Q: How do you find the perfect postdoc?
A: The perfect postdoc doesn't exist.

Whether you've just started graduate school, are halfway through, or are finishing your Ph.D., it's never too soon to start thinking about the next step. For many Ph.D. scientists, the "next step" is a postdoc.

A postdoc is nearly always required for tenure-track faculty positions, especially for positions in research. But here's a shocker for those of you who haven't peered out of the windows of the ivory tower. Only a small minority of science Ph.D.s ever achieve a tenure-track position. A recent Federation of American Societies for Experimental Biology study showed that in 2001 only 14.1% of biomedical scientists surveyed 5 to 6 years post-Ph.D. had tenure-track faculty positions. That's down from 34% 20 years earlier. Although the statistics don't address this, it's safe to say that a relatively small fraction of the remaining 85.9% will ever get a tenure-track position. Sure, the percentages vary from field to field but not by much. Tenure-track professorships are rare no matter what kind of science you do. So if it's time to start looking for a postdoc, it's also time to take a good, hard look at your prospects and your long-term plans.

What are the options? They are too numerous to mention. That's the problem with so-called alternative careers: More people with Ph.D.s work in these careers than in traditional careers, but they're so diverse that they're difficult to categorize. Some Ph.D. (and M.D.) scientists go on to work in regulatory affairs. Some choose science policy. Others become editors. Still others become outstanding schoolteachers. Others choose industrial research--for which a postdoc is not a bad idea but may not be essential. (Of three recent hires into drug-development staff positions at one "big pharma," two were hired straight from their organic chemistry Ph.D.s; only one had any postdoc experience.)

Scientists with advanced degrees do all sorts of work. They typically find the transitions into their new careers easy. And they almost always find the new work very rewarding. Yes, folks, there is intelligent life outside the academy, not to mention more family-friendly policies and better working conditions. So if you have doubts about your evolving bench career, now is a very good time for some serious introspection. Pay a visit to your institution's career office. And check out Next Wave's Index of Monthly Feature Indexes. With hundreds of essays about and by scientists in dozens of careers, this is undeniably the best resource available for scientists considering their away-from-the-bench options.

Still with us? Still want to know how to go about finding a postdoc? Then let's get started.

You've convinced me that I should consider alternative careers, but I'm almost done with my Ph.D. and I'm not sure what I want to do yet. Should I do a postdoc while I'm figuring out what I want to do with my life?

A short postdoc is unlikely to hurt you no matter what career you end up in. It's unlikely to hurt you, that is, unless your postdoc adviser is hostile to the idea of alternative careers.

For many careers away from the bench, a postdoc won't help at all, but by the time you're finishing up your dissertation it might be too late to properly investigate your other options. So get started!

But if it's time to choose and you haven't decided what you want to do yet, or you haven't found the ideal position outside of academia, there are worse things you could do with your time. If you enjoy the work, go for it.

But make sure your boss is open to the idea of an alternative career. Enlightened folks know that there are many ways for a Ph.D. scientist to do well, to make themselves a benefit to themselves, their mentors, and society. But not every PI is enlightened. And a hostile PI can hurt your career, no matter which direction you end up pointed in. So it's best to be up front: Tell potential postdoc advisers that you're considering away-from-the-bench careers. Seek assurance that they consider this a legitimate choice. If you don't get that assurance, consider moving on.

How should I go about looking for a postdoc?

Many published information sources exist, from classified ads in Science (and the online ads at Science Careers), to The Chronicle of Higher Education , to the journals of your discipline's professional societies. Unless you're very interdisciplinary, the more discipline-specific publications ( Physics Today , Chemical & Engineering News , and so on) are likely to be richer sources of information than the more general magazines and journals.

These sources list many good jobs, but you may not need to use them. Very often, the best information isn't written down, and the best opportunities aren't advertised. To invoke a phrase I first heard in connection with sport fishing, "You've got to know the water." Theoretical expertise is a very fine thing, but what you really need is what some fishers call local knowledge. You can have the best fishing gear and know everything there is to know, in a general, theoretical way, about fishing, but if you don't know the water you won't catch much.

In the case of finding a postdoc position, "local" doesn't refer to geography but to your particular subfield or the subfield you intend to work in, if you want to switch fields. If you don't know the water yourself, you need to find a guide who can take you to the fish-- er, the jobs. Your guide may be your Ph.D. adviser, but don't rule out any possibilities. If your adviser seems uninterested in your professional development or lacks the connections to help you, look around, look elsewhere. Look for someone who knows the answers to all the following questions and is willing to help you. Who runs the best labs? Who publishes in the best journals? Who just won a big grant and is likely to be looking for people to do the work? Another question you need to know the answer to: Which scientists in your field do the best job supporting and nurturing the fledgling careers of their laboratory staffers?

