There's an idea I found buried in Neil Irwin’s How to Win in a Winner-Take-All World that I thought deserved a little more attention. Irwin argues (and I’m paraphrasing here) that one of the most important things you can do for your career in today’s world is to understand the business models of the industry you’re in.
Doing so allows you to see some of the shifts that occur as they happen. Irwin likens this to ‘a sailor (who must) understand winds and currents’ as he’s navigating his ship at sea. The sailor might not be able to influence these external factors, but he is willing and able to adapt to them. Focusing on the boat beneath his feet while ignoring the effects of the weather around him means that he's a terrible sailor. Similarly, in your career, you probably shouldn’t focus on the mastery of your work to the exclusion of studying the market you work in.
I’m biased here. I think that understanding business models are important — but then, I work in startups. With startups, the wall between the work you do and the business you’re in is remarkably thin. You can't rely on company brand or signalling effects to evaluate which startup you should go work at. So instead, you learn to evaluate the business of the startup itself.
Another way of looking at this is to say that startup people tend to think more about the broader health of the company they're in, because if the business isn't doing well ... they leave. Irwin’s book, however, asserts that this skill is important for everyone in our current era of globalised, knowledge-oriented, semi-automated work.
Here are a few stories of people putting this idea to practice.
The Business Models of Investment Banks
In the late 1800s, bankers like John Pierpont Morgan Sr practiced a form of banking called ‘relationship banking’. This style of banking saw the banker as far more than a mere provider of capital: the banker was also the ‘lawyer, high priest, and confidant’ — with direct involvement in the businesses they provided capital to.
As author Ron Chernow puts it in his book The House of Morgan: ‘this came about not because the bankers were strong, but because companies were weak.’ Most companies at the time were badly managed and in dire need of capital … which they acquired from the merchant banks. The bankers, in turn, could use their position as gatekeepers to capital to grant themselves directorships or board seats in the companies they worked with.
This kept the banks really cosy with their corporate wards. Imagine, if you will, the CEO of Citibank telling Google and Apple to stop competing with each other today ... and imagine that both companies actually comply, for Citibank owns their asses. Or imagine the CEO of JPMorgan Chase announcing that going forward, all electric utility companies in the US will merge into one giant corporation, under the auspices of their M&A department. This is crazy-talk now, but it was par the course at the turn of the century. In fact, in 1901 Morgan Sr merged Andrew Carnegie’s Carnegie Steel Company with Elbert Gary’s Federal Steel Company and William Henry Moore’s National Steel Company to form US Steel — the world’s first billion dollar corporation, with a combined capitalisation of 1.4 billion in 1901 dollars ($14.82 billion dollars today).
Relationship banking began to erode over the course of the 1900s, as companies became more powerful, and the sources of capital available to them became ever more plentiful. Over time, Chernow writes, companies would come to treat banks as replaceable commodities. They no longer needed to borrow money via loans — large companies in particular had become multinational concerns in the post-war boom, and their massive cashflows and commensurately massive capital needs made it difficult for any one bank to service them adequately. To make things worse, in the 1960s, larger corporations began to circumvent the banks and sell commercial paper at interest rates far lower than what would be offered in a bank loan.
As a result, by the time the 90s arrived, investment bankers were reduced to two functions: they provided advice on matters of corporate finance, and they raised money for companies. But the good times of yesteryear were truly over. Compared to the giants of Morgan Sr’s era, the investment bankers of today experience greatly reduced power and influence for significantly longer hours and substantially more competition ... as we shall soon see — from the eyes of two young bankers.
In Monkey Business, former investment bankers John Rolfe and Peter Troob recount the diminished role of investment bankers in the modern era, as background to their shared experiences in the now-defunct Donaldson, Lufkin & Jenrette. They start with a discussion of the banker’s role as financial advisor:
In general, though, the advisory side of the business has become much more commoditised. The banker no longer has the lock on relationships. The banker's information is no longer highly proprietary. Information on companies is now so widespread that there's very little company-specific knowledge that bankers can truly call their own. The banker no longer brings enough unique added value to the table to necessarily merit a CEO's granting him a lifelong mandate to provide paid advice on matters of corporate finance.
As the bankers' competitive information advantage has waned, the bankers have gradually been forced to change their approach. They can no longer rely on a relatively small number of loyal clients to generate advisory business for them year in and year out. They now have to spend a much larger portion of their time scrambling to find new clients and new business. To justify their existence, they now have to go out and pitch ideas to whomever will give them an audience in the hope that just a few of the potential clients will sign on for the program. And when those clients sign on, the bankers have got to assume that the next time there's advisory business ' to be had with that company, it might not necessarily be them providing the advice. In short, the banking business has become a whole lot more like most other businesses out there — competitive.
