How to Manage a Research Library with Zotero

Keeping up to date with research and managing an increasing number of journal articles and other material is a vital skill of academics and graduate students…

Here I set out my workflow developed over the lat few years using tools such as Zotero, Dropbox , my iPad and Goodreader. Together these tools, and others, see me through from article discovery, through cataloguing and annotating research papers, to synchronising my library across multiple devices, all the way to citing outputs in my own research and preparing accurate reference lists.

Article Discovery

I use a variety of methods to keep on top of research in my field, and to search for articles when preparing a paper or presentation. Foremost is the use of RSS feeds to deliver alerts of new articles in journals I follow.I use Google Reader to manage my feeds and usually browse through new feed articles using Flipboard on my iPad or iPhone. If I come across an interesting article that I think I may like to read and use, I email myself the link. This means that when I check my email every morning, I can navigate to the article and save it in my reference manager for reading and marking up. If I am researching a paper, I tend to search journal repositories such as Web of Knowledge, although more often than not I find Google Scholar finds what I need quickly and accurately. Finally I am finding that Twitter is a fantastic resource for uncovering new research as an increasing number of researchers and research groups have a presence.

Reference Management 

Once I have found a research output that interests me, I import it, and its bibliographic information, into my reference manager of choice, Zotero. I have tried nearly every reference manager over the last few years including EndnoteMendeleySente and Papers 2, but for me none of them come close to Zotero. Zotero is a freely available tool that comes as either a plugin for Firefox, or in a standalone version. It collects all my research in a single, searchable interface and can add PDF’s, images, audio and video files, snapshots of web pages, and really anything else. The software retrieves bibliographic information for a paper directly from the web browser through a plugin that presents an icon in the address bar when it senses information it can save. The plugin seems to be able to pull accurate information from nearly all of the usual journal repositories as well as from Google Scholar. Zotero can also download a PDF of the paper (this only works with some repositories) which saves directly to my library. In the cases where this does not work, I download and attach the PDF manually. Finally Zotero automatically indexes the full-text content of my library, enabling me to find exactly what I am looking for very quickly.

Synchronising my Library

I tend to work in different places a lot and so need my full library to be accessible at all times. To do this I use Dropbox to store all of my papers and my Zotero library data together. This way Zotero is always up to date whether I am working at home on my MacBook or in the office on my PC (although I have to remember to log off first else the sync gets messy). The ability to work across operating systems is another must for me. I tried to do this with Endnote whilst writing my Thesis which had disastrous effects – at one point I couldn’t open the document due to Endnote changing all the meta-data. To date I have had no problems syncing in this way and its fantastic to have my research library in the cloud to access on the go. [Update] There is a risk that using this synchronisation method the Zotero database may get corrupted if you leave Zotero open on more than one of your computers – Thanks to Adam for the warning in the comments below – however I have not had an issue with this to date. I shall leave it up to you to decide the level of risk.

Reading and Annotating

Once I have imported a newly discovered paper into Zotero, I send it to my iPad for reading and annotating. To do this I use a plugin called Zotfile. This gives me three new options in the Zotero menu when I right-click an item in the library; the ability to ‘Rename Attachments’ to any convention I specify (essential for good file management), the ability to ‘Extract Annotations’ from the attachment and the ability to ‘Send to Tablet’. This last option sends a copy of the PDF to a Dropbox folder outside of my main library that I have set up. I then use Goodreader to download the file from Dropbox, and annotate the PDF. Returning to Zotero the ‘Send to Tablet’ option changes to ‘Get from Tablet’ which retrieves the freshly annotated PDF, returning it to my main Zotero library folder. As it does this it extracts my annotations to a separate file, attached to the PDF that contains only my notes. This is great for when I am writing a paper, or preparing a lecture, as I only have to look at my notes rather than re-read the whole article.

Citation Management

I have also found Zotero to be the best citation management tool. I can use Zotero’s word plugin on either my PC or Mac to insert the citations into my text as I write and build the reference list at the end of the paper. If working in Scrivener, which I use for longer projects such as writing books, I insert the references in curly brackets {} and then use Zotero’s RTF scan to replace these with the correct citations and insert the reference list (detailed instructions on this can be found here). Another reason why I favour Zotero is that it includes the vast majority of referencing styles, which makes it easy to change style if adopting a paper for submission in a particular journal. If the reference style you need is missing, it is relatively easy to modify the an existing style and create your own.