Should I seek an individual postdoctoral fellowship?

Sure, go ahead. It can't hurt. The experience of writing the application will be valuable, and there are definite advantages to having your own money. But most fellowships these days--even individual postdoctoral fellowships--require the buy-in of a potential mentor. So most postdoctoral fellowships don't give you as much independence as you might think. Furthermore, the best fellowship proposals generally include preliminary data, so you'll be able to write a better proposal after you've had some time in the lab.

Anyway, you shouldn't need a postdoc fellowship. There's lots of money out there. And although there's hardly an excess of permanent positions, the demand for well-trained, talented postdocs outstrips the supply these days by a fairly wide margin, if anecdotal evidence is to be believed. As one postdoc put it in a recent e-mail, "Postdoc [positions] are a dime a dozen."

Note that I said shouldn't, not won't. The demand for postdocs and the amount of money available to support them varies from field to field.

A prototype is emerging, at institutions that are progressive enough to think about these things, for what the postdoctoral experience ought to look like. It goes something like this. When you're first hired, you're supported by research grants or institutional money. Late in your first year, or during your second, you apply for a postdoctoral fellowship. That fellowship pays your way for the rest of your postdoc.

To conclude: Should you apply for a postdoctoral fellowship while still a graduate student? If you've got time, sure, go ahead. But it's a seller's market. If you are a strong student with a sound publication record in a research area of some importance, you shouldn't have much trouble finding one--or more--postdoc offers from laboratories that are willing and able to pay you. You are a highly skilled worker and a great bargain at prevailing postdoc rates.

Okay, so there are postdoc positions out there to be found. How do I choose among them?

Look to your future with two words: due diligence. It's a cliché these days but a useful one. In case you aren't familiar with the term, it came from the legal world and gained notoriety during the Internet boom when venture capitalists entered the public consciousness. Professional investors have a process for evaluating investment opportunities--typically young companies--before they put their money down. Every early-stage investment is risky, goes the thinking, but there are some risks you can mitigate by doing your homework. As Michael L. Baird puts it in Engineering Your Start-Up:

Investigate and master all the success-contribution factors that will make or break your business: management team, board of directors, markets, customers, products, business plans, and fund-raising.

That's a tall order; none of us ever "masters" the intangibles that go into determining whether we succeed or fail. It's all about minimizing risk.

How does this translate to the world of science? Investors look for viable companies with opportunities for substantial growth. Scientists-in-training look for viable career paths with opportunities for future career development and the appropriate balance of short-term and long-term satisfaction.

What's this last bit about? Choose the postdoc lab that maximizes your chances of long-term success (however you define it) without requiring that you sacrifice too much in the short term. Sure, your future success may be worth some sacrifice (it's up to you to decide how much), but postdoc sacrifices are often gratuitous. There's no good reason why a PI should ask you to make sacrifices--the suffer-for-your-science model is old-fashioned and discredited. Suffer for science if you want to, but realize that you don't necessarily have to, and that suffering won't necessarily help you become a better scientist.

Take the safe road: Choose a well-funded lab with a clear scientific focus and a record of turning out publications and successful scientists. And talk to the people in the lab, as well as lab alumni, to make sure that their lives aren't miserable (or weren't when they worked there).

Does that mean I should never choose to work for a new assistant professor?

This is a tough question. If you've found an assistant prof you really like, and you find her work exciting, then go for it. A personal affinity to the PI and the work can outweigh all sorts of disadvantages. The importance of a good fit cannot be overemphasized, and only you (and the PI) can make that determination. Another factor is the PI's potential; a young PI with a great pedigree and a big, shiny, new, highly scored research grant might be an excellent risk. Finally, younger professors may, in general, be more open to alternative careers than their more senior colleagues. But as a general proposition, working for a scientist without a track record is risky. Make sure the potential upside balances the risk.

What do I need to get out of a postdoc?

You can use the following checklist to assess your offers.

· Respect and professional conduct from your postdoc mentor, now and in the future.

· A reasonable paycheck and benefits, including, especially, health insurance for you and your family. You may have to pay extra to get family coverage, but they should have access to a subsidized group plan.

  • Good publications and appropriate authorship status.

  • Money to travel to conferences and opportunities to network and present your work.

  • A degree of independence, especially in the later years of your postdoc.

  • Strong letters of recommendation (assuming you earn them by doing good work).

