Referencing, Bibliographies, and the like

I will first admit that I am not fully versed in all the different ways to appropriately cite and reference a paper. Having said that I DO list and cite all references in the documents I write.

With word my approach has been to create a text bibliography at the end of my document and use footnotes and cross references to link each item to the bibliography. I ended up doing this since the citations left by word were too short and there were issues cross referencing from footnotes.

Since Scrivener can already do footnotes and has the beginnings of citations book marking (nice and easy to drag and drop website url’s into Document References) would it make sense to add some more referencing capabilities? I have tried Bookends, but this is double work in my view. If I have already done the research dragged/dropped it into Scrivener (including url) why put it into another tool, again, to aid in citations?

I am referring to manual & drag/drop entries and not the look-up work that Bookends does.

If I could do this chore once when writing, it would save loads of time and effort.

When doing research with Scrivener it would be great to just drag a citation/reference over (whatever form MLA etc.) right into the draft. Scrivener, Word, or an external tool can then spit out the bibliography nice and tidily.

Maybe I have things mixed up or am missing something but Scrivener would seem to serve this well.

Thank you for taking a look and thinking about this one (apologies if I missed similar posts)

We have tentatively planned to develop integration with popular citation management products in a future release of Scrivener (along the lines of what a Word plug-in might offer). It won’t be for a while; it is a rather large project, but it’s there on the list.

It doesn’t really make any sense for us to reinvent whole pieces of complex software into Scrivener though. There is a lot of work that goes into the functions these programs provide, and anything else than what they provide would not be adequate to cover the needs of most our users. Not even Microsoft Word goes into that particular thicket.

While it might be relatively straightforward to code citation management for one style, to cater for the hundreds of styles writers may need to use is a major task. Which is why there are dedicated citation apps in the first place.

So to answer your question

The reason is the difference between



While the former provides an accurate reference, it is not acceptable in an academic setting or for publication. The latter two are examples of the same source represented using APA-style and Harvard style respectively. Of course, this is just for the reference list at the end, there’s also the differences in citation style to worry about (Author date? Annotated footnotes? Numbered endnotes? What aboutt multiple authors? Or secondary citations? Undated sources? Translated sources? Classics? Reprints? Audio sources? etc). By using a citation manager like Papers (or Bookends, or EndNote, or…) I don’t need to worry about all these details, I just write and let the software handle the generation of correctly formatted citations and reference list at the end of the process.
Note: all three examples used here were generated with drag & drop from Papers.

I still mainly use EndNote for managing citations in my academic writing, mostly because I have a system that works and collaborators that also use it (and I’m getting full value from my tweaked citation settings), but I use Papers as my main reference repository and am beginning to switch to it for citation management. Even though both of these apps work well with Scrivener, I’m pleased to read AmberV’s comment that further integration with citation managers is considered for the future.

I am glad to hear that additional functionality is being planned.

My suggestion will not replace the other tools that are out there and I will try a few more.

If I look at Bookends in its most basic form or the citation manager in Word, I am looking at the fields they capture. In Word you have 1 of 4 citation forms to choose from (APA, Chicago, MLA, Turabian) and based on fields you have entered the data in, will cite accordingly at the insertion point. Bookends is in this respect very similar though it has over 20 forms to choose from (or create your own it seems). Fill in the fields and drag the source to the insertion point for citation according to whatever form you have chosen.

Essentially I am suggesting the same thing. Capturing the fields in Scrivener, selecting a form (maybe limited to a few like word), and dropping them into the text.

I am not for one second suggesting the search, collaborate, and other functions that BookEnds or EndNote can do. In my field, however, many of the articles and research items I seek are not searchable or automatable. I have to manually enter the data into whatever tool I use. Why have the data in 2 separate tools?

Hence the thought that perhaps this manual data can be captured in Scrivener. The research tools would not lose their place in this simplistic view.

You are describing a level of support we probably won’t be adding. If anything additional happens here at all, it will be to integrate with existing citation managers so that the marker handling is more elegantly handled (rather than using the plain-text placeholders), and maybe some minimal database front-end searching. To get an idea of what I mean, see how Mellel implements integration with Bookends and EndNote. Something like a third References panel that lists all of the sources in the current editor session and makes it easy to look-up and insert cites from your database is what we’re thinking of here.

I must stress this is all long term speculative stuff. For the most part our academic user base is quite happy with the current system which utilises plain-text tokens, a good database on the side, and post-processing the cites and bibliography with the database software. That is the industry standard for anyone not using Word, by and large, and some even prefer it to Word’s heavy handed plug-in approach. It’s modern, efficient and very scalable.

