Scrivener for Academic writers

I am a academic writer and keen Scrivener user. I love the structured approach, snapshots and split scrrens. Scrivener has however three weak points that seriously hinder its use for academic writing

  1. Tables
  • witing scientific paper without tables is not possible. I miss the possibility to copy paste tables , insert rows/columns, merge/split cells etc … Basically I have to switch to Word for tables and then except for reall ysimple ones to copy tham as pictures! Bad…
  1. Citations
    the same is true for citations. I use Sente for bibliography management on Mac and endNote on PC at my university … neither works fine with Scrivener. I have to compile the paper to Word, make citations list and then copy paste this list into Scrivener. The major problem is that a scientific paper undergoes many revisions - from colleagues, after review etc. Once compiled I can not use scrivener anymore for tracking changes to the citations that are made in this revision proces. I end up with two versions of documents. Draft (scrivener) and final (Word) each with a different set of citations. When I then want to start a new project and “reuse” my previous work i am stuck. I have to laborously add the citations again …
  2. As given above, academic writing encompasses a lot of collaborative work - comments, revisions, workng on the same document… Scrivener does not offer this …

Unfortunatelly unless this issues are solved Scrivener is not suitable for academic work beyond the point of a first draft .

Just a thought: think of Scrivener as a drafting tool.
Draft everything in Scrivener, and then output to the Word Processor of choice, scan the Sente references, and then work from there.
I suspect that’s how most people with requirements similar to yours use it.
Scrivener is for working towards that draft, and helps greatly with structuring and all of that. But then, send it to the Word Processor.
Myself, I don’t like looking at the manuscript in Scrivener once the first draft is done, because I don’t get the same sense of continuity. Looking at in in Mellel (or whatever) helps me to look at it from the reader’s perspective, and tweak the text.
As I say, just a thought.


This thread has a good explanation from Scrivener’s developer about tables.

I am also an academic writer and I can see your points. Yet, to me, the main limitation in using Scrivener more often than I do is the fact that my collaborators don’t use it. They all use Word. They even use Word to write a short abstract! The average academic that wants to write something, even something pretty trivial like a note, opens Word first! It’s almost funny. As much as I can’t stand Word, I find myself using it all the time. Even for drafting something, when I do want an immediate feedback from my collaborators.

But, if your collaborators do use Scrivener, then the limitations are minimal.

You can use multi markdown for tables. If MMD table handling is limited for you, you make the table outside scrivener and insert it only after you compile (at some point you got to compile to submit your work, no academic publisher will take the .scriv format)

Sente as reference manager, again scanning only just before submission

Critic Markup for comments and changes on the manuscript between collaborators

The real bottleneck is that everyone else uses Word, not Scrivener itself

We have many academic users, so I don’t think it’s fair to say that Scrivener is not usable for academic work - in a recent survey of 10,000 users, 22% of them said that they used Scrivener for academic papers.

You can already insert rows and columns into table - just Ctrl-click into the row and use the contextual menu, or use the Format > Table menu. Likewise it is possible to split and merge cells using the tables utility panel. You can also copy and paste tables, just like any other text. I do agree that tables are a little weak, but this is sadly down to Apple’s implementation of them in the OS X text system, which is what Scrivener for Mac uses (the same one that TextEdit uses). With us being such a small team (one coder on the Mac version - me - and two on the Windows version) it’s unfortunately not possible to provide our own custom-built text engine.

We have lots of users using Scrivener with Sente and Endnote, so I’m not sure of the problem there. The idea is that you compile to RTF and then process the citations with these apps for the final draft. We have always been very open about the fact that Scrivener is most powerful for writing the first draft, and that many users (though not all) may need to take their work to another app once that is done.

As for collaboration, we support comments and snapshots, but there are no plans to allow different users to work on the same project at the same time, as we do not have the resources of Google - please see the various threads in the Wish List forum about this.

All the best,

1 Like

How many Scrivener users are there, and what is the percentage split between OS X and Windows users?

Or is that hush-hush info?

More than 10,000, less than 1,000,000. :slight_smile: Most of our users are still Mac users, but that’s mainly down to the Mac App Store. Windows users buying from Amazon outnumber Mac users buying from Amazon; the split is fairly even on our own store. But the extra MAS users push the numbers over the edge in favour of Mac users (even though we sell to more Mac users through our own store than via the MAS). David knows all of the stats better than I do.

Appreciate the candour, even with a w-i-d-e margin. :smiley:

Read about a Scrivener trainer making $40,000 a month. Wondered how the user numbers stacked up to make that feasible. Got to be far closer to a 1,000,000 users then. J.O.Y.


