Writing: How Long Does This Need To Be?

In a comment on my first post here, someone asked about writing. Specifically, they asked whether the length of a piece was more important than its substance. I responded in the comments that I don’t think the length of a story/post/article is important at all. But I was thinking this over as I drove home last night, and that’s not entirely true. Sometimes word count matters.

There is a school of thought that writing for the web needs to be short. I don’t agree. Writing, for the web or anywhere else, needs to be GOOD. If you’re not engaging, people won’t read the piece, regardless of length. If your writing IS engaging, people will read it whether it’s 500 words long or much, much longer. And content, in my opinion, is king.

You need to explain to the reader fairly quickly why they should give a shit about what you’re writing. When it comes to science writing, you need to muster your facts. Not only do you need to explain the relevant science in terms your audience can understand (which will differ depending on whether you’re writing for an expert or non-expert audience), but you need to place those facts in context. For example, where do new findings fit in on the continuum of research in that field? What came before? What new directions do these new findings indicate for future research?

So where does word count come in?

If you’re writing a blog post, I don’t think word count comes into play. But a lot of writing associated with science communication is done outside of the world of blogs.

When I work as a freelance reporter, word count is important (you usually get paid by the word, for one thing). When you land a commission to write something, the editor tells you exactly how long your piece will be. Maybe it’s 500 words or maybe it’s 2,000 – but you know what you’ve got to work with before you even start typing. The trick is to make sure you get in all the essential details while still making it a story that people will actually want to read. (I’ll leave thoughts on how to write an engaging science story for another post.)

Now, as a press guy for a university, I face a different challenge. When I write up a news release about research findings I have to keep it short. The ultimate goal for a news release is to interest reporters so that they’ll want to follow up and write their own articles on the research. Reporters get approximately one zillion news releases and pitches per day. If you send them a 1,000 word summary of the research, they (usually) won’t even bother trying to read it. I certainly didn’t when I was a reporter.

I do two things to address this: 1) I keep the release short; and 2) I make the pitch even shorter.

I try to keep the release itself to around 400 words, trying to address only a few core questions:

  • What did the study find?
  • Why is that interesting or important?
  • What are the limitations of the study? (Honesty is crucial.)
  • And (maybe) what are future directions for the work?

But in order to get a reporter to bother reading those 400 words, I shoot them a much shorter email message with the following information: one sentence on why I think they’d be interested; a link to the release if they want more information; whether I have supplementary material, such as images or video; and the direct contact information for the relevant researchers (so they don’t have to go through me if they don’t want to).

This is much more effective than spamming reporters with the full news release right off the bat.

Summary: word count doesn’t matter, except when it does.


21 thoughts on “Writing: How Long Does This Need To Be?

  1. Matt, I agree and disagree. As a professional writer who has learnt the art to writing engaging stuff, it doesn’t matter what the word count is on the web. But as a person learning the craft, which many bloggers are doing today, keeping a word limit on the post helps you ensure quality writing. It forces you to cut down the unnecessary words, and, hopefully, make it worth the reader’s time.


  2. I can see the utility of imposing a word count as a training tool. However, I disagree with the idea that web writing — in general — has to be short.

    Also, taking the time to go over your work — and getting friends or colleagues to proof your work — is always a good idea (regardless of word count).


  3. Thanks for the thoughtful post, Matt. I think your title hits on a crucial point which I don’t think you touched upon in the text, though: ‘How long does it *need* to be?’, as opposed to ‘How long *can* it be?’. The fact that there are no space limitations online can sometimes make us a bit sloppy, I think, and result in texts that drag on for no apparent reason. It ties in with your point about content: whatever length your text is, make every word count.

    And if you’re not restricted in your word count, experiment and find out what kind of story you write best. Some people are great at all lengths, but a lot of us are not. I’ve worked with writers who wrote great, interesting 500-word pieces but couldn’t sustain a reader’s attention over 1500 words, and others who write beautiful long feature articles but have trouble making a shorter text work. If you have a blog, good news: you can experiment, and practice all sorts of lengths and styles!


  4. Agreed. The length of a story is no indicator of its quality. Should have included a quote from my first editor when I was just starting out as a reporter. I asked him how long a story should be. “As long as it needs to be,” was his response.


  5. Troy McConaghy

    Every time a new version of Mac OS X comes out, John Siracura writes a very, very long review and posts it on the web. It’s book-length. In fact, you can get it as an ebook! Yet it’s always a popular article, zooming to the top of many trackers. Therefore long doesn’t imply bad for the web.

    But! We’re scientists here, or at least science writers. In science, you might suspect something, but you need to back it up with data. Surely it must be possible to make a plot of “article length” versus “popularity” (measured somehow, e.g. by total page views, or total shares on Twitter, or total time-reading-article, or something). Maybe someone already did this. (Hint.)


