Companies by revenue per employee

I was surprised by some of this:

We found that Energy companies have the highest average Revenue per Employee, while Industrials and Consumer Discretionaries perform worst on this metric.

Technology companies performed at the lower end of the range on Revenue per Employee; part of the reason for this however, is other companies in spaces like Energy and Healthcare have large non-employee costs that Technology companies do not have.

AmerisourceBergen, a pharmaceutical distributor, tops the list, generating more than $7.9M per employee in 2016. With a reported team of 19,000, which is less than half the workforce of Cardinal Health (37,300) and McKesson (68,000), the company compares favorably to its peers on revenue per employee. Cardinal Health and McKesson‘s RPE were $3.3M and $2.8M, respectively. Overall, Healthcare companies score well on revenue per employee, though they have other huge costs (the costs of administering drugs and health services).

As for the lowest revenue per employee:

It is perhaps unsurprising that Restaurant and Hotel chains make up the majority of the list. What is more striking is that IT providers Cognizant and Accenture have among the lowest revenue per employee in the Index.

There are several useful tables at the link.  I do not think this is making any adjustment for independent contractors.

For the pointer I thank the estimable Chug.


Definitely not surprised on the energy side. I keep seeing how the wind and solar industries now employ so many people with a tiny fraction of total energy production and think "how do they make any money?"

Energy jobs tend to be pretty difficult - long hours, poor job security in sectors like upstream oil and gas, and subpar working conditions. Wages have to be high to attract workers but labor is still often short so there is a very large incentive to increase productivity. Also, lots of labor is contracted out in capital projects. Not sure if that is accounted for in the methodology.

This has nothing to do with the green energy "economy" and everything to do with the standard energy companies being very capital intensive.

Revenue per employee isn't the greatest metric, but it is interesting. I think profit per employee is more in vogue (energy still scores well).

Agree. It's the capital costs. The labor is in building it with a very small labor force after the fact. Solar can be extreme this way. A smallish 5MW site costs $10MM with less than 100 employees building over a couple of months. When complete, it will be remotely monitored and have no permanent staff assigned specifically to it.

That's what I mean though. With less than 15 workers on site at any time a $10 million dollar well can be drilled and completed in less than three weeks. Over thirty years a natural gas well in PA will make like 20-30x more energy, also with very little daily labor.

Like I said, how do they make money? No wonder they need subsidies. Though that gives them lots of scope for reduction in costs, as we are seeing.

Gas provides thermal energy. Wind and solar farms provide electrical energy which is considerably more valuable.

And do new US gas wells produce for 30 years? I would have thought all conventional gas sites would have been well exploited by now, leaving short term coal seam and fracked gas extraction as the typical new well.

New combined cycle plants are more than 60% percent efficient and can be sited near cities. So you are still looking at like 15x if it is all used for electricity that isn't going to resistive heating.

That is cumulative production after 30 years, to give solar a better chance. Most of the gas will come in the first few years, of course. Even if most of the gas comes in a few years, 30 BCF out of one well (8+ wells per square mile!) is an almost unfathomable amount of energy, when 1 BCF per section from a deep vertical well used to be a big deal. But that is today's technology.

This Forbes piece notes that the solar jobs count workers supported by the industry while the oil and gas jobs count people directly employed in extraction.
I wish this piece gave the apples-to-apples numbers instead of just pointing at apples and oranges.

Aria de Capo?

Think about value added per employee, and all makes sense. Pharma distribution has a 4-5 gross margin, energy is almost all gross minus royalties and outsourcing, and tech is pretty much 100% gross.

Yes, it's amazing they could write that and not use the word "margin". It reminds me of the first .com bubble, when companies that were really just taking a small fee for a service/marketplace would try to characterize it so that all of the revenue from the associated transactions was theirs (albeit at a tiny margin that would never get better with scale).

Revenue per employee is just profit per employee plus wages per employee plus non-wage costs per employee. Makes perfect sense that that last one should be very high for energy companies and very low for IT companies.

If a company cannot earn $15 per hour per employee plus profit, that company does not deserve to exist. Besides, who even eats at those kind of restaurants or shops at those kind of stores, gross.

I'm glad you aren't in charge of who gets to exist.

I think that was sarcasm.

I think that was typical liberal logic.

It was sarcasm, and was likely directed at a comment from the thread about guaranteed jobs since someone in that threat did make almost exactly that comment in a serious way.