I can probably guess which trainer you mean. I have no idea how he does it (other than by making his course very expensive and persuading people that Scrivener is just so hard to use that they absolutely must buy his course), but if he really is earning $40,000 a month, then he’s earning a lot more than anybody at L&L.

From this:

Some people are keen to get discount coupons to even buy Scrivener. Other people seem to have a lot of money to spend.

(I only saw the article because someone sent me a message with a link saying that they found the Scrivener tips [and personal email answers] I publish on my website more useful than….)

One possible way of dealing with point 2 in the OP is to reimport your final text after going through the revisions process in a word processor.

I wrote my phd thesis in scrivener up until the point of a first draft of each chapter. Everything from then was done in Word, but at the end of it all, I reimported the final thing and chopped it up into sections. I now use scrivener’s search features to dig through that imported version when I need to refer to something while trying to squeeze some articles out of it.

Using Word in conjunction with Endnote, there is a ‘convert to unformatted citations’ option which allows the endnote codes to be turned back into plain text, which is a crucial step prior to reimporting to scrivener if you want to keep the endnote references.


The key to using Endnote with Scrivener is to use what Endnote calls “temporary citations”. In this way, you can use Scrivener and Endnote side-by-side as you work. (Because you do want to get those cites in there while you are thinking about it!)

In my opinion, ‘temporary citations’ are the preferred way to use Endnote anyway, because it is transparent and not code-dependent (and there is absolutely no reason you need to see a WYSIWYG citation when you are working on your academic paper, so it makes perfect sense to let all the citation formatting and biblio processing be done by Endnote at the end).


My academic involvement is now much reduced, but until this year I was still actively involved. I think Scrivener is far better suited academic writing than Word (or anything else that I’ve tried). I also wrote my doctoral thesis in Scrivener (dozens of drafts!) and I struggle to see how Word can compare with Scrivener. I honestly(!!) do not think I could have written my doctoral thesis in Word.

Tables: Word handles tables really well and this is the only area where I consider Word to be superior to Scrivener. As Keith noted above, Scrivener is limited by the text capabilities of the underlying OS. I don’t write in Word (haven’t for years), but still use it for tables.

EndNote: I had no problem with using EndNote with Scrivener. None.

Collaboration: I compiled drafts to Word documents (Word not needed to do this) and then imported the edited versions received back from collaborators/editors. This meant that I could have both the original text and the edited version(s) side by side as I edited and revised. Incredibly helpful. In the few occasions where I couldn’t visual display changes (e.g. deletions) and it was important that I could, I would print to PDF a copy with all changes visible and import the PDF into my Scrivener file.

FWIW, I write exclusively in Scrivener, on all things Academic.

Recently purchased the Windows version, have yet to install it on the work pc.

As mentioned by others, the {temp citation} feature works a treat. I use SonnySoftware’s Bookends12, and there’s really not much difficulty in invoking Cmd+Y, selecting the relevant reference, and invoking another shortcut key to paste it back into Scrivener.

If that’s happening regularly, I can use Alfred’s clipboard history to quickly search and paste my most recent insertions as well, OR I can even type out the reference manually [Bookends does a fine job at working out what reference I am referring to – so very often {Surname, Year, @101-102} will be enough to have the correct reference identified – and if it cannot work it out, it asks me. [Without muddying the waters, I have constructed workflows that sees the currently selected text (i.e. Author’s name) fed into the Bookends search bar automatically, so a simple tap on my iPad, sees the search conducted over in BE, making the selection of the relevant reference even easier that it already is!]

Once all is said and done – I compile over into Word. Run the Bookends scan, to convert the {temp citations} into final references, and have it automagically generate the bibliography [hallelujah!].

Touch up on the formatting if I must. Done.

I’m not involved in the sciences, so have no need for tables/equations and the like – but works a treat for the social sciences/arts, at least in my experience.

So I guess you could consider this another endorsement from someone who uses Scrivener exclusively for creating academic papers – with Word remaining involved in the very final step, immediately prior to sharing/publishing said papers.

Possibly, possibly not - but what a point! (As long as that’s a metaphorical first draft that includes all the drafts up till the final one.) Let’s hear it for drafting, re-writing, re-fashioning, revising, re-doing, call it what you will, and for Scrivener’s pre-eminence as a re-drafting tool.