  6. If anyone’s done it, I’d like to see it. But I wouldn’t consider myself bound by the results. There are a lot of bloggers out there, but many (most?) of them are not very good. So, the study would have to find a way to control for quality. Maybe sampling within-group? E.g., evaluating popularity of long versus short posts for each of a dozen (or more) “popular” science bloggers?


  7. Hmmm. What seems to drive blog traffic for me is not length, but topic. Anything involving pubic lice, bot flies, or bed bugs is a winner.

    This presents the opportunity for a controlled trial, if I can just figure out how to get all three insects covered in one blog story.


  8. Pascale

    The rule-of-thumb I have seen for blogs is 400-800 words. If it’s much shorter than that, think about it some more. If it’s a lot longer, it better be good.


  9. The myth that blog posts are short (or even worse, that they SHOULD be short) originates in the early days, when many blogs were “filter” blogs, posting links with just brief descriptions. That kind of stuff now goes on Twitter (or Facebook, G+ or other social network of choice). What remains on the blog is serious stuff. Serious stuff takes research, thought, time and energy to do. Bloggers who do not come from journalism do not have a concept of “word limit”. They write until they are done. Thus, the usual blog post is long. Often much longer than an MSM article on the same topic. Unlike a traditional article, the blog post does not follow the Inverted Pyramid, does not cut out important context due to length constraints, and usually includes stuff that only the expert can provide. Thus, a reader will check out the MSM article, but that same reader will bookmark and share the much longer blog post because the blog post is much more informative and contains information and context that turns it into a valuable resource.

    New York Times, Slate, and some individual bloggers have done the analysis. It shows that longreads do better in traffic.

    I did one such analysis myself a few years ago. Out of top 100 posts by total lifelong traffic (out of about 10,000 total posts I had at the time), 98 were long, some extremely long.

    Short posts, quirky stuff, videos, those things may get a spike in traffic, perhaps 60,000 hits from reddit in a few hours. That is worthless traffic. Those are hit and run readers. They may post crappy comments before leaving forever. They are not going to become your regulars.

    But long post accrue traffic slowly over a long period of time. They can keep gaining traffic over several years. They get rediscovered by new generations of online readers over and over again.

    By virtue of length, long posts have many words. Thus many combinations of keyword will bring up that post in Google searches. Bloggers who write such long, serious posts, tend to gain loyal readership – they tend to have something to say from a position of expertise. Thus, they gradually gain quite a high Google Rank. Such bloggers, being serious and expert, also tend to be invited to blog for MSM, or blogging networks hosted by MSM, which automatically adds to their traffic and Google Rank.

    Are some bloggers woo wordy? Sure. I have read many a long treatise by Orac, for example, who is notoriously long. Yet, I find it interesting. Also, I feel rude leaving his posts halfway through – he has earned his respect with me, so if I decide to read his post, I am committing to reading the whole thing. I know what I am getting into, so I will leave sufficient time to read the whole thing. I was told the same about my long posts. Often people leave them for later – evening or weekend – and bookmark them (there are several pieces of software or sites that allow this to be done easily these days). And they really read them whole.


    1. Bora, I take your points, but I do still think that you can’t just trust that readers will be ‘polite’ and read to the end – especially those readers who are finding you for the first time, and don’t feel any obligation to you or have any expectations of what your writing is like yet.

      Whether it’s in print or on dead trees, I’ll find myself ‘devouring’ a long text if it draws me in and keeps pulling me along to find out more. And I’ll drop it for something else if it doesn’t – very often thinking “I’ll come back later when I have more time for this”, but in the meantime my twitter feed/etc. flag up another dozen interesting stories, and that tab sits open on my browser for a few weeks until I finally end up closing it without ever having finished the text.


  10. Just to throw my particular axe down for grinding, walls of text can sometimes, I dunno, tire me. As Bora mentions, you can set it aside to read later – and I do, but perhaps my MTV/Star Wars quick cutscene upbringing makes me want to keep my twitchy hands moving. I doodle or feel like I should move on to something else, and then read in chunks.

    But, (and here’s that axe I intend to drop) visuals in a post may hook me for longer. The wall of text becomes a lovely mosaic of letters interspersed in a mural, and I wind my way down through it. It can be super long, but clever images interrupting, like a cut scene keep my GenX attention span far more easily.

    Some bloggers do this with a genuinely helpful image, underscored by funny captions (Scicurious comes to mind), others break it up with little images, like windows through the wall of text. I’m a visual guy. For me, it helps give the long posts more room to stretch out in my head.