 Over to You…

This is how I manage my research reading, cataloging, annotating and referencing, but I would be interested to know how others approach the problem. I know that Zotero is not for everyone, so if you have a system that you think would work for others, please let me know in the comments..

And Finally….10 points for the first person to point out who the two people in the title image are…and NO cheating…

32 thoughts on “How to Manage a Research Library with Zotero

  1. I think the two people are a young Franz Kafka and Max Weber in a beard. Very useful post.

  2. If I understand you correctly and you’re using dropbox to sync the Zotero library (as opposed to just the attached files) that’s a bad idea that has a very high chance of permanently corrupting your Zotero database. Don’t do it.

    • Hi Adam

      Thanks for the comment. Could you explain more about what you think the danger is? I have used this method for over a year to sync my library across my MacBook, MacBook Air, Windows PC and iPad without any issues.

      There is perhaps danger of corruption if Zotero remains open on more than one machine at a time (a danger inherent in most synchronising endeavours) however I am in the habit of only using one computer at a time and shutting down or hibernating after use. Also my Zotero library, minus PDFs, is synced with Zotero’s online server, so in the unlikely event of corruption, I can restore from there? (I realise I would need to restore links to my PDF library if I did this.

      To reiterate, I have used this method without issue nearly every day for the last year with a very large library and have yet to have an issue…



      • The risk is mostly having Zotero open on two computers at the same time.That’s not a danger inherent in most syncing – e.g. Zotero sync doesn’t care how many computers are running at the same time and for regular file syncing I’d expect dropbox to have some conflict resolution, too – it’s specifically that dropbox and related services aren’t set-up to sync database files.
        There is also a possibility, though smaller, that this may happen because Zotero doesn’t finish writing to the database, e.g. when you’re hibernating a computer but leave Firefox open. That’s why the instructions for taking a back-up tell users to close Firefox before backing up.

        I realize you’re not typically running multiple computers at the same time, but I think it’s a bad, bad idea to rely on any system that requires you to not make any mistake, ever, when it comes to crucial data – any number of scenarios come to mind: forgetting to turn a machine off, a spouse or roommate using a computer etc. One incident can be enough.

        Database corruption may not be visible immediately, so you can lose months of work. It’s your choice, obviously, but even if you think it’s unlikely – do you really want to bet your data security on it never happening? There are dozens of threads on the Zotero forum by researchers who got burnt.

        More pragmatically, as you know Zotero, like all open source software, doesn’t come with a warranty, but you get high-quality support through the forum. If you ever have a problem with Zotero and report it, the first thing you’ll be told is that they can’t help you as long as you have your database stored in Dropbox..

      • Thanks again Adam

        Your comments and worries are certainly valid. I think the lesson here is to use syntonisation methods such as the one I posted with care and at your own risk. That being said, I have not had issues and this method has greatly enhanced my productivity so for the time being at least I shall continue..

        But fair to say – I have been warned!


      • of course. If you could include a pointer to these comments in the blogpost so that your readers are aware of these risks, too, that’d be appreciated.

  3. Hi Alex! I came across your blog post via twitter. I was wondering why you chose Zotero over Sente and Endnote? I’m struggling to find a system that works for me. I’m not a fan of Mendeley and I am loving Scrivener but the thought of exporting to word long drafts for my supervisors makes me shudder (as I start the final year of my PhD). I’m a mac user and don’t get a lot of support from my uni with regards to software.

    • Hi Eijee, thanks for the comment..

      There are a number of reasons I chose Zotoer over Sente and Endnote.

      With regard to Endnote – I found it clunky, slow and un-intuitive in use. Quite often it would not find the correct bibliographical information for a source, plus I found syncing and attaching PDF versions of paper difficult. Zotero is simply better at drawing bibliographic information from a multitude of sources via my web browser, and allows me to sync my library with my iPad via Dopbox.

      I liked the look of Sente and found the interface excellent…however like you, my Uni discriminates against Mac users so I am stuck with using a PC in the office. This ruled Sente out for me as I need something that works across both OSX and Windows. When/If my institution allows me to use my MacBook on the work network I will probably take another look at Sente.

      Finally Zotero is FREE…Sente is $129.95/£83 ($89.95/£58 with academic discount) which perhaps is not unreasonable – but out of reach for many graduate students..
      Endnote is $199/£127 and I would need both Mac and PC versions…OK, I can use the software via my institutions virtual desktop – but I dont like using this as it is slow and clunky and limits me to using the programs installed on the virtual environment (Internet Explorer…Shudder…).