And on capital raising, they had this to say:
Until recently, there weren't many new entrants to the underwriting business. Because an investment bank needs a certain minimum scale to operate profitably, there haven't been many new players willing to make the necessary up-front investment. Increasingly in recent years, though, as the risk of underwriting has come down and fee spreads have stayed constant, the economic return has appeared increasingly compelling for potential entrants to the business. As this has happened, the new entrants have begun to make their appearance.
The first new competitors through the door have been the U.S. and foreign commercial banks. Increasingly, the large regional commercial banks have begun to set up securities underwriting subsidiaries and have begun to hire away investment bankers from the DLJs, Morgans, and Goldmans of the world. New investment banks have begun to pop up with an operating model based on online distribution of IPOs direct to retail investors. As the number of underwriters competing for each piece of underwriting business has proliferated, the spreads have come down.
In Monkey Business, Rolfe and Troob conclude that the insane hours they had put in while at DLJ was a reflection of the increasingly commodified world of investment banking on Wall Street. You’ll have to read the rest of the book to get their stories, but in short, it involves over two hundred sleepless nights at the office, the occasional hooker, unadvisable amounts of alcohol, and doing work of middling value to the broader economy.
Their solution? By the end of the book, both authors had burnt out and switched over to the buy-side of the business, concluding that investing capital was far better for their sanity than the competitive meat grinder that was investment banking.
Profit Centres and Cost Centres
In 2011 I came across a Patrick McKenzie blog post titled Don’t Call Yourself a Programmer, and Other Career Advice. In it, McKenzie wrote:
Peter Drucker — you haven’t heard of him, but he is a prophet among people who sign checks — came up with the terms Profit Center and Cost Center. Profit Centers are the part of an organization that bring in the bacon: partners at law firms, sales at enterprise software companies, “masters of the universe” on Wall Street, etc etc. Cost Centers are, well, everybody else. You really want to be attached to Profit Centers because it will bring you higher wages, more respect, and greater opportunities for everything of value to you. It isn’t hard: a bright high schooler, given a paragraph-long description of a business, can usually identify where the Profit Center is. If you want to work there, work for that. If you can’t, either a) work elsewhere or b) engineer your transfer after joining the company.
Engineers in particular are usually very highly paid Cost Centers, which sets MBA’s optimization antennae to twitching. This is what brings us wonderful ideas like outsourcing, which is “Let’s replace really expensive Cost Centers who do some magic which we kinda need but don’t really care about with less expensive Cost Centers in a lower wage country”. (Quick sidenote: You can absolutely ignore outsourcing as a career threat if you read the rest of this guide.) Nobody ever outsources Profit Centers. Attempting to do so would be the setup for MBA humor. It’s like suggesting replacing your source control system with a bunch of copies maintained on floppy disks.
McKenzie’s observation that software engineering came in two flavours was hugely influential to my career.
The obvious implication from his blog post was that you should — whenever possible — optimise for working in a profit centre. Benefits accrued to you if you worked in a profit centre. They didn’t if you worked in a cost centre.
A few years later, I realised that what was considered a profit centre or a cost centre was a matter of managerial perception, not necessarily business fact. A senior of mine worked in the IT department of a commercial bank in Singapore, and was part of the team that did the revamp of their consumer banking site. While ostensibly a cost centre from the perspective of the bank’s business model, her team was paid well and treated decently, because they were part of some vice president’s ‘digital transformation strategy’.
(As an exercise of applying business model understanding, consider: what is most likely to happen to this team if the VP leaves or is promoted out of the department? And what would you do if you were in my friend’s shoes?)
A third and final implication of McKenzie’s post turned out to be useful when I was coming up with a hiring strategy at my previous job. We were having some difficulty hiring software engineers in Ho Chi Minh City, where our technical offices were located. I realised, after some trial and error, that the market was divided into two kinds of jobs: cost centre outsourcing jobs, and profit centre startups. This wasn’t a clean comparison, of course — at the time, companies like East Agile and Silicon Straits Saigon were both great dev agencies with good internal culture and decently high salaries. These two firms were representative of the better service agencies, and it didn’t make sense to mix them in with the ‘cost centre outsourcing jobs’.
