Best practices and success principles for PhD students and early career researchers

Whenever a new student, intern, or research assistant joins my lab, I often find myself repeating a lot of our lab’s best practices. One of the most common and most frustrating occurrences is receiving an email entitled ‘Dear prof, please read through the attached paper, the deadline is midnight tonight’.

My experiences as a PhD student at the University of Antwerp in Belgium; in my current position as the head of the AI for Audio and Music (AMAAI) at Singapore University of Technology and Design (SUTD) (co-founded by MIT), and director of the SUTD Game Lab, have inspired me to write down some best practices for the early career researcher.

The landscape of research has changed a lot over the past few years. We do not typically spend whole afternoons in the library archives looking for that one dusty article.
Instead we are overloaded with information through a myriad of digital services, all offering high quality manuscripts at the click of a button. Along with these technological advances, the challenges faced by PhD students have drastically shifted over the last decade. In addition, more and more students opt to enroll in graduate school [3], thus making the competition for getting a job and making an impact that much harder.
This competitive landscape has been known to put psychological strain on graduate students [6, 5]. In this article, we offer best practices, or principles for success, for the modern day PhD student and researchers.
Being successful as a PhD student can entail a bit more than just publishing [2].
Throughout graduate studies, it is important to maintain a good relationship with your supervisor, while doing impactful publishing, building up a network to leverage your work, and a myriad of other small things that are vital for your future career. We will go over some simple best practices that will help you become a great PhD student and impress your peers as well as your supervisor.