      With regard to using scrivener – you dont necessarily have to export to word…you ‘compile’ as .rtf – scan with zotero, and print with whatever WP software you need…Pages, Word, Google Docs..etc (Scrivener is not a word processor – it is designed to manage text rather that formatting – you will usually have to export to some kind of WP software to ensure your document conforms with all the complex formatting requirements of a Thesis submission, book, paper etc).

      Having said all this – you should choose what works for you!! This is MY workflow – it is not intended to be a ‘you must do this guide’. If my tips work for you great! if not develop your own approach….then blog it for others..🙂


  4. Zotero seems really good, but I still choose papers 2, because I don’t need to change between so many apps. Importing a paper from your browser (with ‘Open in papers’ plugin), synchronizing, reading and annotating, exporting the notes, organizing the library, emailing pdfs, and citing are all done through one app. Also this way I don’t use my Dropbox’s space for pdfs. As for the cross-platform, they recently released a version for windows. I only wish they had one for Linux as well.
    So what am I missing and you go for Zotero?

    • Hi Firefly

      I really wanted to like papers 2. It has perhaps the best looking interface and yes, it can handle synchronising and annotating, and there is an iPad app. Unfortunately It just keeps crashing for me and is incredibly slow. I have tried it on two separate occasions on two different MacBooks but I just cannot get it to import my (rather large) PDF library without freezing and loosing data. This, coupled with the price, $79 (59 with academic discount), and the fact that I would need to purchase both Mac and PC versions, was a deal breaker for me.

      Finally I like the fact that Zotero is open source, and has a very active community on their forum which is great if you have any problems or feature requests. I also like the fact I can adapt the reference styles easily to match any journal requirements – makes life so much easier when revising publications.

      But as always, use whatever tool works best for you!


      • FYI – While I agree about the virtues of open source, Papers2 uses the same citation styles as Zotero (and Mendeley) – which they can do because it’s an open source project and all the styles are CC licensed.
        To their great credit, the people at Papers have actually been very good about contributing new styles and fixing old ones since they adopted CSL. I believe the even give you a free license for contributing a style.

      • Hi Adam

        I didn’t know that, thanks…it’s useful info – especially the potential for a free license…

  5. Pingback: Organising myself « Between the Lines

  6. Pingback: Three Apps For Building Your Research Collection | TeachThought

  7. Pingback: Graduate Students Need Apps To Build Their Research Collection, So Try These!

  8. Thanks for this post on your workflow. I have been thinking a lot about this and blogging about it a bit too. Here is my most recent post, which indicates again the many sharp tools I use to practice my scholarship in a digital age. An update to this is coming though, as I integrate ZotPad into the cycle. That app has brought me back to Zotero in a more robust way.

    • Thanks for the comment Christopher, and the link to your blog post. I had been flirting with the idea of moving to Sente for a bit (not a fan of Mendeley) for the ipad app sync etc. However I think that I may actually stay with Zotero after reading your post🙂



  9. Alex, thank you for the post.
    I am trying to set all at once to start using Zotero on all my computers togather with GoodReader on iPad, and I have one question and one advise:
    The question. Zotero does duplicate each and every PDF in it’s own ‘storage’ directore, and in a terrible way. Is there any way to avoid this duplication and make it working with my own files in my own directories, (like photographic databasing software do, Lightroom, CaptureOne, ACDSee or whatever)?
    Why do I want to avoid duplications: 1) what the reason to loose disk space? 2) it will be possible to syncronize ALL papers with GoodReader (yes, without database), which is very important for me.
    The advise. To avoid danger of database corruption during synchronization of Zotero database on two computers, you’d better do it in the following way:
    1) Zotero syncronizes database ONLY using it’s own servers – it will be done with no danger, but you can not open on Computer2 files added to Zotero on Computer1;
    2) ‘storage’ directory is syncronized through all computers with SygarSync/Dropbox/whatever – so you CAN now open files on any of your compuetrs, and it will not hurt your database until you try to edit the same file simultaneously on both computers.

    • Thanks for the comment Vladimir.

      I use the Zotfile plugin in Zotero which allows me to specify a specific folder for Zotero to store my PDF files. Each time Zotero automatically imports a file, Zotfile moves it from the default Zotero directory to my preferred folder avoiding duplicates.


      • The Zotfile/Dropbox combination is great, but does have the downside that it only works when you can place the dropbox folder at the exact same filepath on all machines. That means that syncing with one Windows machine and a Mac/Linux computer won’t work (since currently file-links in the Zotero database are absolute links).