I eventually came up with the term ‘good tier’ and ‘bad tier’ to describe this split. I lumped the good dev agencies with the product startups and called them the ‘good tier’. The bad tier, on the other hand, consisted of bad cost-centre outsourcing companies. This latter category made up the vast majority of tech jobs in the city.
Our hiring strategy worked like so: we would hire fresh grads out of university, and pay them 80th-percentile salaries. For senior roles, however, we set our compensation lower than the top of the market, and targeted candidates from the bad tier. My hypothesis was that there were several good engineers in the bad tier companies that wanted out, and it would be relatively easy for us to hire them. As a result of their prior experience, their salary expectations weren’t as high. This worked out for us (I didn’t have to sell a crazy expensive new comp system to my boss) and it worked out for them (I promised to train them so that they may leave the team after two years, if they wish — something I’m pleased to say has occurred).
Looking back on it, the returns to my career from McKenzie’s simple observation has been pretty large. In 2017, I wrote an essay titled The Two Tiers of Singapore’s Tech Companies, which applied my ‘good tier/bad tier split’ to the Singaporean labour market. The essay went viral in Singapore’s tech community. As a result, the conversation around hiring and compensation has changed rather considerably (I sometimes see derisive comments on Singapore’s subreddit going ‘this is so obvious’ — which is the best compliment ever); my friends and I still regularly use ‘good tier’ and ‘bad tier’ to gauge companies we want to work for.
Business models matter when it comes to jobs.
The Joys and Limitations of Working For a Consultancy
Here’s a final example.
I sometimes joke that a tech consultancy is a fancy term for what is essentially an over-glorified outsourcing firm. You may know it by other names, of course — an agency, a service firm, a dev shop. The core idea is that these companies perform a service for a fee. In tech, the service most commonly performed is development work of some sort.
Service businesses are looked down upon by the startup world. Startup culture is built around VC returns, and so startups are expected to scale: growth is expected to be exponential, distribution must be viral. All of these are true and good.
But consultancies have a number of strong benefits. There's practically no business model risk: you receive a steady stream of income for as long as someone is willing to pay you for dev work. Consultancies also benefit from a simple cost structure: they tend to have low fixed costs (rent, computers) relative to their variable costs (salaries).
Here’s Jonathan Siegel, in The San Francisco Fallacy:
I tried to combine the best of both worlds in our consultancy: the intimacy, collegiality, and creative fire of a startup with the sustainability and family-friendly environment of a more stable company. Using the company as a development lab for my own projects (and those of other team members) helped keep everyone stimulated and the atmosphere creative and supportive.
Consultancies aren’t immune to the business cycle, but they can have an intrinsic countercyclical thrust. In a boom, you can scale up, but only as quickly as you can hire. In a bust, you can cut your margins and cut staff, responding immediately to the changing market rather than relying on economic forecasts. One of the virtues of consultancies is that they can do well in a downturn by helping the larger companies around them downsize, streamline, cut costs, and gain efficiency.
The point I’m making isn’t that consultancies are good and that product companies are bad (or vice versa) — merely that business models matter when we’re talking about work. In this particular case, understanding the tech consultancy is important to you if you’re thinking of working at one.
Let’s think through the implications of the tech consultancy business model. Because the majority of costs in a consultancy are variable costs (read: staff salaries), consultancies live or die by their ability to control costs in a downturn (read: fire lots of people). The likelihood of a layoff happening in a consultancy is a direct function of the pipeline of deals in the company, the amount of money you have in the bank, and the salary costs on a per-month basis.
What this means: if you work in a consultancy, and you observe the pipeline of deals drying up (or overhear that the partners are cutting margins on each deal), it might be a good idea to prepare for a layoff.
But we can get more nuanced than this.
Think about what happens to a consultancy when it grows. Assuming no external capital, consultancies grow using the profits they’ve made from past deals. This then means that unless the owners of a consultancy are prudent and control growth in a thoughtful manner, rapid growth is a sign of diminishing profits (or a diminishing amount of cash on hand). This makes rapidly growing consultancies more susceptible to variation in deal flow. In other words, it makes them fragile.
I observed this first hand in a consultancy near us in Ho Chi Minh City. At first, I was envious of their size, rapid growth, and internal culture (we were a competing consultancy; this was before we pivoted to product). But over time, I realised that the continued pressure to bring in and close new clients made it incredibly stressful for the principals of the firm. Any small variation in deal flow would mean severe cash problems for the firm. Within a few years, the company sold itself off for parts.
