Good enough

Last year, when I wrote a post with advice on strategies (and reasons) for working more efficiently, the first strategy on my list was:

  1. Recognize what is “good enough”. As the saying goes, perfect is the enemy of good. And recognize that “good enough” will vary between different tasks. It’s okay if the email you are sending to your lab about lab meeting isn’t perfectly composed.

In this post, I want to go into that idea more, since I think it’s really important (and since it’s one I need to continually remind myself of!)

Recently, I was helping some of my lab folks with a project that required, among other things, taking lots of pictures of Daphnia. The picture needed to allow us: 1) to measure her length (which requires seeing the eye and the base of the tail spine clearly), and 2) to see her coloration (as we were interested in how that changed over the course of infection.) There were a lot of Daphnia who needed their photos taken (I’ve already forgotten the exact number, but I think it was about 125), and we had to do other things with the animals, too. So, one thing that was important was that we get the photos taken relatively quickly — which is part of why I was helping out.

As I sat in the lab taking the photos, I got into a routine where it was taking me about 1 minute to take each photo . . . but only if I managed to keep myself from obsessing over trying to get the perfect photo. In this case, this photo:


(look at all that crud in the background!) or even this photo:


(ack! there’s a fiber on her!) was just as good as this photo (taken on a different day with an older animal):


In fact, those earlier two photos were even better for this purpose, because speed was important, both so we could get all the photos taken in a reasonable amount of time, and because leaving the animal on the slide for a while to get a pretty photo would risk stressing her out.

This was hard for me, because I love taking pretty Daphnia photos, and am proud that several photos I’ve taken have been on journal covers. But I had to keep reminding myself that, in this case, perfect was the enemy of good.*

This general idea is one that comes up a lot in science, and it’s a really important one. A key place where it applies is when writing. With writing, instead of perfect being the enemy of good, perfect can often be the enemy of done. For some people, trying to write something that is perfect makes it so that they don’t get anything written, which is a major problem. Sometimes, people end up paralyzed because they feel like nothing they write is good enough, and so end up continually deleting what they wrote (or never getting started in the first place). To tackle this, we use a strategy in my lab that we call ‘barf and buff’ (or, sometimes, ‘barf then buff’). It’s not an elegant name (and it’s not one I came up with, though at this point I don’t recall where I first heard it), but it gets the point across. Let the words spill out onto the page (however they come out), then polish those. Or, as the author Jodi Picault put it when asked about how she fights writer’s block (emphasis added by me):

I don’t. Writer’s block is for people who have the luxury of time; I started writing when I had three kids under the age of 4. I used to write every ten minutes I got to sit in front of a computer. Now, when I have more time, I function the same way: if it’s writing time, I write. I may write garbage, but you can always edit garbage. You can’t edit a blank page.

Okay, so perfect is the enemy of good (or done). But if you don’t have to be perfect, how do you know what is good enough? That requires thinking about your goals (and audience). As one example: If you are sending a manuscript to Science or Nature, you will likely spend a lot of time on the cover letter, as you are trying to convince the editor that your study is interesting and worth sending out for review. If you are sending a manuscript to a journal that sends pretty much everything out for review, spending days crafting the cover letter would be a waste of time.

To give a different example: when reading papers, sometimes you really want to dig into the manuscript deeply — perhaps because you are reviewing it or because it is closely related to your research. But other times, you might be reading it just to figure out one piece of information — say, what spore dose people use when working with a particular pathogen, or whether a particular organism is found in a particular location. It was based on reminding myself of the importance of “good enough” that I decided to give up on my #365papers goal for this year. I realized that it was making me feel like I needed to read all papers thoroughly (because I had set that as the bar for what papers “counted” towards my goal). That meant I was starting to spend time reading papers more carefully than I really needed to read them so that I could add them to my tally, rather than focusing on what was good enough for that paper. By stopping tracking what I was reading, I felt more free to read papers as much (or as little) as I needed for a particular goal.

In my experience, the concept of good enough is one that is really important for students to learn. Perfectionism can be a real problem (even though people often talk about perfectionism as a good thing). Trying to do things perfectly can be crippling (and associated with anxiety and depression). You will not design a perfect experiment. You will not write the perfect paper. Accepting that and recognizing what is good enough is a very important part of being a productive scientist.

Of course, figuring out what is “good enough” might not always be straightforward and can take time and experience. Hopefully this is a place where mentors can help if you’re unsure. But, in the absence of feedback from others, two things that help me are: 1) thinking about my goals (as I discussed above), and 2) thinking about what the consequences would be of missing the mark and not being good enough. Often, when I think more about what the consequence would be of missing the mark, it’s not nearly as dire as I initially imagined.