        (I’m a little confused about the “duplication” – Zotero saves PDFs in its own storage folder, but not anywhere else. I don’t see how that duplicates files unless you keep a separate folder/file structure with all of the files – which kind of takes away the benefit of having files attached in your file manager anyway. For people who like their PDFs in a human-readable folder, I suggest looking at virtual folders, available on all operating systems, which provides that w/o any file duplication)

  10. Interesting to hear about your system and use of Zotero. I have a similar system:

    1) I get literature all sorts of places; From colleagues, various online databases etc.

    2) I maintain a BibTeX-database using JabRef. It does not have the nice metadata-extraction that e.g. Zotero and Mendeley have (in fact it has it, but it, on average, only retrieves the author and title). In turn, it integrates seamlessly with LaTeX (that I use for writing documents and articles). In JabRef, each reference is linked to (an annotated version of) a PDF (that opens in an external PDF viewer; Mendeley opens in an internal viewer, and I personally prefer that former).

    3) All literature is stored in Dropbox. I have two folders: Original and Edited. In the former, I keep copies of all literature that are not annotated, and in the latter I have the annotated versions. That way I always have a “clean” copy, e.g. for forwarding to colleagues.

    4) The literature is read and annotated on my iPad in Goodreader.

    5) I access my BibTeX-database (that is in my Dropbox) on the iPad in PocketBib: It syncs with the BibTeX-database via Dropbox, and it has an internal PDF-viewer through which it can open the PDFs (that are in the literature folder in my Dropbox).

    I have considered moving to Mendeley or Zotero – because of their functional metadata-extraction – but am afraid I will loose a little bit on the integration with LaTeX. What do you guys think about my system vs. going to the more dedicated database systems like Mendeley and Zotero?


    • Thanks for the comment Jakob.

      Your system sounds vary interesting too. I don’t use LaTeX – it is probably a little bit too much for my purposes. The real issue I think that this highlights for all of us is that we are increasingly using a workflow that includes reading and annotating papers on the iPad. It seems a little crazy that we still have to jump though all these hoops to get this basic functionality to work. I guess we could de this with Papers or Sente who offer integrated iPad apps, but these are not without their problems (or high price tags). Hopefully we will see a good Zotero iPad app soon, or a Mendeley one with better functionality..


      • for that specific purpose, ZotPad seems perfectly sufficient already, even without the planned improvements, any reason you’re not using that?

        As for LaTeX – the standard workflow I’m aware of is to work, organize & annotate in Zotero, then export BibTeX and either use that for writing directly, or to import into JabRef and use that as a frontend for writing. Zotero doesn’t let you change citation keys atm, so if that’s important to you you’d do the latter. I know of a lot of people who use Zotero (or Mendeley) with TeX, I’d warmly recommend it.

  11. Pingback: Reflections on Academic Blogging « Dr Sustainable

  12. Reblogged this on As the Adjunctiverse Turns and commented:
    Is anyone else tracking, collecting and trying to organize online resources about adjunct and academic labor? We need all the help and advice we can get on how not to drown in the information. The organizing principles laid out here should help even if you aren’t collecting for formal research purposes or don’t use the same tools described in the article.

    PS this link is from today’s #etmooc Daily on but seemed more appropriate for here than MOOC Madness, mostly for the serendipity of coming across it while multitasking email, rss feeds and exchanging correspondence about developing an adjunct/contingent faculty bibliograpy ~ a worthy project, especially if coordinating multiple existing efforts.

  13. I agree about not using DropBox. Use the built-in syncing support in Zotero.

    I recommend that people get a free WebDav account so that all your pdfs can be synced, not just the bibliographic information. I have 3gb for free from

    Be warned: I tried a bunch of different WebDav providers and not all are compatible with a large Zotero library — some have a limit of 1000 files per folder which you will eventually hit😉.

    Second warning: There can be firewall issuses/ For example, Zotero WebDav syncing does not work within the hospital where I am located — I have to vpn to the university to successfully sync.

  14. Pingback: The Cloud Academic: Part 1 | Dr Sustainable

  15. Pingback: Citing Sources and Zotero « OSU-Cascades Library

  16. Pingback: Reflections on Academic Blogging | Dr Sustainable

  17. Pingback: How to Manage a Research Library with Zotero | Dr Sustainable

Leave a Reply

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 )

Google+ photo

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

Connecting to %s