The implication here: you should be doubly careful if a consultancy is growing rapidly (and you don’t see any sign of external investors putting money in to create a cash buffer). Assuming no capital injection, the firm is either growing from retained earnings, or it has taken on a loan — which makes the firm even more fragile, at least when loan repayments kick in.
But we can get more nuanced than this.
Why are tech consultancies so often engaged in a race to the bottom for fees? Why are so many tech agencies stuck in a local minima of growth? And, on the other end, why can certain non-tech consultancies make a ton of money? Compare firms like McKinsey, Boston Consulting Group, and any superstar law firm to the typical tech agency. Why is there such a discrepancy in outcomes?
In Managing The Professional Service Firm, David Maister describes the pyramid of work that all consultancies engage in:
- Brain projects. “In the first type (Brains), the client’s problem is at the forefront of professional or technical knowledge, or at least is of extreme complexity. The key elements of this type of professional service are creativity, innovation, and the pioneering of new approaches, concepts or techniques: in effect, new solutions to new problems. The firm that targets this market will be attempting to sell its services on the basis of the high professional craft of its staff. In essence, their appeal to their market is, “Hire us because we’re smart.”
- Grey Hair projects. “Grey Hair projects, while they may require a highly customized “output” in meeting the clients’ needs, involve a lesser degree of innovation and creativity in the actual performance of the work than would a Brains project. The general nature of the problem to be addressed is not unfamiliar, and the activities necessary to complete the project may be similar to those performed on other projects. Clients with Grey Hair problems seek out firms with experience in their particular type of problem. In turn, the firm sells its knowledge, its experience, and its judgment. In effect, they are saying, “Hire us because we have been through this before; we have practice at solving this type of problem.”
- Procedure projects. “The third type of project, the Procedure project, usually involves a well-recognized and familiar type of problem. While there is still a need to customize to some degree, the steps necessary to accomplish this are somewhat programmatic. The client may have the ability and resources to perform the work itself, but turns to the professional firm because the firm can perform the service more efficiently, because the firm is an outsider, or because the client’s own staff capabilities to perform the activity are somewhat constrained and are better used elsewhere. In essence, the professional firm is selling its procedures, its efficiency, its availability: “Hire us because we know how to do this and can deliver it effectively.”
Brain projects command the highest fees with the highest margins, but require skilled professionals to deliver them from scratch each time (imagine a major, novel court case, like the Microsoft antitrust lawsuits of the early 2000s). Grey Hair projects do not require so much customisation, and may reuse artefacts the firm has built from prior projects (imagine the somewhat cookie-cutter cost cutting projects done by McKinsey). They command medium-level margins. Procedure projects command the lowest fees amongst the three, and must compete on efficiency and cost.
The answer, then, is simple. McKinsey and ilk go after a mix of Brain and Grey Hair projects. They structure operations so that novel Brain engagements produce knowledge artefacts that may be reused in Grey Hair projects. The vast majority of tech consultancies, on the other hand, perform Procedure work, with no way to climb up the pyramid. They are stuck in a race for lower costs, lower prices, and higher efficiency.
Not all tech consulting companies are Procedure-oriented firms, however. Some manage to climb the pyramid of work. In Singapore, Pivotal provides software consultancy services at the Grey Hair level: they sell engagements of embedded software consultants that train the client’s engineering teams, during the development of some software project. Consequently, they charge very high prices: something on the order of S$10k per consultant per week. (Don’t quote me on this — I’m extrapolating from numbers I learnt years ago).
I should note, however, that this isn’t a complete picture. Pivotal is augmented by a product business (their Pivotal Cloud Foundry), and Maister’s ideas go beyond just the pyramid of work. But a simplistic understanding of Maister’s book is sufficient for our purposes: you should evaluate consulting firms based on the mix of work they perform. The higher the position they occupy on the pyramid of service work, the more lucrative a career with them becomes.
So what have we learnt today? We’ve considered three different business models in three different industries, and explored how an understanding of each of these models may inform career moves within them. Often, these moves are built on second and third order implications of the underlying business dynamics of the firm.
Before writing this essay, I’ve occasionally wondered at the recommendation — often given at commencement addresses — to “read about the history of your industry”. I think I now understand why this is the case.
Reading stories about your industry forces you to become familiar with the business models that underpin your job. Firm dynamics have a tendency to show up repeatedly in a given industry. Learning what these are, what they look like, and why they exist should make it easier to understand what’s going on as you navigate your career.