Year End Reviews

At the time of writing this, it's that time of year once again. Along with the usual joys of Christmas cheer and such, a lot of businesses are wrapping up matters for Q4 and 2023 as a whole. As part of this process, year-end reports seem to be a common theme for many. At least among the ranks of white-collar folks. In theory, a year-end review or report seems like a good idea. After all, it's an opportunity to reflect on the year as a whole and think about the goals that were set and how they went. Moving past that phase, setting goals for the upcoming year is typically the next step in the year-end process. In a lot of cases, the new goals are similar to the old goals. Dates and figures might shift but the intent usually doesn't. Here in lies the major problem. If you have abysmal goals to start with, you're essentially stuck with them forever. Making matters worse is the fact that compensation, merit increases, bonuses, and the like are usually tied to the results of these goals.

Why exactly is this a problem? Well, if the goals are good, then it isn't but for the vast majority of organizations, the goals are usually completely and utterly asinine. We should probably hop back to the basics before I continue with this rant. What exactly is a goal? The Oxford Dictionary says "the object of a person's ambition or effort; an aim or desired result". So what then is the difference between a good goal and a bad goal? It's not rocket science and a very large number of books have been written on exactly this topic. Cutting through all the crap, a good goal is something that adds value as defined by the person setting the goal. Therefore, a bad goal would be something that decreases value. If you're going to apply goals to the individual level, then I would posit one more criterion to sort good goals from bad goals. Is it within the power of the individual being assigned the goal to achieve the goal within the timeframe specified for the goal in a measurable way?

Concisely, does the goal add value and can the person assigned the goal actually achieve it? Most of the goals assigned to employees across numerous organizations don't actually meet these requirements. Consequently, they are stupid goals. I'll provide a couple of quick examples from my own career to drive this home a bit.

In my line of work, we have a daily report that contains a cross-section of quality data from the previous day. Management has made it a goal for me to read these reports on a daily basis. So far this year, I've read over 240 of these reports. On top of that, I've also documented the date and time that I read each of said reports. From start to finish, reading one of the reports probably takes about 10 minutes. 9 minutes to read and 1 minute to document. So as a consequence, I've spent 2400 minutes reading these reports this year which equates to 40 hours exactly. Let's say we bill engineering time at $50-150 per hour, so that ends up being $2-6k of time spent reading these reports. Breaking it down further, $1.8-5.4k reading and $200-600 documenting that I did in fact read the reports so no one can accuse me of not reading the reports when I say I read all the reports during my year-end review.

So, one, are reading these reports value added? Probably not. If I was tasked with evaluating a quality issue, the first step of the process would be pulling all of the warranty data from the last several years and reading through that to determine failure modes and relevant trends. The only value argument for reading the reports would be that I'm essentially building a mental situational awareness model of ongoing warranty issues. If we didn't have a quality engineering department, this would probably make sense. However, since we have dedicated staff being paid to do exactly that, having other departments doing the same thing seems like a waste of time.

Two, are reading the reports a goal that I can accomplish? Yes. I can spend the time required to read all the reports.

Taking the full balance of this particular example, reading the reports could probably go either way. If I were to attempt to get out of reading the reports, I'd probably attack from the angle that it's costing a few thousand dollars with no direct value to the bottom line. That same time could instead be spent doing project work that will improve the bottom line.

Next example. Projects, in general, have timelines. A very common goal is to meet these timelines. Beating the timelines is good. Being behind the timelines is bad. The timelines are typically pulled from the tuckus of the person doing the project planning for the particular project in question. This goal is disastrous for reasons you'll see why shortly.

So, one, are setting timeline goals a value-adding activity? It depends. Goodhart's Law states that "when a measure becomes a target, it ceases to be a good measure". The value add of a timeline is a predictable product launch. Based on the timeline, activities can be planned accordingly to ensure the project is completed successfully. It can also help with budgeting where finances are involved. The problem with timelines lies in the fact that life is inherently unpredictable. This is where Goodhart comes into play, unfortunately. When you tie employee compensation to meeting timelines, you create a perverse incentive structure.

Typically when working on a project, there are several ways that the project can be completed successfully. Take a mechanical device that consists of four parts. For these parts, you have some options in the design phase of the operation. There's an old saying that goes, "When in doubt, make it stout," which essentially means overdesign components you are unsure of. In the case of your four parts, you could make them out of beefy steel, guaranteeing that they'll work under any loading. This path would result in the quickest design solution. However, on the flip side, you could also spend a good deal of time calculating the exact loadings and customer usage profiles for the mechanical device. With that information, you could then design the four parts using the minimum required material to ensure proper function and longevity. This latter approach will take much more time.

Now, let's say you have an employee working on a project to design such a machine. The employee realizes that he or she can get a good performance review and a bonus by beating the timeline. What then should that employee do? Should the employee spend a bunch of time designing the best possible product that is optimized to ensure no resources are wasted? Or should the employee use the quick and easy stout design so they can beat their timeline targets?

Anyone who has any knowledge of human nature knows exactly what the typical employee will do in this situation. As a consequence of the timeline goal, the company in question will get value from the beaten timeline in exchange for the added cost per unit of the mechanical device that is produced in an unoptimized state. Sometimes the economics work out for this but most of the time, the business in question receives a loss for this kind of behavior.

If this were the only strike against timeline goals you might be able to use them but unfortunately it's time for number two. Can an employee meaningfully affect the timeline of a project to accomplish a timeline goal? No. The answer is a resounding no. In most cases, especially when working with external suppliers, a project is going to take about as long as it's going to take. Speeding a project up comes at the cost of cutting corners and dumping resources into it. Slowing a project down comes at the cost of leaving money on the table that could be realized by getting the product out sooner. The correct answer here is to find a balance.

My personal preference is "I'm going to do the project right and it's going to take as long as it takes, deal with it". This is the approach that maximizes value to the customer and is therefore the correct approach in a business context. In light of that, timeline goals pass the first test. It could be argued that they add value. However, they fail the second test because employees can't meaningfully influence timeline goals without cutting corners. Cutting corners results in an inferior product. Like usual, you can get away with it for a while but do it for too long and it's going to come back to bite you in the butt.

Set stupid goals, accomplish stupid things. That's the takeaway from this essay. If you want to set good goals for your employees, make sure the goal adds value to the organization and make sure the employee has a great deal of influence over it. This will get you the best results. Setting a goal that doesn't add value is dumb. Setting a goal that an employee can't influence is also dumb. Don't be dumb.