And with that, I think I’ll say this post is good enough. 😉

* Interestingly, at the time that I took those first two photos, I thought the pictures were really not good. But when I went back to them for this post, they don’t seem nearly as bad as they did to me at the time. Conversely, the “good” one doesn’t seem as good. That part often happens — I’ll be happy with a photo after spending a while trying to get a good one, but will be less happy with it when I come back to it later.


28 thoughts on “Good enough

  1. “”Barf then buff” is a good strategy. The similar aphorism I use with our students is: “Don’t get it right, get it written”. As you rightly say, you can’t edit a blank page.

  2. I sometimes think this is the #1 lesson I end up having to teach graduates students. We’re all survivors of a filter that filters for perfectionism. And then we get dumped into grad school and faculty jobs where perfectionism is no longer valued (at least not as much as productivity). Indeed, I sometimes think of having a finished thesis as the graduation mark of having moved past perfectionism (it sure holds up a lot of theses anyway).

    I often think about this as finding the 80/20 spot. Have I reached that place where I’ve got 80% of the quality with 20% of the work. Because that remaining 20% of the quality is going to take 80% of the work (4 times more than I’ve already invested) and is almost never worth it.

  3. Anne Lamott’s Bird by Bird is great for writing advice like this and helped me get through my PhD. She describes Shitty First Drafts which are exactly what you’re talking about – her example is finding the starting sentence of her next draft on the second page of her first shitty draft.

    The ‘bird by bird’ advice (so appropriate for my dissertation) was to break the work into steps and just start doing them (based on her brother needing to write a school report on some group of birds and he just needed to do it, bird by bird).

    • Thanks for the recommendation! I’ve been thinking of having the lab (or others) do a “how to write” reading group where we each read different books and then compare and contrast strategies. It sounds like I should add Lamott’s book to the list of options!

      And, yes, breaking things down is so important. “Write dissertation” is way too daunting a task. “Write Chapter 2” is, too. “Make Figure 2.1” is much more manageable!

      • This is also a big premise of the Getting Things Done system of David Allen. All projects have concrete “next steps”: things that should be doable at one setting. When you put down a project, write down the next step. When it’s a writing project, try to leave it in a place where the next step starts easy. Beginning a tough journey is easier rolling downhill.

        Anne Lamott is a patron saint of all beginning writers. When I post doc’ed at Livermore, and she lived nearbye, I cold-called her just to say how much Bird by Bird meant to me. She was very gracious, even when I began telling her tales of Reviewer 2.

  4. Meghan, I think this is an important message – as somebody who, by nature, tends to err on the side of sloppy (at least by scientist’s standards) I’ve often felt that the dangers of perfectionism were rarely acknowledged. I’m not proud of the fact that I’ve occasionally submitted work that could and should have have been labored over more diligently but, it’s always bothered me that the down side of being a perfectionist is rarely discussed. When people say “Geez, I have to stop being such a perfectionist.” it’s often a humble brag rather than a sincere admission that there is a sweet spot and they haven’t found it. The fact that I can’t figure out a humble brag version of “I’m so sloppy.” says, I think, something about how we perceive the two sides of this issue.
    And a related point – it’s often struck me that people interpret what I think of as superficial errors (e.g. typos) as evidence of sloppy thinking – as if, rigour about the presentation and rigour about experimental design and interpretation are unavoidably connected. I’m not convinced that’s true – I would love to know if there’s any empirical evidence that they’re related. Jeff

    • “The fact that I can’t figure out a humble brag version of “I’m so sloppy.” says, I think, something about how we perceive the two sides of this issue.”

      Good point.

      Am now trying to think of what the humble brag version of “I’m so sloppy” would be. Maybe humble bragging about how many papers you’ve cranked out lately even though none of them is going to set the world on fire?

      “it’s often struck me that people interpret what I think of as superficial errors (e.g. typos) as evidence of sloppy thinking – as if, rigour about the presentation and rigour about experimental design and interpretation are unavoidably connected. ”

      Another good point. Meaning “point with which I agree.” 🙂 Yes, both in science and in other walks of life, I’ve never bought into the blanket argument that sloppiness or bad behavior in context X should cause others to assume that you’re also sloppy or behave badly in context Y. I mean, I suppose there might be a positive correlation in certain contexts, but my experience is that there are many exceptions, including contexts in which the correlation is negative. People are high dimensional and compartmentalized.

      • I feel like I’ve seen something talking about spelling ability, and how it doesn’t correlate with intelligence. That’s somewhat different than Jeff’s point, but along the same lines.

    • Yes, I agree that it can often be a humblebrag! I think it is because of it’s potential as a humblebrag that it’s a cliche answer to the “what is your greatest weakness?” interview question.

      I have strong tendencies towards perfectionism but work to keep them under control so that I can actually get things done. Early in grad school, I noticed that some people who were doing interesting work were never writing it up because they always felt that, if they just collected a little more data, they’d have a stronger story. But, if you keep collecting a little more data so you have a stronger story, you never end up writing up, which is bad for multiple reasons (your career, and not sharing your findings with other scientists).

