Hacker News Reader

Comment by danpalmer | Hacker News Reader

I know there's a common theme to look at the sizes of tech companies at times like this and ask what do all the people do, but, bear with me here... what do all these people do?

Just taking a few other data points, Twitter (when it was Twitter), had around 7k people, and strikes me as a similar sort of product to LinkedIn, at least the same ballpark. Google has somewhere around 180k (according to public info), and has many products that feel like a LinkedIn sort of size.

Whatever you may think of my other examples, LinkedIn really does seem like an outlier? Am I missing a major aspect of their business? It's a social network, a recruitment service, an ads business. Is that 22k people?

View on HN

I'm sub-contracted to a blue chip manufacturing Fortune 250, who everyone has heard of. Before COVID, I was strolling through the cubicle warren, and really noticing all the things pasted on to every carpeted vertical surface. No matter where you stood, you could see 2 or 3 copies of every notice: local office news, flyers about insurance, community volunteer efforts, secondary education programs, etc.

I suddenly realized that, in this company of ~60K employees, there was a veritable small army of people who were involved in this effort: designing, printing, having printing done, hanging, taking down, etc.

All of that is gone now. Some companies are getting through this economic turndown by simply slowing down hiring, and reassigning people from jobs about which people ask "yeah, but what do these people actually do?" If we get through the end of next year's first quarter without a layoff, I'll give props to senior management.

Big companies are an organism in their own right. I’ve particularly noticed this with HP, HPE and IBM.

All of those are US originated and, I suspect, are like mini-countries for a reason. That reason been a lack of a social system in the US. The company provides those offerings that you would expect by a European government.

Hence, you have the HR emails, posters on the wall and so forth.

From the link, LinkedIn does $15B in revenue, or $681k per employee with 22,000 people. That seems pretty reasonable to me, even though I don't know what all of those 22,000 people do.

This is a great way to think about it, thanks!

This sounds roughly in line with Twitter's (circa 2021) at ~$700k, but a long way from MS/Apple/Google who are all in the $1-2.5m range. Even Amazon is at ~$500k and has many low paid workers, so would likely be much higher if focusing on the tech/product/etc side. It's not terrible, but maybe a little low?

How exactly is that a good way to think about it? What proportion of employees at a tech company are tasked with bringing in revenue and why would that be expected to scale linearly?

Google is a terrible point of comparison because we have plenty of direct evidence that thousands of people at Google are doing nothing of any consequence to the business

It's a rough measure of business efficiency. You're right that many people aren't directly related to bringing in revenue, that's inefficient (although necessary at some level), and so they change the denominator and reduce the overall amount.

The fact that Google has ~$1.5m revenue per employee suggests that it's a more efficient business than LinkedIn's at $700k.

No it doesn't suggest that at all lol

How so? Open to being persuaded here, I only have a basic understanding of this, but you've not provided any alternative?

The real way to look at it is via the marginal revenue minus the marginal cost.

Generally speaking, it makes sense to keep hiring more workers if you're getting a net benefit from hiring each one. That being said, at a certain scale the organizational challenges may eat into your net marginal profit. (eg unclear ownership and diffusion-of-responsibility, more layers in the org chart making it harder for decision-makers to hear from the people actually doing the work, longer deprecation cycles, etc)

Hiring is hard and slow, so it's definitely possible that the other companies you used as examples were theoretically understaffed. (if you ignore the costs of hiring, anyway)

Apple is an outlier because they sell hardware, which carries lower margins than pure software/information services. LinkedIn at $700k vs $1.5m at Google (perhaps one of the best business models ever) doesn't feel bad! For comparison, here are RPEs for other large successful tech firms:

- Oracle: $310k

- Salesforce: $416k

- Intuit: $830k

- Adobe: $645k

- Workday: $317k

- Meta: $1.5m

Studying these numbers confirms that Google and Meta are indeed extraordinary businesses. Which is not to take anything away from LinkedIn, as they operate in a different market.

More interesting is the hypothetical question: if you were in charge of LinkedIn, how would you prioritize achieving an RPE near the top of the industry against other priorities (perhaps such as increasing market share)?

And to your point, Google and Meta are indeed extraordinary businesses. That’s why everyone wants/wanted to be Google or Meta.

Even being somewhere in the ballpark of a Google or a Meta is incredible.

