behaviors of a group. Thus, managing behaviors is managing culture, in so many words.
The other idea that’s even more valuable to me today is to tailor what data I use for performance to the interest of the employee I’m managing. Some employees respond better to low customer survey scores, some respond better to verbatim feedback from customers. While I might give different feedback based on which of those my employee’s interested in, in either case, the future behavior should be better customer service.