Given that retail companies take most of their revenue and shift it onto other companies - i.e. many companies have massive externalised employee costs - I fail to see why this data is at all interesting

What are some ways one might use this metric? It doesn't seem all that interesting to me, since it has less to do with productivity of employees than with overhead and COGS.

Agree. Better to use revenue per employee as a measure of productivity within an industry than across. Nothing to see here.

Refiners have high revenue per employee because they book revenue on total gas/jet fuel/etc sold, but they must buy an entire barrel of crude to create that revenue, so gross margins are low. Refining is also much more capital-intensive than labor-intensive. Medical device distributors have a similar dynamic (high revenues, low gross margin)... AmerisourceBergen's gross margin was 3% last year.

Not sure if any of this warrants surprise: You would expect to see highly capital-intensive companies with high turnover at the top of this list with highly labor-intensive companies with low turnover at the bottom.

I think this measure is, exactly, the measure of the productivity by employee (or the firm-level version of the GDP per capita)

Drug dealers can have high rates of return? Who knew? - at least when distributing in America. It is reasonable to believe that such a high cost middleman does not exist in any other industrial nation's health care systems, all of which cost at least a third less than America's.

That this text appears in connection with AmerisourceBergen Corporation is not a surprise at all - 'the Company provides drug distribution and related services designed to reduce costs and improve patient outcomes.'

Further, it appears from this - 'With more than 19,000 associates and over 150 company-owned offices around the globe, AmerisourceBergen is improving product access and transforming the healthcare supply chain every day. Its 26 U.S. pharmaceutical distribution centers, 4 U.S. specialty distribution centers, and 2 Canadian distribution centers to maximize supply chain efficiencies mean AmerisourceBergen is well equipped to provide its customers with the knowledge, reach and partnership to shape healthcare delivery.' - that AmerisourceBergen buys globally, but only sells into North America. Which probably works well in connection with proclaiming this - 'As part of the largest global generics purchasing organization AmerisourceBergen is well positioned to help customers capitalize on the dynamic changes in healthcare.'

"Drug dealers can have high rates of return? Who knew? – at least when distributing in America. It is reasonable to believe that such a high cost middleman does not exist in any other industrial nation’s health care systems, all of which cost at least a third less than America’s."

The USA has too much commitment to science? We could check the supplement industry to find out.

'I thank the estimable Chug'

Nice job using the thesaurus.

"What is more striking is that IT providers Cognizant and Accenture have among the lowest revenue per employee in the Index."

Not that surprising. E.g.: Accenture has an army of outsourced developers in India/Philippines etc. that perform services through an on-shore service manager that speaks to their clients. Revenue/employee does not make too much sense for that model.

I was going to say the same thing. Accenture, like most other big IT consulting firms, uses low-cost labor wherever they can. They offshore as much as possible, use H1Bs workers from India in the US. I've seen them use Spanish workers in Europe (Spanish workers have a lower cost-of-living than Northern Europeans so they have lower salary demands).

But, really, that's just the market right now. IT consulting is in a weird spot. There is downward price pressure from Indian consulting firms. At the same time, many companies are outsourcing their IT departments and moving to Cloud solutions for their business software. This means that businesses are moving away from customized solutions for their own businesses. Without the internal staff or architecture to support customized or home-built software solutions, the need for IT consultants starts to slip away. Some of the consulting companies are now moving into the outsourced IT department business, which is really not good for the revenue per employee metric, since that requires 7 by 24 staffing.

Spanish workers don't have significantly lower cost of living than northern Europe: What they have is a local market that is in shambles: High unemployment and a lot of local firms that get contracts through corruption, not quality, so costs are all the employer cares about. Therefore, you find a lot of 30 somethings that live with their parents, so you can pay them little and they still come to work. The cost of living in middle America is very similar to Madrid, but a firm in Kansas City will start people out of school in the 60s or 70s, instead of 15-20k.

This is why I laugh when people talk about the economic miracle of Spain after the crisis: It's really hard not to get catch up growth when you have a college educated population. But Spain's crazy labor laws, corruption, and weak justice systems do their very best to keep the country down.

As far as big it consulting goes, yes, they are in a weird spot all over the place, but it's not necessarily because of less customized solutions, but because their quality is so weak that running in Amazon with no support is not just more cost effective, but is a far better experience. It's pretty crazy, given how expensive cloud offerings are, but badly managed and staffed IT departments, just like terrible consulting companies, are really that bad.

If anything, that should make the revenue per embloyee higher, not lower.

No, the opposite.