I don't know about LinkedIn, but a comparable company would be Indeed - something like 12-15k employees. The majority of the company is sales & marketing. Their customer base is basically every company in the world, so having each sales person nurture a few dozen to a few hundred relationships is not uncommon.

But how can you pay the ceo millions when 22,000 other people need to make a living wage?

Resilient systems are rarely 100% efficient. Duplication of key infrastructure and roles maximizes uptime. While I'd bet LinkedIn suffers from large-scale organizational inefficiencies, each team probably is working to maximize resiliency.

Sometimes inefficiency can help retain talent. Sometime in the 1950s, Thomas Watson Jr., the second president of IBM, wanted to retain a bunch of employees at the Poughkeepsie location (which is still an important location for IBM to this day), but IBM didn't have enough work at the time to justify keeping them. According to the story, Watson told these employees to redo the flooring in a massive Poughkeepsie building (Building 012, to be precise) and to take their time. The tactic worked; the flooring was redone (think school gym flooring), and work eventually picked up again for those employees. I heard this story from long-time employees at the IBM Poughkeepsie location.

A now-dead comment says that LinkedIn is way more complex than Twitter, and I'm open to being persuaded about this, it could well be true, it's not obvious to me that it is though.

Sure there are more buttons and inputs, the product does a bit more, but when talking in terms of thousands of employees these sorts of things don't tend to matter a lot, and instead I'd focus more on whole services/businesses, and major complexities such as scale, international businesses, etc. These seem pretty much the same for Twitter/LinkedIn?

You can click on the time of a “dead” post and click “vouch” if you think it’s okay.

New user posts are often “dead” by default and waiting for a long-tenured user to bring it to the front.

> it's not obvious to me that it is though.

Their recruiter portal stuff is pretty complex, which doesn't really show up in the twitter comparison.

I find it just as humorous, every time this question is asked, that people throw out the "it's just a X, Y, Z" without ever thinking too hard about the roles needed, beyond engineering, to fulfill X, Y and Z at scale. X, Y and Z can usually also be broken down into many, many subparts all with their own supporting cast.

Yeah I get that, but I think there are two sides to this.

One is that it is possible to guess ballpark figures without knowing the intricacies, and I think anyone with experience in a big tech company of this ilk can probably make a reasonable estimation (to within 50%?).

The other is that it's easy on the inside of these situations to justify each individual role, and miss the fact that the system is too complex and a simplification may be able to drop significant numbers. I'm not suggesting it is too complex, I fully realise there's a different sort of scale involved when you operate a large globally available service like this, but duplication still happens, engineers still make things more complex than they should be, etc.

LinkedIn for Business makes LinkedIn much more than Twitter in respect to product offerings and in terms of support, marketing and sales as those are much more important when dealing with business customers. All of those products need to be developed, supported, updated, etc. The teams doing that however also need to be managed, so additional HR, Managers, etc. I think it is just that all of these different departments quickly add up to quite a significant amount of people.

A significant amount of the enterprise SaaS company is sales and marketing.

As for engineering, a significant number of people are required to keep various services running. But more importantly, a lot of engineer time is wasted in optics and metrics for management.

If management is culled and their way of working is changed from measuring dumb metrics to keeping things steady - the number of engineers required will reduce, albeit only a little.

Anytime you’re a software company, you can do an equation of “we pay X millions a year for service/product Y. Could we spin up a team and do it for cheaper?”

Large companies are often replicating what other companies just buy because it works at their scale.

Don't know of linkedin, but I remember Twitter went full on to Scala, then decided to create documentation and learning resources for it. Then proceeded to work on its compiler.

Though LinkedIn wrote Kafka, which from afar seems to be a better investment.

The real problem is that too many people has a preconceived notion that profit-driven corporations are economically rational. No, incentives to grow in size are everywhere from individual middle-manager careerism all the way to boost-numbers-to-look-good-for-wall-street exec careerism, like good luck having a juicy IPO on a 50 man crew.

WhatsApp got $19 billion with 55 employees.

and it was bought by Facebook, it didn't sell itself to the superficiality of Wall Street. Do you think it would've able to get there with 55 employees? I do not. Do you have an example of that?

The only use for the linkedin web page is keeping in contact with current or prior contacts. A dozen people can do that job.

It being comparable to the headcount of others still does not answer the question of what do all these people actually do.

Not to entirely rebut your question, but there is also an online education business (LinkedIn Learning/Lynda).

Reply on news.ycombinator.com