[bbatchelder]

Too many times in my career have I seen an employer lose an employee, who has a ton of institutional knowledge and performs adequately or even excellently, only to replace them them with someone that ends up costing much more money, and then takes a lot of time to get up to speed (which also slows the folks who help get them up to speed).

It is like they haven't taken the time to even make a rudimentary calculation on the costs involved in keeping the employee versus replacing them.

[returningfory2]

This is a common response to this issue, but I think there is a logical flaw in it.

Employers don't know ahead of time which employees will leave and which will stay. In order to improve their employee retention, employers have to pay _all_ of their employees more, not just the ones that eventually will leave. This changes the math entirely.

For example, suppose it costs 50% extra to replace someone who leaves. In retrospect, sure, giving the sole employee who leaves a 10% raise each year to retain them makes sense. But if you have to give all of your employees a 10% raise per year, just to retain that one person, it doesn't make financial sense anymore.

To be clear I'm not endorsing this system at all! But from a pure financial perspective it makes sense to me and is why, I suspect, it persists.

[andrewflnr]

It's almost like you need to identify which employees are key contributors and pay them appropriately. Imagine that.

[datavirtue]

"Employers don't know ahead of time which employees will leave and which will stay."

That is their fault. I used to work in a well managed company. It was never a surprise when someone left as career progression was always an open topic between colleagues and supervisors.

You don't know when people are going to disappear if they are crouching and ducking around every corner to avoid management.

[eftychis]

I disagree related to the SF Bay Area and the Tech world specifically. It is usually written in the wall who is going to leave if things do not change. Now you could say the upper management can't read the wall, and the middle management might not want to share what has been stated -- or they kid themselves.

[duped]

Good managers and HR leaders will recognize when people are teetering on leaving. Not always, but I've known a few who have given raises to keep people around and knew that others were planning on leaving before they announced it. In a healthy organization you can have frank conversations. Or even ambiguous ones. It's just rare.

[raducu]

> For example, suppose it costs 50% extra to replace someone who leaves. In retrospect, sure, giving the sole employee who leaves a 10% raise each year to retain them makes sense. But if you have to give all of your employees a 10% raise per year, just to retain that one person, it doesn't make financial sense anymore.

Yeah, so basically this argument is realpolitik, machiavellianism, game theory -- things that work 9/10 years, bring in profits but have enormous hidden costs that ravage companies/economies/countries in that 1/10, 1/50, 1/100 years.

The argument is "be a dick as much as you can get away with, because people are sheep and being a dick is a virtue actually". Until a lot of people start being dicks, then we cry about the long lost art of not being dicks to eachother.

[phpisthebest]

>Employers don't know ahead of time which employees will leave and which will stay

I may be a bad example, but I can assure you every employer I have ever left new I was unhappy in my role before I left.

I always have frank and open conversations with my managers about my expectations for growth, wages, etc. These are not Ultimatums, but more "In 5 years I would like to be" type conversations.

Now as a manager I have gotten similar feedback from people that work from me, sometimes you have to listen closely to understand what they are saying in reality as many are not as direct as I am, but the feedback is there

I think managers just simply ignore it in most cases

[justsocrateasin]

Yes but sometimes they do - at my last job at a large corporation I was paid 80% below the median rate for an engineer in my area/YoE/etc. - I threatened I would quit and showed them a counter offer that was 30% more than what I was currently making and they said "sorry nothing we can do" so I quit and found a better job.

[foobiekr]

I would agree, but employers actually usually do know who is a flight risk.

[peteradio]

Presumably not everyone is as valuable or risky.

[gsibble]

Oh, I've quit 3 companies where I was the CTO/lead engineer where they had to replace me with 5-10 people (backend and frontend devs, QA, devops, plus a manager or two, etc.) costing them enormous amounts of money.

Many times they've hired me as a consultant where I make 5-10x my hourly rate for several months after bringing the new hires up to speed.

Companies place very little emphasis on retention and retaining institutional knowledge. They don't seem to understand that employees who have worked at a company for years developing systems and architecture know where all of the secrets are.

[ok_computer]

Though it may cost more, spreading functional knowledge across a greater base of people would be more resilient. Even if a company needs to pay an individual 10x salary consulting for a year, it would be in their long term best interest to spread a fraction of that knowledge across a larger employee base and document what is tangible. There is a limit (different for different people) across which an individual cannot cognitively support an operation. With years more growth every person would eventually reach their personal limit or lose interest. It helps to have teams of people that are smart enough to work through new problems and document procedures, playbooks, current states and retrospective reports and rely on institutional processes to maintain a coherent operation. Sometimes money isn’t the singular factor at work.

[ryan_lane]

In most companies, it's good to let you leave, if they need 5-10 people to replace you. You take this as a point of pride, but I see folks like this as organizational bottlenecks.

When I leave an organization, my goal is for them to either directly replace me, or for them to not need to replace me at all. I do this by ensuring that I don't silo my work (by ensuring others are working with me, or under me), documenting everything I work on, and occasionally changing roles.

My specific goal is to help a company grow, not to make them depend on me.

[r930]

Not to mention the social load on integrating a new person to a team. Depending on the depth, breadth and number of interactions with other individuals, this causes others to have to also get up to speed on the new person's strengths, weakness, quirks, etc.

[lloydatkinson]

I am seeing this recently. Imagine a place that hired dozens of really good developers in 2020. Now imagine that instead of focussing on retention and paying well, they focussed on ignoring the waves of developers leaving and instead on exclusively hiring graduate developers. Now imagine all that domain knowledge that left months ago... compounded by the fact there is only five developers left from 2020...

Totally not a real story at all...

[panny]

But accounting and HR will time your bathroom breaks. I guess the watchers don't like being watched, which is why retention isn't a bother for them.

[toast0]

> But accounting and HR will time your bathroom breaks. I guess the watchers don't like being watched, which is why retention isn't a bother for them.

Maybe they should tune the company food to reduce bathroom occupancy. :P

[gsibble]

Ha. I had the CFO once call me angrily asking where I was......I was in the bathroom. She told me to get back to my desk. I took my time.