Mendeley Integration

I wish Scrivener was fully integrated with bibliographic management software Mendeley!

Could you elaborate a bit on what you mean by that?

Hi Amber,

I’d like to be able to use a “Cite while you write”-type function; that is, to be able to do this: (which I can do with Word)

This kind of thing has been discussed various times before, particularly with regard to Zotero. If you conduct a search of the site for Zotero, Endnote, Bookends and Sente you will find the relevant threads.

I wouldn’t hold your breath!


We have plans to look at something like this down the road. I don’t know if Mendeley will make the cut, and I don’t even know what this will end up looking like, or if it will even materialise. All I know is we have it on the 3.x list and a few wild ideas about it.

It’s low priority because there are good citation managers out there that use the basic RTF scanning method. We’ve found that in fact most people prefer this method over the plug-ins once they learn how to use it. In the meantime, I would highly recommend looking into one of these that already works well with Scrivener. Bookends and Sente both receive high marks from Mac users. If you want to ultimately stick with Mendeley, then help lobby for RTF scanning. The more they hear from users that people want to use any word processor they please, the more inclined they will be to do something about it.

Hi Amber,

I’m one of the those that use Sente and prefer the RTF scan method. However, I would like to see better integration with Scrivener, particularly in terms of bibliography generation. I’ll try to explain:

  • One of the things most of us have to deal with is word limits in documents. Many times, bibliographies can be ridiculously large, taking up as much as 1/4 of the final word count. It is therefore virtually impossible to understand where we are in terms of word count until the the scrivener file has been compiled and scanned with a reference manager.
  • What I’d like to see is, Scrivener allow a reference manager access to our documents in the compile list. This can be done in a ‘read-only’ manner since there is little difference to the word count between a reference field code (i.e. {Auge, 2009 @10}) vs the final output (i.e. (Auge, 2009, p.10)). A separate document could be ‘read-write’ (perhaps an indexed file) to allow the reference manager to compile the full bibliography (which is what adds so much to the word count). Obviously, this would only help users in having a much better sense of the final word count and for draft compiles - but I think it would be extremely helpful.
    Perhaps this could be an intermediate step before the 3.x “wild ideas”?


I’m curious to know in what context the bibliography counts towards your word count? I’ve yet to encounter a situation in academia where this is the case. Every journal, thesis requirement, student assignment, or grant application I’ve investigated specifically excludes the bibliography (and usually also tables, figures and appendices) from the word count. The closes I’ve seen is some research ethics applications where they limit the number of references (rather than the word count of them).

Can you explain your usage?

I think they meant it was counted towards the wordcount in Scrivener when using this referencing system, not towards the word count needed for submission - hence the asking for a way to have the RTF scan notations excluded from the word count.

Actually, nom, you are correct in your thinking of my usage. With the exception of grant applications, a lot of assignments I have specifically include the bibliography in the final word count (student assignments and conference proceedings/edited books). Journals seem to go the same way (example here:
I appreciate thorough bibliographies, so I do not like this kind of submission, but it seems to be endemic in the architecture / urban planning disciplines.

Thanks for the explanation. In my area (psychology) reference lists don’t count towards word counts and psychology students at my university are specifically instructed to exclude them. Interesting to know that this is not the case in other disciplines.

Thanks for the link and the extra info.