Setting up workflow with DTPro, Bookends, & Scrivener

I’ve had DTPro for years but rarely used it.
I use Bookends constantly, and now store my PDFs in its iCloud folder.
I also store research-related PDFs etc within Scriviner project files.
I’d appreciate help in understanding how best to set up my work-flow with these apps. I’ve no experience with scripts, etc…

I just want to add that I have tried two things:

  1. Indexing the Bookends folder in iCloud, where I store my PDFs associated with Bookends entries
  2. Importing references from Bookends via the File>Import menu.

Both produced good results, but I’m not sure yet if they are redundant.

More importantly, I’m not sure how DTPro will update as I add new references and PDFs to Bookends. I see the option to “Update Indexed Items” – which I assume will deal with the documents folder now in my DTPro Inbox based on the Bookends folder in iCloud. I do NOT see any obvious way to trigger DTPro to refresh the Bookends now in its Inbox, however.

Thank any and all for you for suggestions regarding how best to integrate these applications for academic research & writing.

The command File > Import > Referenes from Bookends skips already imported references or updates them if necessary.

Thanks, cgrunenberg, that indeed works to update the references, but what about the associated PDFs?

All I can think of adding at this point, is “what do you hope to achieve with DTPO”?

That might give clarity to some possible suggestions.

I too have Bookends and Scrivener, along with DTPO.

Scrivener is treated solely as the place I write (I sometimes drop PDFs in there, but that is on the odd occasion when I am only working on my laptop, and don’t have the external monitors available).
Bookends is used solely to manage the actual referencing inside my output-documents (along with automating the bibliography generation etc.) – I don’t use it to annotate or search, since I am far more comfortable using DTPO for that purpose.

So in my case, DTPO is the central hub of my ‘research-activities’. But that does not appear to be the case for you? Which has me wondering what you are hoping to use DTPO for?
The answer to that question, will most likely guide what possible approaches could work for you…

Thanks so much. I’m really just looking to improve my work flow.
Bookends is great, and it seemed terrific to link PDFs with the references, but I’ve been frustrated that searching within Bookends doesn’t search the linked PDFs. In fact, if I’m not mistaken, at best I get a result when a word is in the reference, but if I have long notes in the notes field, Bookends won’t take me right to the word I’m looking for.
Scriviner has also been good, and I thought it convenient to dump everything project related into its research area: PDFs, JPGs, notes, websites, whatever. Doing that has its problems too, as I don’t always remember that something is in there, and I also end up with duplicate PDFs with different annotations, which is a pain.
I bought DTPro years ago and never really figured out where it was supposed to go, but was inspired this week to revisit the app and see whether it might not help me overcome some of the above.
I am also ideally looking for a solution that will work well with my mobile setup, i.e., my iPad Pro. Having the PDFs in iCloud is pretty handy. I don’t sync them within Bookend’s iPad app, though. I use PDF Expert to sync that folder.

Bookends is no doubt an incredibly powerful tool, with plenty of add-on functionality in regards to PDFs and annotations that have been added over the recent years.

But as mentioned, I was already all-in with DTPO before setting up BE, so I simply continued along the way I had been doing things. As it stands, whereas both DTPO and BEs annotation tools are sufficient, my particular approach to annotation has only been properly catered for by Goodreader, and as such – I do all of my annotations there.

I do sync to BEoT on my 12.9" iPP, but since I essentially have a duplicate of all my academic PDFs (a ‘clean/original’ folder that Bookends sees; and then importing the same PDFs into DTPO, through a service-action on that folder), the synced PDFs in BEoT are the ‘clean’ versions of the PDFs. I then also sync to BEoT on my iPhone – but this is without the attachments.

My thinking on this point is that I always have my iPhone with me - so if I am in a library somewhere doing research, and I up and leave my desk to walk into the book shelves, should I come across a book that looks interesting, I have my BE library on my phone, and can quickly check to see if I have that source already. In this instance, I don’t need to ‘see’ the actual PDF attachment, I just need to verify if it is something I have already found.

That being said, 2 points in this regard:
(a.) I now have a 256gb iPhone 8+, and still have plenty of space left. I may as well take the plunge and sync all the PDFs across to BEoT on my phone as well, since it would fit comfortably.
(b.) I have done a full sync of most of my DBs from DTPO to DTTG2 on my phone as well – for the same reason: I have the space – so between the two of these apps, I will in any event have everything on my phone, references and PDFs.

Regarding why I sync my BE library to BEoT on my iPP >> given that I am quite comfortable writing in Scrivener iOS, I can insert citations into my writing using BEoT, and since I then need to be able to check the particular page number of the source I am citing, it makes sense to have the full attachment available in BEoT.

But all of the above, is speaking to my approach on iOS primarily.

Over on the Mac, as mentioned - I live inside DTPO. I annotate there – albeit by means of roundtripping the PDFs via DTTG2 into Goodreader, and back again – and I set-up my linked annotation notes etc. inside DTPO. All my RSS feeds in regards to Academic Journal publications are dumped into DTPO, as well as any news articles/websites relating to the topic. Any emails I receive regarding my research, is also fed into DTPO.

In essence therefore, I treat DTPO as the canonical repository for everything academic related, and only ensure I duplicate into BE/BEoT for being able to use the latter over on iOS, or as my reference manager on macOS. That being said, I do have the odd application that I use on the side as well – such as LiquidText, and very occasionally MarginNote – but those are for very specific use-cases.

Regardless, DTPO does the heavy-lifting when it comes to searching within my Academic library, and its ability to allow me to completely rearrange my library on the basis of what I am doing at the time, using Replicants/Duplicants/SmartFolders/Workspaces etc., simply cannot be reproduced in any of the other apps mentioned above.

So I guess to get back to what you asked initially, about Workflow examples, I use Scrivener to write/create, BE to manage my citations/automate bibliographys and citation styles, and DTPO to do everything else (sorting/retrieving/searching/annotating/connecting/storing etc.)… That’s really all I can think to add from my perspective - yours will no doubt develop quite differently, but hopefully this will give you a start!


I’m not sure if you are an academic or not. I am, and this is my current (perhaps overly complex) workflow, which uses bookends, DTPO, and scrivener. Perhaps of use for thinking about your own needs vis-a-vis your workflow?

1 Like

There are loads of different ways of connecting the three pieces of software. Personally I store everything in DevonThink. For reference material I will index a Finder folder, which I will also use as the Bookends reference folder. I will also create URL links between Bookends and DevonThink enabling me to quickly jump backwards and forwards between the two. I use Bookends primarily for citation and will annotate PDFs in either PDF Expert (iOS) or Highlights (Mac). Scrivener is simply where I write (in Pandoc). I don’t store any reference material in it, rather I will create a link back to reference material stored in DevonThink. I have a number of other tips relating to Bookends and DevonThink on my website:


Some good stuff in there kseggleton - thanks for sharing.

Hi Cassady,

My apologies for resurrecting this old thread. I’m considering integrating bookends into my research and workflow, but I’m not entirely sure how to start. I am familiar with several different strategies for integrating Devonthink and Bookends from this point forward (I like your process, for example). However, I’m unsure of how to import my existing pdf libraries from DTPO into Bookends, without disrupting their filing structure in DTPO. Am I missing something obvious here?


I use Bookends, Scriviner, and DT3. I too keep my PDFs in Bookends and store them all in iCloud Drive. As you know, the advantage to that is that if I annotate a PDF on one device, the annotation is on all devices (they share the same PDF).

I index that Bookends folder in DT3 (I don’t import the PDFs, just index the folder). What I love about this is that I can do a search in DT3 across all my PDFs. I do this a lot in my research. Newly added PDFs added through Bookends are automatically indexed the next time I open the that database in DT3.

1 Like

Trouble is , if there is no pdf in iCloud BE library , it won’t appear in the DTO index , and easy to miss

  • any solution found ?

I would also like to know if there’s a solution to this…!

I’m an academic and also own these three tools. I’ve just not ‘gotten going’ with them into a workflow, although I believe they can offer me a lot for a powerful workflow. I read various posts, but it often seems that examples shared throw in yet one more tool, or talk about scripts or such, which just feels overwhelming to me.

  • When I’ve tried working with DTPro I like what I see but feel like I’m probably missing a lot. I
  • Bookends looks sensible enough and posts here suggest it is the best reference manager to integrate w DTPro. I currently have all my tagging of relevant articles in Mendeley and am loath to lose all that data.
  • I wrote my last article in Scrivener and liked it a lot, but end up cutting and pasting all the sections back and forth between a Google doc to get input from collaborators.

I am wondering if anyone who is a part of this community is expert at using these three tools together AND consults with others to create customized tweaks. I’m happy to compensate for this expertise. It would just be so much more efficient it seems to hear about some options, and to ask questions and then devise my set up with access to someone who knows the settings, etc.

I can explain more if anyone is interested in such a gig. I guess part of the issue is that I like learning through interaction and have a limit for technical prowess (no programming/ scripting know how).

@sdexter – being new to DT myself, I don’t have anything to say about most of this, except on one point:

It seems like one point of friction here for an integrated workflow is that AFAIK Mendeley doesn’t offer much possibility of integrating with DT (or anything else). I have no experience with BookEnds, but I recently switched from Mendely to Zotero (b/c I didn’t like what seemed to me the increasing lock-in that Mendeley is pushing). I believe I was able to preserve all my tags.

And then, it appears that Zotero can be made to integrate with DT. Again, I have no experience with this, but here’s a thread that seems to suggest that Zotero+DTmay work about as well as BookEnds+DT:

If you are interested in looking more into Mendeley → Zotero, I can probably dig up some links and maybe some of my notes on the transition, if that would help. BTW I only ever annotated PDFs outside Mendely (and now Zotero) – if you’ve been annotating inside Mendely, I’m not sure how the Mendeley → Zotero migration wold go.

I have three Macs syncing DEVONthink through Dropbox. I have Bookends store the PDF attachments in the iCloud Folder for iOS Sync (/Users/user/Library/Mobile Documents/iCloud~com~sonnysoftware~bot/Documents). I then have DEVONthink index that folder. The problem is that, whenever I add a new record with an attachment to Bookends, the file is listed twice in the Bookends indexed folder in the other two computers when I open DEVONthink in them. Is this being caused by using the iCloud folder to store the Bookends attachments? Should I store the Bookends attachments in a directory in my drive, and then index THAT folder? Thanks for any insights about this.

I haven’t had much success using iCloud sync with bookends. I have never used iCloud sync for anything else, but have used Dropbox for years. The errors I got with iCloud and bookends sync (newly added bookends items just not showing up on other devices) were ones I have never had with dropbox. So I switched to putting the bookends PDFs in dropbox. Then I indexed dropbox folder in DEVONthink. That has been working for me. I think I’ve seen others on these forums opine that iCloud sync is not particularly robust. HTH!

Thank you for your response. I am wondering if others are indexing the Bookends iCloud attachment folder in DEVONthink without incident.

  • Do you happen to be using a VPN?
    • If so, which one?