Tips for Early PhD Students

Getting a PhD is about doing science

Getting a PhD is about learning how to journey into the unknown and come back with new information to tell the world. This involves getting lost or confused with results that don’t seem right— not the absence of results, but results that didn’t fit expectations. Uri Alon’s concept of being lost in the cloud is spot on (see his TED talk). Doing research is very different than reading papers or designing experiments on an exam, and your goal during your PhD is to practice working through unexpected results.

That great idea from an exiting grad student is probably better than anything you’ll come up with during your first year in lab. Do not be afraid to take a ripe project that was already designed or to take on a project as a co-first author. You might come up with a decent idea in your second year but truly great ideas are much easier to come up with once you’re already immersed in a field. Executing someone else’s great idea still will give you plenty of space to be creative. In fact, you’ll probably have more space to be creative when executing someone else’s great idea than when struggling with your kind-of-decent one.

Early success gives compounded results. If you hit the ground running, you might have promising data to show off to others early on (in your department/program or in publication). People will become aware of you and you can begin to build a brand. You will be exposed to more ideas. People will want to collaborate with you. You will get more opportunities to practice other skills of being a scientist— writing papers, giving talks, etc.

Spend a lot of hours in lab during your second year (or first year in lab). Most PhD students spend a year or so working around the clock, usually in their last year. I suggest doing this during your second year instead. Spend as much time as possible generating data, reading relevant literature, and talking to your labmates about science. If you can, put off teaching requirements and any additional classes until later. Make a friend who is in med school or law school (one with grades) and realize that you are not working all that much. There are at least 3 reasons to work long hours early instead of during your last year : (1) As mentioned before, early success will give compounded rewards. (2) Working hard early will benefit you if you need to switch projects; if a project looks mediocre after 4 years, it will be harder to switch than if you see the light after 2.5 years. (3) During your 3rd or 4th year, it is easier to keep research going with fewer hours in lab, but you probably won’t be able to do so in your 2nd year.

Think about the entire publication process as you as you choose and design your project. Will you be able to get a publication out of null results? What might your eventual figures look like?

Focus on questions. This is relevant even if you’re explicitly doing method development. How will your future results change understanding? How will they change what is possible?

Prioritize and optimize your time. Spending a day planning and strategizing can often save you weeks. A rule of thumb I like is to spend at least 20% of your time strategizing rather than doing (this includes planing experiments).

Picking a lab

Pick a lab that publishes the kind of papers you want to write. Would you prefer to write several short papers? One field-shifting paper? Choose carefully and keep your personality in mind.

If you work best when receiving external validation, pick short projects (e.g. you enjoy getting good test scores back) . A project that you think will take a couple of months is ideal –it will probably take a couple of years.

When choosing a lab, keep environment and culture in mind. Every factor about your environment can be important. If you love the bike ride into lab, it might be easier to keep up morale during those rough times. The most important factor is that you enjoy collaborating and discussing hypotheses/results with the members of your lab.

Working with people

Learn to write great emails. Have your advisor or a senior labmate check important emails to collaborators or people you need help from. Do this for each new type of email and even after you think you’ve got the hang of it. What makes a good email? A summary of what you want from the reader should be in the first sentence, bolded, or otherwise easily visible. Each paragraph should have a single point. A good email is positive in tone — try changing every “but” to “and” and you’ll notice the difference. Use as few words as possible. One strategy is to get out all your thoughts out and then restructure the email with the above points in mind.

Be generous and courteous to others. Working well with others is a requirement to do good work. Regardless of how intensively you collaborate with others, you will have co-authors, advisors, referees, editors, administrators, technicians, and many others to please. Be kind. Acknowledge that everyone is human. Have reasonable expectations. Use the improv “yes, and” policy (described really well in this section of Bossy Pants, by Tina Fey) when you can.

Learn how to get what you need from your advisor. Each advisor has their weaknesses. Usually, these weaknesses do not come from ineptitude or malice, but from lack of attention to some aspects of being a mentor that matter to you. If you need more feedback, figure out how to get it from your advisor.  If your advisor isn’t helping you network, ask. You may find that someone who seems to be terrible at X is quite good at X when they try. Or that you can create a system around their shortcomings— but you have to be deliberate about it.


Take ownership of your PhD. There is no simple procedure to follow that will result in success for you. Doing what your advisor says or wants may not be the best path. Be proactive. Set goals for yourself and think about how your choices are contributing to your goals.

Become an expert in your subfield. Read a lot. Go to a small conference or workshop if you can. If your project is a new area for your advisor, make sure you know more about it than him or her. Talk to people who know more. Seek out criticism.

Twitter is the best way to stay on top of your field [update: still true in mid 2018]. Create a twitter account JUST for science. Follow journals, blogs, and other scientists who tweet/comment on/retweet interesting papers. Start with one scientist in your field and look at their following/followers list to find more. Unfollow people who fill up your feed with things you don’t find interesting — if they write something great, someone else in your list will retweet it. This will give you something nutritious to read when you’re looking for a break.

Twitter can also help you network. This can be as simple as retweeting papers that you like. Other scientists in your field will notice and follow you back. This may take time, especially if you’re unpublished– but when your first publication rolls out you’ll have a brand and people will follow you.

Grad school is hard. Talk to older graduate students about your emotional and scientific struggles. Take care of yourself, and listen to your body if its showing signs of stress; your body isn’t just a platform for your head.

Face the reality that the probability of getting an all-star academic job isn’t that high. The landscape is competitive. You may get the academic job of your dreams if you work hard and make the right choices (some people will tell you its impossible; it isn’t), but there is no guarantee. Hone skills that are marketable outside of academia, like programming and writing. Even if you eventually stay in academia, the confidence and security of mind will make those existential tear-filled sessions about graduate school shorter.

Talk to your labmates a lot about science and the scientific process. Learn what everyone in your lab does, what techniques they use, and why. Be able to describe their projects, at some level of detail, to someone else from another lab. Talk to them in more detail after lab meeting. Ask them about that pretty graph they’re making. If your lab is small, do the same for another lab you like. It is an easy way to make sure you understand multiple techniques and concepts deeply — which will help you (1) come up with new ideas, (2) know what kinds of science you do and don’t like and (3) remain critical of work outside of your area of expertise. Your labmates are a great way to learn more about the different stages of inquiry and publication, so that you can prepare yourself for future hurdles.

Be proactive about getting feedback. If your advisor asks you to help draft a section of a grant or referee an article or similar — make sure that they send you the final version so you can learn from their corrections.

Don’t invest too much in common wisdom. Don’t invest too much in this list. Some pieces of advice get repeated so often that it seems barbaric to defy them, but really have little grounding.

How can this list be improved? Leave a comment!

Tips for Early PhD Students