I used to work in an organisation that made a shibboleth of re-drafting. Perhaps, as is sometimes the way with shibboleths, it became outmoded. But before it did, it had great value - for clarity, concision, the improvement of communication, above all, for the benefit of the understanding of the consumer of the communication. I remember that the record number of re-drafts in that organisation was something like 54 - yes, Michael, if you’re reading this, I’m thinking of you - and of course you can have too many - there’s a sort of bell-curve of improvement when after a good few revisions the quality of the improvement drops away, and the editorial skill then lies in judging just where the curve is turning down. But on reading these author quotations about rewriting, I see that our total of 54 was easily exceeded by Roald Dahl’s. (This too from the quotations is one I’ve always particularly liked:
Ernest Hemingway: I rewrote the ending of A Farewell to Arms, the last page of it, 39 times before I was satisfied.
Interviewer: Was there some technical problem there? What was it that had stumped you?
Hemingway: Getting the words right.)

Of course, academic writing is different - but not that different. Complexity surely demands clarity, and clarity usually requires getting the words right.

Couldn’t agree more, Hugh.

I sense a kind of kind of disconnect between those of us who’ve been using Scrivener for years — particularly those like you and me who’ve been with it since Mac version 1.0, who took it on board at that time as a “drafting” tool, and what already looked like becoming the best drafting tool ever — and those who’ve come to it over the last couple of years, who hear about this great “writing app” and plunge in expecting it to do everything and produce the perfect output file.

We old lags are perfectly happy with the thought of compiling a draft in RTF and then importing to Word, NWP, Mellel, Pages or whatever word-processor or into Final Draft or other scriptwriting program we favour to finalise the formatting, or to use another program to create the complex table we need to insert. But more recent users, who haven’t been involved in all the discussions on the forums over the years and who come straight into using Scrivener in its present form, seem to baulk at the thought of needing to use another program in the downstream phases of their work or for producing specific content — tables, equations whatever — to import.

I think also, the onward “simplification” of using computers generally and the encroachment of the style of tablet and phone apps with their thousands of templates and all their presets which encourage you to look at the output, without paying attention to the nuts and bolts of what it is that the preset or template is doing and how it is made up, all this encourages new users to expect something similar of Scrivener. And they then seem to be distraught when for some reason the output is not exactly how they want it, or the template is not what they expect.

And I throw my hands up in horror in these and other forums when I see a post claiming that it is too much to be expected to use two mouse clicks to achieve something, and asking for a one-click solution to be programmed in!

Thus, for the OP, Scrivener is not suitable for academic work because it is only a drafting tool; for me, it is superb for academic work, precisely because of its power as a drafting tool.

Disgruntled of Exeter
a.k.a. Mr X

Another academic who has to say, everything I write is drafted and polished in Scrivener. I use Bookends with temporary citations, MMD ⇒ LibreOffice (and TeX equations are converted to images via TexMath[1]). The ODT is converted to docx for collaborators with track changes on. Any changes are added back to Scrivener, using footnotes for comments back to collaborators. Rinse, spin, repeat. My only problem is one collaborator who refuses to stop using endnote CWYW on his manuscripts, so I just use Word but it is not really my draft so no problems really[2].

MMD tables are good enough for my needs, but my field (Neuroscience) rarely uses tables for much anyway.

Scrivener is an amazing academic’s best friend IMO, the epic benefits of managing research and organising ideas far far outweighs the minor steps of compiling…

[1]: — note that conversions of equations has to happen before Bookends scan as it will choke on the TeX {} otherwise…
[2] I have occasionally created a Scrivener project and linked his docx as research, so I can benefit from collecting PDFs and notes while still having to edit in word…

Academic user here also. I rarely co-author but often send material to colleagues/friends for them to read and comment on. Usually this is in the form of a Word doc, which I open in front of me on one screen when returned, open Scriv on another screen and implement any suggestions I want to take on board. Maybe that’s a time consuming approach, but I find that as I incorporate these suggestions I am also revising and re-editing myself, so it works. The returned comments are filed away in Devonthink.

As for citations, I experimented with citation managers but found them to a pain in the proverbials. When I write, I manually insert citations :open_mouth: . A quaint approach, but my guess is that citation managers save people less time than they imagine.

In sum, I find Scrivener excellent for academic work.

Not me. Saves me more time than I can imagine. And I hang out on these forums, I have a good imagination… :slight_smile:

On this point we are in 100% agreement. :smiley:

Scrivener does convert DOCX comments to inline annotations, so one can actually see these if you import the collaborators version into your research. I tend to print to PDF (word keeps all the track changes and comments visible) and import that instead, but I do like working in split-editor mode in Scrivener…

As for not using a citation manager, wow! Perhaps we just get lots of papers rejected (aim for Nature, end up in Journal of Latvian Neurology!!! :wink: ), but just having to reformat when resubmitting in itself would take ages. Being able to search, read and manage refs in one place is cream on top. But to each their own! :smiley: