This essay kicks off a series called Figures of Merit, in which we  apply analytical methods in non-scientific environments. We’ll start by exploring the relationship between time and money for early-career scientists. The question: how much is our time worth? Not only is this this a difficult question to answer  – but for academic scientists beginning a career in industry, it is a question rarely asked. If you want to skip the analytical part, here is the message: I can’t tell you the value of your time, but neither can you, unless you know how to measure it.

For the data set in this analytical exploration, I’ll use a few key metrics from my own career. A career, after all, is an experiment – and like any experiment, the outcome largeley depends on our ability to observe something meaningful.  Of course this is an experiment with N of one, and not representative of all scientists. But if we consider ourselves to be in the business of reducing uncertainty (I do), and if we think career planning represents a lot of uncertainty (it does), then the best tool we have is to reduce that uncertainty is a plan. Even with a limited data set, we have something interesting to compare: the difference between Plan and Reality.   

The timeline below shows my plan as a grad student (the top line in blue): spend a couple years as a postdoc, then get a Real Job, beginning an unbroken trajectory towards Success. Success, in this case, was loosely defined as paying off loans, buying a house, and achieving some financial security. Reality (bottom line, in black) turned out to be a different story: the postdoc was longer than expected, jobs were hard to find, and keeping a job was even harder.

The first thing you might notice is the quantity of black circles: each one represents a new job. Less obvious, but even more meaningful, are the blank spaces: those represent time without work. Needless to say, this was not part of The Plan, and had many consequences not easily shown in this chart.

Much of what matters can be hidden in charts. While the visual above does highlight the number of job transitions, it does not show the impact, or cost. Here are where Figures of Merit come in. Measuring progress in academia was easy. Papers published, impact factor, grants awarded – these are the currency of research scientists and easily recogized by everyone in the field. However, this currency carries little weight outside the academic world.

For the goal I was trying to achieve (financial security), a more universal metric would be something related to money. I considered plotting student loan debt as a function of time,  but decided against it because a) it would reveal my embarassingly naive financial planning and b) it is convoluted by other decisions (having kids, my wife’s post-graduate education, buying a home, etc), and c) there is nothing interesting about a flat line.  

In my case, a more useful figure of merit is retirement savings. This is not on the list of priorities for most early career scientists, but it will catch up with all of us sooner or later. Consider:

• Paying for retirement is a necessity for everyone, regardless of career path.
• For those of us who spend extra years in school, we have less time in which to earn and save before retirment.
• Many of us live in high-rent areas, leaving little room in the budget for long-term saving.

The graph below shows my cumulative retirement savings since my PhD, in units of months (I estimated the monthly cost of retirement, assuming my overhead is roughly the same as today). Moving from left to right, there are three periods of my career reflected with three different slopes: the first is the 3+ years as a postdoc, the second is my 4 years working in startups, and the last (with the steepest slope) shows my two years to date in a large company.

There a few lessons I take from the graph:

1. My retirement savings as a postdoc was significant, even though the salary was less than half what I made later in industry. This is mainly because the university took automatic deductions to my pre-tax income.
2. I saved roughly the same amount in the two-year “corporate” period of my career compared to the first 6 years combined.
3. My retirement savings were essentially flat during the “startup” period, even though my salary was considerably higher than a postdoc. Four job transitions meant four periods of gaps in salary, gaps in health insurance, gaps in free snacks, etc.

The savings vs time graph is one way to illustrate opportunity costs when considering different career paths. These opportunity costs are not obvious when job hunting. After life as a grad student, almost any job can seem lucrative – but it depends on the metric. Thinking about long-term savings as a metric may make us consider job stability just as much as salary. On the other hand, employers at startups may offer not just immediate income, but (potentially) imminent wealth & success.  This makes the evaluation of stability vs income more difficult, because the income is partly theoretical.

When I talk to early-career scientists about the decisions they face, I ask them to identify their figures of merit. If they are concerned with long-term financial stability, I bring up the lessons shown here. But this is just one example for one individual. More importantly, each person can identify what is important in their life. Money, fame, vacation time, respect among peers – these are all legitimate ways to measure a career. The real question is: can we have the foresight to plan our careers, making choices to optimize for the things things we care about?

OK – we’re scientists. We know it’s easy to apply analytical thinking to a bunch of things that have already happened. The trick is to use that thinking learn from observation, build models, and plan ahead. We do it all the time with our experiments. Now how would it look if we apply the same thinking to our careers? I look forward to hearing from readers – what are your figures of merit?

The pressures of stagnant research funding, limited academic jobs, and increasing numbers of graduates are creating a bulge in the postdoctoral population cleverly termed “the postdocalypse.” The academic community is addressing this issue – slowly, laboriously, and often reluctantly (see “The case of the disappearing postdocs” below). For the postdoc working on the 5th revision of a manuscript while the grant money runs out, the problem is much more acute.

Understandably, it’s hard for academic scientists to look beyond the current experiment and the next paper. I frequently hear from scientists at the stage of “I’m graduating in a month – now what?” Or, “My postdoc funding is running out – now what?” In the language of entrepreneurship, PhD students and postdocs should be planning an exit strategy.

A knowledge gap big enough to drive a small business into.

Procrastination is a poor strategy for planning a career, but it works as long as there is a next stage. Masters, PhD, postdoc (2nd postdoc)… Those transitions may be difficult, but they occur within the same academic ecosystem. The post-academic transition is something else. The skills, strategies, and tactics are simply different, and this is the knowledge gap that waits at the end of the academic phase. There are many voices bringing visibility to this problem, but let me offer the following as one viewpoint. The market has recognized the difficulty of PhD scientists transitioning to jobs outside academia, and has responded with a small but growing industry: Post-postdoctoral career training. This is the business of training science PhDs and postdocs so that they can begin careers in industry, science policy, or other non-research fields.

There is a harsh way of painting this picture that’s hard to ignore: A decade of scientific education and training does not make you employable. This is a statement that can make postdocs angry, professors defensive, MBAs smug, and university PR firms nervous. But what gets lost is that the career training that most scientists need is not “instead of” their education.

The career development programs I’ve seen have three main offerings: One is to help scientists understand the culture and expectations of industry: for example, putting a priority on teamwork and deadlines. Another is to help grow a professional social network so that more opportunities become available. Finally, there is the process of realizing how many skills are acquired during doctoral education, and which of those a person wants to develop as a profession. The world needs scientists – it just doesn’t need scientists to do the same things they did as grad students and postdocs.

We’ve got procrastination down to a science.

Early-career scientists put off career questions until after the very end of their academic runway. This is largely a matter of urgency. Graduate students are master procrastinators. We finish plotting the data Tuesday morning because the deadline to print a poster is Tuesday at noon. How urgent does preparing for a career feel, when we don’t even know exactly what it is or what that means?

Scientific research is hard. But for grad students and postdocs, at least there is a framework: Advisors, peers, technicians, and instruments are all there to support the research. So it is no small challenge to ask scientists to simultaneously carry out their research and prepare for a career. But the research has milestones and deliverables (experiments and papers). Career preparedness has no milestones and no deliverables; at least none that will be identified by your committee. This is the vacuum that post-postdoctoral career training fills. It gives  a framework for bright scientists, who will solve many types of problems along their academic path, to address a problem they have not thought about.

Expecting universities to deliver combat-ready scientists is a tall order. Those with strong industry collaborations tend to offer more career-development opportunities, but this is not the norm. Some institutions are responding to the needs of graduates with skills training and career workshops, but it’s no surprise that academic institutions tend to focus on academic career training. Professors can be great scientific advisors and still not recognize how important transferable skills, like networking and communication, will be for a non-academic career. This does not show a lack of caring, just a different perspective.

For science PhDs, getting useful data on career outcomes is not easy. Only recently, graduate departments have begun tracking career outcomes of their PhDs, and the data is sparse. Even with data in hand, knowing about career outcomes doesn’t lead directly to career preparedness. Of all the stakeholders in this game, it is grad students and postdocs that have the greatest incentive to look ahead to their career path.

As scientists, we deal in data. Now is the time to treat your career like a research project, even if it’s a small one. How will you go about getting data to prepare for your post-graduate career? Many people are willing to help you: Some are your advisors. Some are strangers. Some do it as a business. Some want to help because they’ve been there. You have the time and resources to prepare for careers you never even knew existed. The only real risk is in not asking the question.

Links

Finding the post-academic career path. A companion article in Lab Without Benches lists post-postdoc training services.

The Case of the Disappearing Postdocs. More scientists are going directly from PhD programs to industry jobs – is the value of postdoctoral training in decline? By Beryl Lieff Benderly, Science, 2015