You have 100 employees that are billed out at $25/hour while only costing $10 versus 10 employees being billed out at $250/hour while costing $100. (I think that math works out to 2500/hour revenue and 1000/hour costs in both cases). So the company earns the same revenue and profits, but the revenue per employee is much lower.

If anything, that should make the revenue per employee higher, not lower.

Revenue per employee has lots of major problems as a useful measure.

* Energy companies can get huge revenues by buying and reselling expensive oil, but the profit is a tiny fraction of revenue.
* If a company employs its own cleaning staff instead of buying it as a service from a cleaning company, that greatly changes this measure. The actual efficiency of the company has not changed at all.
* Apple is probably dragged far down by employing its own Apple store staff.

I agree. My first reaction was that revenue per employee is a pretty useless number.

Revenue per employee might be useful in comparing companies within the same industry, using it as a measure of efficiency, but I don't get the value of comparing it across industries.

The local water utility has a higher revenue per employee than the local auto body shop.

So what.

This is, like, a rookie error a business analyst @ McKinsey would be embarrassed committing.

You people. An "error" would be drawing an inappropriate conclusion from the analysis. They are only presenting it as an interesting, different way of slicing the data. You might disagree over whether it is interesting, but it's not an error.

What do you mean "you people"?

Bill and carlospln.

Carlo, You have a problem.

First, you didn't offer any support.

Next you did an ad hominem.

I've taught at a graduate business school and use IO econ in my practice.

Revenue per employee across different industries is irrelevant. Revenue per employee is relevant for comparisons within the same industry.

Here is something from Ivestopedia on the proper use of revenue per employee and how it is used:

"Factoring in Company Industry
To evaluate revenue per employee, a business compares its results to other companies in the same industry. Some industries, such as banking, require a large number of employees to staff physical locations and answer customer questions. The banker should compare his company's results to competitors in the same industry."

Read more: Revenue Per Employee
Follow us: Investopedia on Facebook

FWIW, I thought carlospln was agreeing with you--calling the article, not your comment, a "rookie error".

If it is, my error and apology.

I was agreeing with you.

Carlo, I apologize to you.

This is not a useful metric. They should compute something like Gross Margin per Unit of Labor Cost (a dimensionless ratio). And even that metric would have ambiguous interpretations.

It's always good to use "revenue" as a metric when the useful number is of course "profit." It confuses the rubes. And if it helps you stick it to the pharma industry, all the better.

I admit it's not quite as good as saying that oil businesses that deduct their expenses are exploiting a "loophole" but it's still very effective.

"Technology companies performed at the lower end of the range on Revenue per Employee . . ." Apple, Facebook, and Google are the outliers, generating very high levels of revenue per employee, Apple because it has high non-employee costs, Facebook and Goggle because they derive most of their (very high) income from digital advertising and don't have that many employees. Why the huge disparity among the tech companies? Is it because the tech companies don't generate much revenue or because they have so many highly paid employees, or both? I've commented before that Facebook and Google capture nearly 70% of digital advertising revenues (and an even higher percentage of the marginal dollar spent on digital advertising). Yet, money continues to pour into the tech sector. Do investors believe all these tech startups can duplicate the success of Facebook and Google by capturing some of their revenues from digital advertising? That seems unlikely, as the scale of those two advertisers can't come close to being matched. Or do investors believe that Facebook and Google will acquire the tech startups for an enormous prices as with WhatsApp (or the much lower priced Instagram)? In other words, do Facebook and Google drive the tech sector? If that's the case, is the tech sector an illusion?

We'd see a different picture if Amazon wasn't one company, but two, one being their retail operation, and another AWS: You'd probably see AWS far closer to Google and Facebook than anything else.

The disparity in tech companies also has to do with their maturity: If a 5 year old tech company can keep growing 100% YoY while having revenue matching their costs, or has to drop to 5% a year if they want to focus on revenue per employee, you'll see then go for the 100% growth rate every time.

Money is pouring into the tech sector, but most of it isn't for companies that plan to make their money on ads: That part of the market looks pretty tapped. Nowadays you use a big company for your ads, if you are ad driven, but consider their cut the cost of doing business, just like your credit card processor. Most interesting startups are doing things that the large majority of the public would find way too boring. And all of that said, the money going to the tech sector is shrinking: While a few years ago it was very easy for anyone to get 3 rounds of funding, nowadays it's quite difficult to raise seven figures, much less eight or nine like the top startups have, without demonstrable progress towards a pot of gold that is worth at least 10 billion. Go to a site covering startup news and look at big C rounds: Some are going to terribly mismanaged companies, but they are always in sectors that really have the power to reach google-like levels of revenue per employee.

And yes, revenue per employee is not the best metric. Look instead at a different flawed metric, like market cap per employee, including latest valuation for private companies. We either have to believe that the EMH is completely bonkers, or that there is much opportunity behind tech companies.

That advertising revenue is going is no one's surprise. Digital and television are locked in a creative destruction horn battle. The question is and always has been, is advertising better or is marketing more important. If advertising is better than digital is the way to grow. If marketing is importantly than television budgets should continue to advertise their viewers. At least in advertising, the marketing is clear.

Revenue is not a great measure because of the reason that you mention what I would like to know what companies have the greatest value added per employee.

As most people have already chimed in, this is a dreadfully brain dead analysis. Gross profit or value add per employee are much better metrics for comparison across industries. Top line revenue is meaningless as a cross industry comparison. Really, nothing to see here.

Part of the answer may be found in the commonalities among these energy companies not on that first table: Gulf, Texaco and Mobil. They were slow to modernize, prone to hiring rather than contracting out maintenance and thus among the first devoured when winter came. Valero runs a refinery that one of them built with 1/9th the permanent staff. EOG is another good example. I attended a meeting recently at which its CEO explained how they'd managed to drive down costs while driving up production; and mainly it was due to predictive analytics run on the vast amounts of data derived by seismology and an amazingly sophisticated array of downhole sensory devices. Trigger warning: cognitive dissonance ahead -> Some of the energy companies on that list are more tech-nerdy than the tech companies with which we associate the stereotype. They're barely more than a bunch of engineers, geologists and petro-quants disrupting the E&P model they established six months ago.

The most interesting thing about this is that the same people who are carping about how this is a meaningless metric are many of the very same people who you would expect to lambaste, for example, Walmart for paying low wages when they have massive revenues.

What's the logic from that statement.

Please explain. Dare you.

Americource Bergen, McKesson, and Cardinal, are distributors, basically logistics companies that specialize in medical products. Because of accounting rules, the cost of the products they distribute are included in their revenues, but really it's essentially a pass through. In fact, their margins are slim, and they are liable to disruption in an existential way (management is trying to diversify away from these old business lines). So, really dumb to say they "score well" on revenue per employee, or to call them healthcare companies.

IT companies hire the cheapest workers they can find. Capitalism, I spose...

Facebook and Google sell your data without giving you a cut. Take heart, tent seeking is still in style. If we all aspire to be rent seekers the world will be a belter place - greed is good.

Alfred E. Neuman for pres! Yay! It's our time.

Uh, where's that spell checker? :)

Tent seeking is good, but belters are brutes.

Here's a 2020 pres campaign poster, leaked by the big bear. Don't laugh, remember Trump!

In an age of automation this leads me to wonder if there is some societal worth to inefficiency? Looking at large retailers from the S&P 500, I see the following:

Retailer Revenue Employees RPE
Amazon $136,000,000,000 351,000 $387,464
Walmart $486,000,000,000 2,300,000 $211,304
Kroger $115,000,000,000 443,000 $259,594
DG $22,000,000,000 121,000 $181,818
TGT $69,000,000,000 323,000 $213,622

Assuming some of these retailers will be overtaken by, or will rise to the efficiency levels of, Amazon, then we should see massive declines in employment in the retail sector. Productivity gains are lovely, but have we reached a point a certain amount of inefficiency should be tolerated just to give people something to do?

What's the revenue per employee at the IRS?

15% (-fed rates) on top of the agency cost per year/revenue of the tax collection.

I think market cap/employee is a little more useful here. At tech firms most high-earner compensation comes from stock, which is literally a dilution of the market cap.

It's also a flawed metric but you'll find it's more useful in discovering which tech firm pays the most :).

Technology companies are largely unprofitable scams.

Back in the early '90s I recall being astounded that ARCO had about 2000 employees.

So what? This is driven in large part by (a) how a firm or an industry is organized, specifically whether it outsources a large part of its functions, or does them in-house with employees, and (b) how capital intensive it is, with lots of capital substituting employees leading to higher numbers but meaning little.

This treats employees as if they are a bad thing, you get better numbers with fewer employees and more capial substitution for them.

Return on invested capital is what counts. That includes employing people when it is more efficient to do so than deploying more capital. All the rest may be interesting but can only be understood with a deep dive into the structure of the business and its markets.

Comments for this post are closed