      So, as you said, there’s a sweet spot that we need to aim for. I think it helps a lot to know which end of the spectrum you tend to fall on so you can know whether you should be more on the lookout for perfectionism or sloppiness.

      • Re: never writing up because there’s always a bit more data you could collect, Peter Morin once told a labmate of mine who was close to finishing “You can always do N+1 experiments, where N is the number you’ve already done.” His point being that at some point you just have to decide your thesis is good enough.

  5. I had to keep reminding myself of this post yesterday as my 5 year old helped me fold laundry. She folds the wash cloths and cleaning rags. She gets them folded in the general pattern that I use, but not as neatly. It takes all of my restraint some days not to refold them. But, really, what would be gained from refolding the cleaning rags so that they are in a neater pile? So, I just kept telling myself that it was good enough.

  6. And this advice will only be even more relevant if the recently suggested alterations to basic paper structure are realized:

    My experience and views on this topic differ though. Stream of consciousness writing, followed by x rounds of editing, doesn’t work for me. The underlying problem usually boils out to some degree of unclear thinking and I’m better off working to clarify my thinking beforehand than putting down a bunch of words and trying to correct them later. Few experiences in science are worse than trying to salvage poorly written crap, IMO, one of them being trying to salvage poorly written R code. Or as Bob Dylan said “I’ll know my song well before I start singing”, which is also exceptionally good advice when you’re in front of a crowd singing.

  7. At the end of my Behavioral Ecology class for 3rd yr students, I gave a lecture [ a bit tongue in cheek] on why the students did not spend as much time studying MY class as I would have liked.
    I quickly skipped to ‘ how the heck does one decide how much time to study’ any one of the 5 classes you have that term. Of course the answer [ by decree] was to treat it as an allocation of time problem, and to force the students to draw gain curves for the time put into class XXX, to add up the total gains, and to solve for the optimal time to give to each class…with the big constraint that the total sum of times must equal time available for studying [ we already took out party time!] The students found this quite amusing, particularly since the simple answer is to equalize the marginal gains for all the classes.
    I did not invent this game, an economist name Richard Mckenzie did:
    his final qualitative message……’ any thing worth doing is not necessarily worth doing well’.

  8. Great post! I generally agree with most of this advice, but I think that it may be also worth considering that the viewpoints and constraints of a trainee and a PI are often different, and the messages we send based on perception of perfection/sloppiness are important. For instance, if you spend time on the cover letter of one student’s paper destined for Nature, how will the other student who’s paper is destined for the Proceedings of the Milwaukee Beer Distributors B feel? While that may be the best use of time for the *PI*, I think it has to be weighed against what our goals are for our trainees. Another example: I have had people in the lab for whom English is not their first language (or otherwise have poor writing skills). I emphasize to them that *everything* they write, from papers all the way down to casual e-mails, is a chance to practice, and that they should always write properly. To set an example, I always try and write properly when communicating with trainees, including even the most casual e-mail. I think it does matter, but perhaps I’m just deluding myself to justify perfectionist behavior… 🙂

  9. This is one of the things I spend the most time on as a graduate advisor. The training we get that makes us good undergraduate students and careful student researchers, backfires when we hit those more self-directed career stages. Throw in a little anxiety and some high expectations (either internally or externally imposed), and you’ve got a recipe for disaster.

    So, thanks for this. I’m going to share it with my students, each of whom is cranking away on a draft of one kind or another!

  10. Very nice thoughts. It’s really important to avoid getting frozen by perfectionism. However, it’s also important not to write anything quick and dirty, in order to get rid of the task, and feel like getting the job done. As usual in life, the trick is to find the middle ground.

  11. My mantra while I was finishing my dissertation was the following: Lorne Michaels (producer of SNL) famously would tell writers: “The skits aren’t done because they are perfect. They are done because it’s 11 o’clock”. It really helped me see deadlines as finite and accept that it was NEVER going to be perfect.

  12. Nice post! There’s a quote, apparently from Da Vinci, that I really like: “art is never finished, only abandoned.” It’s a good description of how I feel when I submit a paper.

  13. Pingback: Recommended reads #88 | Small Pond Science

  14. Pingback: What if we make a class better for student learning but unsustainable for faculty? | Dynamic Ecology

  15. Pingback: Rough drafts, getting words on the page, and the pain – and pleasure – of writing | Dynamic Ecology

  16. Pingback: Comparison of ways of visualizing individual-level Likert data: line plots and heat maps and mosaic plots, oh my! | Dynamic Ecology

  17. Pingback: Some things that helped me make it through a particularly busy semester | Dynamic Ecology

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

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