Hacker News Reader

Comment by subway | Hacker News Reader

> Some of these facilities have thousands of employees. A few hundred firings in a year seems, at a glance, plausible.

The article mentions these numbers are for a specific warehouse, and provides the total number of employees at that warehouse (~2500).

If you're firing over 10% of your workforce annually for performance, you have fucked up expectations.

View on HN

One in ten per year sounds about right for these sorts of jobs. I bet if you drilled down you'd find that approximately that same 10% of people is getting fired from every other jobs too. There's always a minority that is screwing up fairly often. Combine that with rigid bigCo policies on timeliness, performance, etc and you can see how 10% per year get fired.

I wonder what the rate is at places which skew towards similar entry-level positions; for example, what % of people get fired annually in the fast-casual restaurant industry, or in retail? I don't know.

I do think that in these sorts of industries, it's important to be able to hire people with a "probationary" mindset, because if firing weren't possible an employer would become a lot more risk-adverse. It's actually really important to people with no credentials or who might have a risky profile (a criminal conviction, for example) that they be allowed to prove themselves and start building a resume, so we should be careful about incentives in these cases.

Doesn't "rank and yank" traditionally advocate firing the bottom 10% of your workforce as part of the annual review process?

I'd personally consider the whole concept of a vitality curve to be rather fucked up.

It's been all but abandoned by the biggest names behind it (GE, MS).

I don't think the fact that some companies have abandoned it means much.

GE's peak days as a company were when they enforced a high bar using a curve. The same is true for Microsoft, whose better days ended around 2005. Although their stock performance in recent times has been strong, it is largely riding on the refresh of contracts relating to Office and the transition of clients from on-premise to cloud. There has been little actual innovation from the company, and although the culture has improved under Satya Nadella, employees regularly talk about workplace politics, low-quality hires, etc.

I would disagree re: MS. They are doing pretty good in enterprise space, their bread and butter. Also whole XBox ecosystem is getting stronger as well

> If you're firing over 10% of your workforce annually for performance, you have fucked up expectations.

Or a seriously broken hiring process.

How do I properly filter applicants for low-skill jobs? If I'm hiring for an Amazon warehouse, I'm not going to ask for a resume, and background checks only catch certain things. How else am I going to know you'll be a good worker in the warehouse other than hiring and seeing how it goes?

>How do I properly filter applicants for low-skill jobs?

You hire anyone with a pulse and fire anyone who's performance is not satisfactory. If you're Amazon's size you implement some sort of system using mostly objective metrics (e.g. a no-show counts as X points but a no call no show counts as Y points and you get fired at Z points) so you can quantify performance then you figure out your churn rate and size your pipeline to compensate.

This is all post employment. It’s not a predictive hiring model which is needed but pretty much impossible at this scale. Sounds horrible but it’s like asking for a model to determine which bums will use your charity for food instead of crack.

More than that, for low skill jobs you don't want strong filters. If someone comes in and says "I was a slacker who never showed up, but now I'm reformed", "I got fired for being drunk at work but I learned my lesson", or "I just got out of prison for some awful crime and want to try honest work": you should hire them just on their word that they are reformed. Of course in all cases their word isn't worth much, but these are the first jobs of those who actually are reformed and you need to keep giving them fifth chances. If we don't slackers, drunks, and criminals need to be supported by society in some other way, which becomes a reward for bad behavior.

So for low skill jobs you hire anyone, fire the ones who aren't good workers and promote the rest.

“We only hire the best, and 10% of you are losers”. I never managed to square that circle

Not sure this is their thinking, but here's one way to square those concepts:

A person can have stellar past performance, but turn out to be a poor fit for the particular culture, workgroup, or work of the new employer.

It's not that difficult. The best would be those hires able to rise up and maintain enough productivity to keep them out of the bottom 10%, who, by definition, are the losers.

That's an easy one. The bar is constantly being raised, so you were the best when hired, but now you're not. :(

Netflix fires almost 10% of their workforce annually.

Netflix also compensates the ever-loving fuck out of you up until that moment.

I'm fairly certain most jobs at Netflix aren't low-skill warehousing.

Don't be so sure, they still run their disc rental service out of 17 distribution facilities. They added a lot of automation in their heyday but there's still a fair amount of low-skill work to be done.

https://www.cnbc.com/2018/01/23/netflix-dvd-business-still-a...

Yes, but for any given job at Netflix, you go in knowing other most gigs are going to pay at least 30% less for the same work.

Amazon isn't paying their warehouse employees a premium rate to offset the risk of falling in the bottom 10%.

Do you have proof??

> I'm fairly certain

I bet Fedex and UPS have a higher churn rate. I quit after the second day working in the hub.

Remember, this isn't the churn rate we're talking about. It's involuntary separation. I'd imagine the annual churn rate is way above 10%. I don't doubt FedEx and UPS have high churn rates, but I also suspect they don't fire 10% of their package handlers annually.

That 10% fired doesn't adjust for churn tho. Over the course of a year there were at minimum 2500 employees working there, we don't know how many unique people worked there over the course of a year to find out their firing rate.

I work for a mid-size manufacturing company, and we fire (or they quit) way, way more people than we keep

UPS does a lot of seasonal hiring during Christmas. This includes workers in distribution centers as well as helpers on delivery trucks.

I don't think companies really care. I have seen mass quit recently on a well oiled machine. Tech force, bad management practices. Nothing really changed and no one seem to care. In the end the quit rate was only increasing (the time that I quit)

Reply on news.ycombinator.com