    1. Your comment really got me thinking, Glendon, because when I read things in print I first read the whole story and then go back and take in the images and captions. Online, though, I think I do tend to take text and images as they come, and having images scattered throughout a post does, as you say, make it look less like a wall of text when I scroll down it to see how long it is and whether I’m going to bother to read it right now.

      The exception is when I run things through Instapaper to read offline (and in greater eye comfort) on my eReader: that seems to strip out all the images, and sometimes garble the captions into the main text. So posts that rely heavily on images just don’t work (and get annoying). Anyone know of a workaround/alternative to Instapaper that would allow me to keep those images?


  11. Thanks, Matt, for this interesting post. You’re off to an excellent start with this discussion and the thoughtful comments it has generated. It’s topical for me, since I’m new to blogging, and I like details so I tend to write long articles. I can say that the freedom that blogging gives me is very seductive, since I’m not used to having no limits. The post I put up today was over 2500 words and could easily have been split into 2 or 3 articles, but I didn’t want to break it up. Once I get used to the freedom I should become more disciplined.


  12. Michael Kenward

    Nice bullets Matt. They are pretty universal for all areas. I sum them up when talking to scientists about the media as:

    Who (did it)?
    What (did they do)?
    When (did they do it)?
    Why (did they do it)?
    How much (did it cost/is it worth)?

    The last one often gets overlooked. But I do write mostly about science and business.

    The idea that blog posts should be short is an extension of the notion that on-line readers have the attention span of a gnat and will not scroll down to the second screenful. That is the factor to look for in any experiments. Is it true? A simple bit of browser technology could test that.

    The issue of length in on-line writing first arose because, unlike paper, there is nothing there to say “That’s enough”. No end of the page to bump up against.

    The appearance of ever longer posts owes much to the death of the printed page, and the associated disappearance of subeditors. Most long and rambling on-line stuff is that way because too many writers are semi-literate and do not have the watchful eye of a curmudgeonly old geezer, for that is what many were, saying “That’s enough, Ed”.

    Then there is the belief among many writers that they are so important and influential that everything they write is a gift to humanity. Me, I’m just a hack who has been around the block a few times.


  13. Hey Michael, as a fellow hack who has also been around the block a few times, I know where you’re coming from. I often think back to the remark my first editor made (mentioned in above comments) that a story should be as long as it needs to be. But how long is that? I rarely, if ever, post anything without getting a second (and preferably third) set of eyeballs on it. Good editors can spot mistakes, but they can also ask: are you SURE you want to be that self-indulgent? 🙂

    And you’re “Who/What/When/Where/Why — and HOW MUCH” was written, word-for-word, on a piece of paper I had taped to my computer when I worked in a newsroom. It was right next to an index card that my editor taped up that said “Why do I care?”


  14. Michael Kenward

    “Why do I care?”

    Nice one. All too easily forgotten. More broadly, anyone writing anything needs to ask the question “Why should anyone care?”

    On writing, one not so curmudgeonly sub editor I worked with, who later went on to become a leading light at Nature, had a subtle approach. He would not harangue the writers and editors he worked with. He would quietly approach them with an “I don’t understand this bit.”

    It didn’t take many of those to persuade the person on the receiving end that they needed to do something about the words they were processing.

    Try that on-line and you could end up in the middle of a flame war. Too many writers are convinced that THEY CAN WRITE.


  15. Tom Webb

    Nice post Matt, and I take your point. However…

    As a reader, I have to read a lot of papers (because that’s my job) and I like to read books when I have time. Blogs are something I browse in idle moments, of which I have too few, and so I very very rarely read anything longer than c 1000 words – i.e., I’ll scroll once, but seldom more. Most of the people whom I know read my blog are equally busy and have similar habits. So when I’ve written blog posts which stretch much over this, I’ve either edited them (which usually does the trick), or, if I really feel like I’ve got more to say, I’ll split them into two posts.


  16. I think the optimum is a something lengthy and substantive but that can be distilled down into something short and memorable.

    Like Bora, some of my most popular posts have been long. But they needed a super-short description encapsulating the point before they spread by recommendation.

    This post describes how a paragraph of standard academic prose suddenly jumped to a conference meme after getting distilled down to its four word essence: http://neurodojo.blogspot.com/2012/10/the-heilingenberg-rule.html


  17. In my view, a blog post should be as long ‘as it needs to be’ – there’s no absolute word limit beyond which it’s “too long”.

    But… an ideal blog post doesn’t need to be very long. A post should encapsulate a single idea. What this means in practice with science blogs is that the natural length ought to be fairly short, say 300-800 words very roughly.

    If you have several ideas, split them into several posts. This is easier for the reader and it also aids you as the writer, because it helps you to organize your thoughts.

    My philosophy is that if it could reasonably be split into two, it should be.


Leave a Reply

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

WordPress.com Logo

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

Google+ photo

You are commenting using your Google+ 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