In the late 13th century, Pope Benedict sent a courtier to Tuscany to check out the art scene and recruit an artist to do some work at St Peter's. The courtier collected drawings from the greatest masters in Sienna and Florence and one morning went into the workshop of Giotto. When asked for a small drawing to show “the big guy”, the artist dipped a pen in red, braced his arm against his side, and in one clean motion of his hand made a perfect circle. “It is too little and too much,” he told the courtier, who, despite suspecting he was being duped, included it with the other drawings when he presented the collection in Rome.

Giotto got the job. And even today, his `O' resonates. What style, what panache, maybe even a bit of theater. One quick gesture, admittedly a perfect one, and the greatest opportunity of the age was his. That's the way to get a job!

O, please. Vasari (who told the story a couple of centuries later) had a good sense of the dramatic - his account even begins with Giotto, as a boy, being discovered drawing in the dirt and recognized for his talent. But let's get real: an O, no matter how perfect, from Guiseppe, Maria or Ed would have been left behind in the courtier's hotel room. Giotto got the job because, by the time he was considered, he'd already done brilliant work in Assisi, and he was known throughout the region for being one of the best. The `O' (if it ever existed) was secondary.

Here's my point. You're not going to get the job you want because you gave a great talk or had a super interview, or even because you published a really nice paper. All of these things are important (and can be hugely important) but they aren't enough.

I'll start with the CV. This `Life Course', like a résumé, is a short outline of training, skills and experience, but includes something else: publications. When I evaluate a potential candidate for a job, be it post-doctoral fellow, faculty member or industrial researcher, or for a promotion to tenure, I skip all of the stuff at the beginning of the CV and go right to the heart: the publication list. I look at what, where and when things were published, and I have a look at the papers. There are other things I care about (is this a person I can enjoy working with?), but the first cut is always on publications.

And let's get something clear: all publications count. I'm astonished when a post-doc tells me that someone else has told them that publications on which they are second or third author and review articles are irrelevant. This just isn't true. It is true that, in the absence of primary publications (first or last author, depending on the job), the publication list will lack punch, but it is also true that a list that includes only these says “does not play well with others”. Primary papers and a good sprinkling of the other types of publication tell me that you've got a good, balanced performance record and might be an excellent person to consider. Is this you (if you're not sure, ask around)? If it isn't, you may need more time - perhaps find another lab that has a good publication record and add to your accomplishments.

Assuming you've got a publication list we are happy with, now put yourself in my place (I'm the guy doing the hiring). Why do I want you? Because you're going to be a terrific scientist? Well that would be fair, but often it isn't enough. I want to hire someone who is terrific but also going to help me. Will you provide me with technologies that will ennoble my research program? Will you collaborate with me on exciting new avenues of exploration? Will we synergize?

How will I know?

I'm going to know because you're going to tell me. And you're going to tell me because you've spent the time to figure out what it is that I work on, and the guy next door, and the woman down the hall. Yes, you're going to do your homework, so that when we sit down to hear about your work, we'll see all the wonderful possibilities. That's right: job talks are different from seminars. They are about the future not the past (although of course the past makes the future seem possible or even likely).

If you've never heard a good job talk, that makes our mission here harder but not unmanageable. Remember, you're trying to convince a handful in the audience (sometimes only one or two) that you're the one we want. Presenting good data is a great start, but you've also got to show us that you've got a program that's ready to rock. Don't simply present published work (we can read that). Tell us all about what you're planning to do - not just at the end but throughout the talk. Tell us where this is all heading. And remember a very important trick: be excited about your work. Who in the world should be more excited than you are? I'm astonished how often speakers forget this, as though showing enthusiasm for what they are doing is simply garish. If you can leave us dying to know how this will all turn out, then you're on the right track. And you might do something else in the process: you might make us think that it will be fun to work with you!

Afterwards, we'll sit down and chat, and you'll tell me exactly how we're going to do this together, how we'll synergize, because you've already thought about it. I'm ready to hire you already.

All this applies to a lab, a department, a company, a journal - pretty much any science job you can want. It isn't easy. But you must do it, because the next candidate in line will and, if you don't, that's who is going to get the job.

Your publications are your resume - the credentials that you bring to the table. But the seminar and the meetings with your prospective employers, that's your `O'. Make it perfect. Then, when you get the job you not only want but will also be good at, come back and tell me all about it. You're buying.