  • Avoid sending this all too common email to your supervisor: `Dear prof, please read through the attached paper, the deadline is midnight tonight’. Always alert all co-authors on manuscripts well in advance before submitting them. While supervisors love receiving news that a new (draft) paper is ready, a better approach would be to share conference manuscripts at least two weeks in advance so that your senior co-authors can plan their time. In the case of journal papers, a much longer time frame is needed to properly structure the manuscript and jointly prepare a writing plan.
  • As a junior member of the team, pair up with a more senior member (preferably a postdoc) for close collaboration. While professors are happy to assist you, they have many responsibilities in the university and are not always able to help with low level coding and small decisions due to time restraints. Communicate very regularly with this person, and report on larger milestones with your supervisor, so as to maintain a proper research direction.
  • Thoroughly proofread your manuscript before sending it to a supervisor or co- author. Take a course in academic writing and polish your English skills. If English is not your first language, enlist a proofreading service or leverage technology and use tools like Grammarly in addition to the former. This way you can get the most out of your professor’s time, after all, you want them to give you research input, not spend time checking spelling.
  • All authors listed on a manuscript need to have made a significant contribution to the work. This could be through coding, ideation, editing, advising, among others. Do not just include friends as co-authors. In addition, everyone needs to have read the final version of the manuscript before submitting.
  • If your manuscript contains software code, make it available, either through open source or with a limited licence (although open source will get you more users!). Whether you make your code available online or not, also keep a private repository that you share with all your fellow lab students. This will enable your teammates to leverage your work. Labs should be collaborative, not competitive.
  • To preprint or not to preprint? Due to the slow turnaround of traditional publishing channels, and the fast pace of modern day research, it has become very popular to make a preprint available as soon as your manuscript is accepted. Preprints can be published through the preprint service available in your university, or through popular preprint archives such as ArXiv. Clearly indicate, however, the original citation information in a footer in the manuscript as soon as your manuscript is accepted by a publisher. This way you direct your citations to the correct (peer reviewed) manuscript. Early career students may not yet realise the importance of citations, but they are key to advancing your scientific career and proving that you do impactful research. One could argue that it is better to have one paper with 100 citations, than a preprint with 40 citations and a paper of 60 citations. Finally, always check beforehand if your publisher allows you to make a preprint available.
  • Publish in impactful journals and conferences that are relevant to your research topic. While there is a lot of controversy about this, the leading evaluation metric used by many universities today is the (field normalized) impact factor of a publication, as defined by leading organizations such as Clarivate and Scopus. Until the dominant evaluation metrics change, high impact publishing will create the most opportunities to advance your career. In many fields, conferences are mostly for learning about latest advances and building a network, hence they only require short abstracts. In computer science things are a bit different and longer papers are required to present at conferences. If you go decide to prepare a paper for a conference, always check beforehand that it has a high CORE ranking: A* is the most prestigious ranking a conference can get. Even when you publish your work in a conference, consider expanding the paper later on and submitting a more complete study to a journal.
  • Use a collaborative writing tool for your manuscripts. This will facilitate feedback from co-authors and joint-writing sessions. If you are in a more technical field, write all your manuscripts in LaTeX. LaTeX offers many benefits, including easy citation management, stable working environment (does not crash), easy cross referencing, automatic labeling, and layout optimization [1]. Online tools for collaborative TeX writing include Overleaf, Papeeria, and Authorea.
  • Attend conferences and network. Building a network is essential to expand your collaborations and to get future job offers! At conference you will learn about the latest developments in your field. If you volunteer, you often get free access.
  • Share your research on social networks, through blog posts, and academic networks such as Scopus, ORCID,, Mendeley, and researchgate. Citations are important for your future career so you need to get your research out there.
  • Become a reviewer for related conferences and journals. This will give you access to the latest research in your field before it is published! When you review manuscripts, provide thorough reviews (not just a few sentences). Remember, your own manuscripts will be vetted thoroughly as well.
  • Join professional organisations such as IEEE (engineering) and ACM (computer science). These will give you access to valuable resources such as magazines, paper archives and tutorials.
  • Keep mentally and physically fit. Exercise helps keep your brain plasticity high, i.e. making it easy for your to learn new things [4].
  • Maintain fixed office hours. In a 2019 survey of 6,300 early-career researchers [7], many of them indicated issues with work-life balance. There are often students, who start their research at the end of the day, much like they were studying for their undergraduate degree. This will not work in graduate school. Maintaining fixed office hours also creates an end time, after which you can go out and relax without guilt. Granted, you may have to keep longer hours, especially before deadlines, but a strict schedule will help you stay mentally healthy and get an early start.
  • Own your project. When you meet with your supervisor, show up with all tasks completed, and ideally, the next step in mind or already started. Read a lot and be creative. Do not just show up expecting your supervisor to give you small pre-chewed tasks. Being a PhD student means doing research, finding gaps in the literature, and tackling those.
  • Be responsive to email. It will facilitate collaborations, make work quicker and help you get ahead faster

The above list is far from exhaustive, nor does it pretend to offer the absolute truth that works for everybody, they are based on the personal experience of the author.

Hopefully these pointers will help you on your way to build a successful career as a researcher. Good luck to everyone out there pursuing a PhD!

By Dorien Herremans, Assistant Professor at the Singapore University of Technology and Design, and at the Institute for High Performance Computing, A*STAR, Singapore


[1] Victoria Baramidze. Latex for technical writing. Journal of Technical Science and Technologies, 2(2):45-48, 2014.
[2] Anne-Marie Coriat. Phd merit needs to be defined by more than just publications. Nature human behaviour, 3(10):1007-1007, 2019.
[3] G Croucher. It’s time to reduce the number of phd students, or rethink how doctoral programs work. The Conversation, 18, 2016.
[4] KI Erickson and Arthur F Kramer. Aerobic exercise effects on cognitive and neural plasticity in older adults. British journal of sports medicine, 43(1):22-24, 2009.
[5] Hossein Jenaabadi, Naser Nastiezaie, and Hamideh Safarzaie. The relationship of academic burnout and academic stress with academic self-efficacy among graduate students.
New Educational Review, 49(3):65-76, 2017.
[6] Kendall Powell. Work{life balance: Break or burn out. Nature, 545(7654):375-377,2017.
[7] Chris Woolston. Phds: the tortuous truth. Nature, 575(7782):403, 2019.

Featured image credit: National Cancer Institute

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.