The fundamental point I think everyone misses is that as organizations get larger, the network cost of running is higher and nobody has solved that problem. Smaller problems are more quickly solvable and no amount of story points will change that.
Understanding and measuring productivity can be valuable to a business. Sure they can't measure productivity for some roles as precisely, simply, and accurately as say counting the number of jelly beans in a jar.
But that doesn't mean productivity can't be measured in a way that is useful to the business. There's just tradeoffs involved.
"Y.T.’s mom pulls up the new memo, checks the time, and starts reading it. The estimated reading time is 15.62 minutes. Later, when Marietta does her end-of-day statistical roundup, sitting in her private office at 9:00pm, she will see the name of each employee and next to it, the amount of time spent reading this memo, and her reaction, based on the time spent, will go something like this: Less than 10 min.: Time for an employee conference and possible attitude counseling.
10-14 min.: Keep an eye on this employee; may be developing slipshod attitude.
14-15.61 min.: Employee is an efficient worker, may sometimes miss important details.
Exactly 15.62 min.: Smartass. Needs attitude counseling.
15.63-16 min.: Asswipe. Not to be trusted.
16-18 min.: Employee is a methodical worker, may sometimes get hung up on minor details.
More than 18 min.: Check the security videotape, see just what this employee was up to (e.g., possible unauthorized restroom break).
Y.T.’s mom decides to spend between fourteen and fifteen minutes reading the memo. It’s better for younger workers to spend too long, to show that they’re careful, not cocky. It’s better for older workers to go a little fast, to show good management potential. She’s pushing forty. She scans through the memo, hitting the Page Down button at reasonably regular intervals, occasionally paging back up to pretend to reread some earlier section. The computer is going to notice all this. It approves of rereading. It’s a small thing, but over a decade or so this stuff really shows up on your work-habits summary."
--Neal Stephenson, Snow Crash
Your employers aren’t optimizing for your personal productivity and happiness. They don’t mind if you waste 1000s of hours with poker cards and Fibonacci numbers if their business goals are met. This can mean making one person upset and less productive if it makes progress towards a bigger goal.
Just because you don’t see the progress and don’t like the process doesn’t mean progress isn’t being made.
This is a common misunderstanding about market economies vs communism btw. In market economies there's a floor to the incompetence. In communism there is none. The ceiling isn't really different, but since there's a floor in one you will get a natural average higher in market economies. Pro-market people get this wrong all the time.
Imagine if the tables were turned. Developers present in all sales meetings and asking each individual sales person how much they think the call to this client is going to take in terms of story points. And how about we also go to HR and start interrogating them how many story points it takes to give somebody their office key? Busywork might be fun, but companies can do with much less, and achieve the same, or more than they do now.
After 15 years in the industry and being both an IC, a manager, and currently a VP of Engineering, I can confidently state that the best strategy one can take to achieve their business goals is to leave the software engineering to the software engineers and step the heck out of their way.
Wasting the time of people actually doing the work by arbitrary rituals you read about in a blog somewhere is not a good strategy for success. If we must go bother someone and treat them like children and force them to play cards, let's do it with another department for a change.
I think we should do this!
> Busywork might be fun, but companies can do with much less, and achieve the same, or more than they do now.
A dangerous assumption to apply to all companies. Also the trick is deciding what "much less" looks like. Better hope you pick the right things.
I'm not saying it can't be done. It's just hard. And in some cases, you'd rather just be 10% less productive (but still productive) by following a process you hate.
Are you able to elaborate on why this is true please?
The company as a whole makes a profit, how do you know what it would have been if feature X wasn't implemented? And ten people had a role in deciding to implement it, four in implementing it.
How can I possibly put a number on my contribution?
Your number won't be exact, but it doesn't have to be and double counting among the other people isn't always bad.
That I work on the front ends was only partly my choice, we all move around depending on where we need people at that time, and other people decide what needs to be implemented.
If we were to fire all developers, we would probably lose hardly any sales the first year. But over time, licenses would go down more and more. So part of this year's profit is due to the work of people who left us years ago.
I don't see any point in claiming some specific part of our profit. People would laugh at me, probably.
I remember once being a similar conversation regarding a training program I was working on, and the customer was trying to assess the value of the training not on the improved performance of the employees, but on some measurable "knowledge unit" that had been transferred to the student (regardless of their ability to retain it).
It was beyond frustrating.
The goal is to measure people's results. It might occasionally take a relatively-long time to find a seemingly-simple solution. But if an engineer /always/ takes a long time to find every solution, and upon inspection the problems were not actually difficult, then you most likely have a low-productivity engineer on your hands.
1. solved the original problem, took 2 days and thousands of lines of code 2. changed the problem, took 20 minutes 50 lines of code 3. changed the problem in another way, took 2 days and thousands of lines but this solution solved TEN problems
Now, which one is "better"? It makes no sense.
This is daily life in every tech company. It’s true that in some cases it’s hard to accurately rank between two people (because the work is multidimensional) but in many many cases the discrepancies in productivity are obvious.
So you're saying that it's impossible to measure developer productivity at all, since it's cognitive work? Strong disagree.
All of us here that work with other developers know that some of their peers get more done than others. Or conversely, that there's a few in the team who don't seem to get much done. The minute you make even that simple assessment, you've measured productivity.
One of the jobs of eng managers is to then make this measurement as accurately, realistically, and fairly as possible. This guides bonuses and raises, which should reflect people's contributions.
I'm not defending McKinsey's proposal here, but I agree with the general notion that it's possible to measure productivity and this is something management should always be doing.
Especially because there are also skill differences and the differences in job title and salary that come with that. So the level of difficulty of the work people get to do is also different, and so on.
The problem is people realise they can gain as much from playing political games as they can from improving productivity. Once that starts everything goes to shit.
I don’t understand your position. Is it that all developers are equally productive, or that they do vary in productivity but it is immeasurable, an æther, with literally zero signals one could possibly mine to see how valuable a developer is to the team?
Soccer is a good metaphor because it's EASY to measure/see the value a good player that does not shoot a lot of goals make. In programming this is not the case.
It's certainly the case that managers that want hard numbers and don't know how to code won't be able to understand the situation at all. This is why this McKinsley nonsense is so dangerous.
"Better" isn't linear. It can multiply, and it can be highly dependent on the rest of the team.
Can you elaborate?
We take exams at school for basically the same reason.
You have to set up an institution or structured lessons to get the higher levels of cognitive measurement.
A company can use an educated guess for cognitive ability. Usually made up from completed tasks.
Selling features that don't exist is a side-effect of the company culture.
It's easy to observe the business culture in the West is often well divorced from academic truth seeking. As a matter of fact and good business.
I’m not sure you can make such a broad statement.
This argument is blatantly false. (I mean the argument not the claim it's arguing for)
As just one counterexample, writing a mathematical proof often requires a lot of creative problem solving (it's not done a lot in school for this very reason). However, checking that a proof is correct is a rote task, even automatable (see proof asistants).
You can't measure any productivity that has any cognitive component.
Devs think it's ok when sales productivity is measured like this, and then they don't understand that this is the reason sales are selling features that don't exist. We need to all stand against this madness. This goes for teachers, sales people, programmers, doctors, everything.