Like you say, the idea is to have one singular repository for your sources so you aren’t typing this stuff over and over. The best way to provide that is to make utilising your central repository easier from other software, not to duplicate a small subset of the functions from that repository in every piece of software related to writing, so that there are now thousands of weak citation management features in the world, instead of a good solid handful of strong ones that thousands of programs hook into. That’s just a modern principle of good software design.

Again, nothing here is assured. These are all just ideas right now; a few lines of text in a list of maybes that is a mile long. Don’t get your hopes up, is what I mean to say.

And to reiterate, I do suggest looking into Bookends RTF scanning, and considering keeping all of your source data in Bookends, not Scrivener. The only thing that should be in Scrivener is a piece of {text like this} where the cite marker will be placed. I mean, you can work other ways of course, but this is the most efficient route to take.

I’m curious to know, which field and what type of data? Modern citation managers automate almost all data entry of academic material, and large amount of other material as well. In Papers, I only occasionally need to manually enter anything, and when I do it’s usually just to append or correct a couple of components (rather than entering everything).

Remember also that the information that needs to be recorded is different for information sources. Websites, journal articles, monographs, book chapters, e-books, paper books, Twitter posts, abstracts, conference proceedings, newspapers (both online & paper), brochures, online forums, Facebook, magazines (again, differences between online and paper), court decisions, patents, legislation, etc all have different data that needs to be recorded and subsequently listed. How much programming time would be needed to address all of those data sources? And to automate the data entry? And to create the database to store it all? And to identify all the different referencing styles, let alone prepare and maintain an accurate display of each? And to program the incorporation of all this into the text editor? And who knows how many other behind the scenes bits and pieces need to be done to support all the above? Further, while Keith was doing all that, he wouldn’t be improving Scrivener. Given that all of this has already been done, very well, by a number of independent apps, this is why there is a need for two different tools.

At the professional level, managing references is a complex, messy, business. Speaking for myself, I’d rather use Scrivener for writing and use specialist apps for citation management. If there was a more seamless workflow between them, that would be fantastic, but it’s already quite simple to use two together.

Finally, if you have the reference information entered in a citation management app, you don’t need to import it into Scrivener. The whole point of citation managers is to manage all the citation and reference details so you don’t need to. In my thesis, for example, I simply had an in-text note that looked like “{nom, 2013}” for each reference to a source and that note was subsequently converted by EndNote to a correctly formatted citation and a matching entry (again correctly formatted) added to my reference list. No double entry needed, the data is only in one tool.

I am not an academic, as a management consultant I often write market reports and sources tend to be news agency web sites, reports published by associations, among others. Most of the info I have does not register in BookEnds or EndNote and needs be entered manually.

I take the point on board, however, that it is better to use 3rd party citation tools since they are better adapted for this task and integrate them into my work flow.

Now I need to choose between BookEnds and EndNote. Both seem highly rated though BookEnds looks easier.

Thanks for the comprehensive answers and pointers.

Take a look at Papers2:

Beautifully designed, and caters for websites, reports, patents, etc. And you can search the database without launching it (and also insert citations into Scrivener very easily).


I second Martin’s recommendation: check out Papers 2. It has a free trial period and is very easy to use (much easier, in my opinion, than EndNote). Given that your sources are mainly websites and reports, it should be fine for that. Papers 2 is always my go-to app when conducting or revisiting research.

Having said that, EndNote have just launched version 7 for Windows and Mac users who purchase version 6 now can upgrade for free to version 7 when it is released. EndNote is probably more flexible than Papers, and certainly has more citation options when writing, but is also more complicated and far more expensive.

I use both, and think they are both excellent, but if I had to choose then it would be Papers. They both have free trials, so give them a go.
I haven’t tried Bookends for years, so can’t comment on it.


I have also encountered this problem, but also found a solution that works well for me. My university already has a license for Endnote and as explained in the following video Scrivener works really well with Endnote:

I also discovered recently that the Format Paper step explained in the video is not really necessary. You can just as well have Endnote open, open the RTF document in Word and click the update bibliography button on the Endnote Tab. This will also convert all the Endnote tags into proper references and add the bibliography to the end of the document.

I hope this helps,

Kind regards,


ps. For web references I use Zotero and export to Endnote, or when I have a lot of them I add them as endnotes in the Scrivener document.

That is true, if you have Word you can format everything using the Endnote plug-in (I believe this is also a possibility with Bookends, but don’t quote me on that), and the result should be identical to processing the RTF file. The route shown in the video avoids Word entirely, so some people prefer that.