  • Training and experience in grant writing, paper writing, lab management, and other career skills.

  • The opportunity to take some of your postdoctoral work with you when you achieve independence.

A couple of these points require some explanation:

Authorship: If you did the work, it is reasonable to expect that you will have the opportunity to write the paper. If you did the work and wrote the paper, then you should be first author.

Taking work with you: Some PIs are very liberal here; others are stingy. Don't expect too much and be willing to negotiate.

In an ideal world, you'd get a formal offer letter that would spell out lab policies on these matters. In the real world, this isn't very likely--not, anyway, for the less tangible issues such as authorship and independence. You certainly have a right to expect--but may not get--an offer letter that spells out benefits, holidays, and work-schedule expectations.

What practical steps can I take that will help me increase my odds of a good postdoctoral experience?

Ask for a copy of the postdoc-PI's CV. Some PIs may be taken aback by such a request, but once you politely explain that you wish to learn more about their work, they will, very likely, provide it happily.

What should you do with this information? A CV is, or should be, a record of the PI's whole scientific career. It should list publications, with complete author lists; lab alumni and what they're doing now (although this part may not be up-to-date); and the PI's funding record over the long term, including recent grants. The usefulness of this information may be obvious, but for the sake of complete clarity:

  • The publications list will tell you whether the PI is well-published, if he/she has been publishing a lot recently, and whether lab staffers are typically listed as primary authors.

  • The personnel list will tell you whether alumni of this lab do well after they leave. Track some of them down and call them up. Don't e-mail, call. People are more candid on the telephone. If there's no such list, use the publications list and call your prospective PI's co-authors.

  • The funding history will tell you whether you're likely to lose your support in a year or so.

See this interesting article for more on "due diligence."

I've been offered a position in the laboratory of Professor X. But he doesn't have any money to support me. What are my options?

I can think of two reasons why this might happen. First, in a research climate that is (in relative terms) awash in money, the scientist who made the offer doesn't have any--not a good sign. Or the scientist has money but has already offered it to other scientists, whom he or she deemed worthy of an actual financial commitment, in contrast to your good self. Not to put too fine a point on it, then, but one of your options--maybe the best one--is to look elsewhere. What sort of mentoring can you expect from a PI who is willing to take your labor but unwilling to pay you?

Enough pontification. It's a very good lab, and you want that job anyway. What are your options? You can work for free, or you can apply for a fellowship. The National Institutes of Health (NIH) is the major source of biomedical fellowship money. In the physical sciences, the National Science Foundation (NSF) and the Department of Energy (DOE) are the big spenders. But other federal agencies exist, as do literally hundreds of (mostly biomedical, typically disease-specific) foundations that offer fellowships. Check GrantsNet for some ideas.

I'm from overseas. I've been offered a position in the laboratory of Professor X. But he doesn't have any money to support me. What are my options?

See above for my views on PIs who offer positions without money. But assuming you're still interested after reading that, read on.

U.S. federal money is for U.S. citizens. NIH, NSF, DOE, the Department of Homeland Security, and so on reserve their fellowship money for Americans. Fortunately, those disease-specific philanthropies I mentioned previously often support scholars from abroad. Your options, although more limited than those of your citizen colleagues, are not insignificant. So you might also want to search GrantsNet.

What can I do to make the postdoc I choose as good as possible, even if my choice proves imperfect?

Your choice will prove imperfect. Count on it. But it could still be an excellent experience. Be professional and have a thick skin.

Speaking of imperfection: What did you mean when you said, at the beginning, that there's no such thing as a perfect postdoc?

Every situation is different. It really is all in the fit. The perfect postdoc for one person may be a disaster for another. A few really good PIs have strong management skills that allow them to build productive relationships with many different types of scientists, adapting to the various personalities in the lab. But this kind of flexibility is rare among PIs. Your postdoc choices may well be limited to mentors with distinctive yet homogeneous and unwavering management styles. One may be autocratic, demanding, and (by the standards of some) uncommunicative. Another may be motherly, supportive, and (by the standards of others) meddlesome.

Choose well now and prepare to adapt your working style to the management style of your postdoc PI. Do some introspecting: Do you deal well with frank criticism, or do you need a bit of soft-pedaling? Very demanding PIs may work well for very thick-skinned postdocs, but an opinionated, strong-willed PI could spell disaster for someone who needs more nurturing. You have to decide for yourself which category you fall into.

Comments, suggestions? Please send your feedback to our editor.

Jim Austin is the editor of Science Careers. @SciCareerEditor on Twitter