If you (or a global pandemic) take the bench away from the scientist, what do they do? They write reviews of course!

As many of us are now far too familiar with, crafting a review article presents a series of unique challenges. Unlike a manuscript, in which the nature of your data inherently shapes the narrative of the article, a review requires synthesizing one largely from scratch. Reviews are often initiated without a well-defined scope going in, which can often leave us feeling overwhelmed, like we’re faced with covering an entire field.

With these challenges in mind, here are a few tips and tricks to make review writing as painless as possible, for the next time you lose your pipette:

  • Do: Push to refine your article’s broader motivation as much as possible, and as early as possible in the writing process.
    • We often think of reviews as being a kind of scientific op-ed. You aren’t just giving a summary of papers, but you are putting your scientific view or story out there. That is often why review articles come out right after major papers. They can help to put your ideas into a larger context. Reviews are not only a survey of the field but one that is driven and framed by a particular viewpoint. [But how does this differ from a perspective?]
      • Defining this viewpoint can be extremely helpful in limiting the scope of your literature search, preventing the overwhelming feeling of having to read every paper ever — focus your time and energy on deep-dives into those papers most important to this motivating viewpoint.
    • To this end, we advise thinking about the “Who” and “Why” of your article, which will then define “What” you will be reading and writing about. At a minimum, you should be able to identify an approximate target audience, of which you are likely a card-carrying member. [Don’t feel like you’re holding a card quite yet? That’s okay!]
      • Ask yourself: Who do you want reading your review? What could you cover that would be most helpful to them?
      • This will be an iterative process — the focus of your review will likely change significantly over the writing process, as you read more papers and start organizing your thoughts.
  • Do: Use other reviews as a resource, especially early in the process. Look for the gaps that your article will address.
    • This can be extremely helpful in continuing to refine your article’s overall motivation, as defined above.
      • For each review, ask: What are their take-home messages? How can you differentiate your own from each of these?
      • As a member of the field, look out for things you wish they had covered: “I wish they had a figure on this, I wish they discussed this, I wish they clarified this…”
    • Take note of their references (and lack thereof)
      • Are there key papers that they missed?
      • Are there key papers that have been published since these reviews have been published?
    • Cite other reviews to save yourself some writing! If a tangentially related topic is outside of the scope of your review, it’s commonplace to reference other reviews for the sake of brevity, and to recognize their hard work: “X is outside of the scope of this review, but is covered in-depth here [Ref]”).
  • Do: Identify a handful of key recent papers that you will focus much of your discussion on.
    • Base your choices on the gaps you identified in other reviews, on papers published so recently that they haven’t been included in other reviews, or any papers you feel you can discuss from a unique perspective (even if other reviews have covered them).
      • For each paper, ask: What was known before this paper, what did this paper show, and what are its limitations?
    • Be strategic and purposeful when you are reviewing the literature, reading each paper through the lens of your review’s specific motivation.
      • It’s important to accept the fact that it is impossible to read, let alone discuss in-depth, hundreds and hundreds of papers.
      • Depending on how each paper will fit into your article’s narrative, it may only be necessary to review specific sections or figures. [I don’t have to read every word of every paper?!]
  • Don’t: Skimp on the outlining process
    • Given the unstructured nature of a non-data-driven article, this is a hugely important step in the process that will make writing infinitely less painful.
    • Try to get as granular as possible — the finer you drill down here, the less work you’ll have to do once you start writing.
      • Which key papers are you going to discuss in which sections?
      • Outline subsections and transitions under each major section.
    • Engage with your PI early and often in the process of crafting your outline, and try to get explicit approval of the finished product before you start writing — this can save you from a lot of painful backtracking later!
  • Don’t: Be discouraged if your outline keeps changing as you get further into the writing process
    • Writing and structuring your review should be iterative as you continue to refine, read more papers, and start to actually get words down on the page
  • Don’t: Summarize the results and rehash the discussion of papers you are citing
    • The most helpful reviews synthesize the findings of multiple papers into a cohesive take-home message.
    • Think about how specific findings relate to your overarching motivation for this article
    • Think about how different papers relate to each other — do different studies align, or do they contradict each other?
  • Do: Think about the figures early and often, and do them justice!
    • Keep in mind how people generally skim articles, by skimming the figures — reviews are no different
    • Planning the figures and how they will fit into the broader framework of the article is as important as outlining the text
      • Figures should be included in your structural outline
    • Treat these figures as a potential future resource for you/your lab — try to create something that will be useful outside of the scope of this review as well
      • For example, many people pull schematics from their own reviews to use directly in background slides of future presentations
  • Do: Diversify the list of authors and labs that you are citing
    • Do not narrow your literature search to only include papers from larger journals
      • While you cannot avoid citing and discussing major, high impact papers from larger journals, consider that these have likely already been discussed in great depth by other reviews given their high visibility. Good research exists in smaller journals, and you can do your part to cast a light on this work.
    • You can provide a fresh perspective by looking outside your field for analogous research, provided you can find a creative way to fit it into the scope of your review’s narrative.

Blog post written by Caleb Perez, with input from Tyler Toth, Viraat Goel, and Prerna Bhargava.


Reviews versus Perspectives- It’s important to draw the distinction between reviews and perspectives here. Although we believe that both should review the field in the context of some overarching scientific viewpoint, perspective articles allow the author much more freedom to craft a more opinionated argument and are generally more forward-thinking. If you have that freedom, definitely use it!

Belonging to a group- Of course, the extent to which you can do this may be limited, depending on how familiar you are with the field. First-year graduate students getting into a new field, for example, may not have as great of a grasp on the gaps in the field — you may have to lean on the advice of your PI and colleagues to help guide you here, especially in the early stages of the process before you start your in-depth literature search.

How to read a paper- There are many situations in which a narrower, targeted paper review is warranted. As one example, imagine a section of a review in which you are comparing different technologies for application X. In this context, you may only need to do a detailed review of the methods sections and any figures they have that benchmark their method for your particular application of interest. The rest of the paper is less relevant, so there’s no need to waste your valuable time and energy.