Workflow for Tagging Dialogue

Is there a workflow in Scrivener for tagging individual characters’ dialogue as I write my manuscript? I’d like to compile a list of all of a given character’s dialogue (as part of the later editing process) in order to check for consistency. (Note: I don’t consistently use quote attributions that make this easily done via a keyword search, e.g. ending each quote with “Phil said” or “Phil exclaimed”.)

Actually the official feature known as “keywords” might work. How granular are you going to want to be? Per-document?

If it’s all the way down to each line attributed to a character, you could either use inspector comments, or I’d suggest splitting at each paragraph and using Keywords to tag which character is speaking/acting. A Keyword search on that character’s name would leave you with a collection of just what they said/did.

It’s not possible to use the compile process in this way. You can use keywords and other measures to flag which documents within your binder that individual characters mught appear in, speak in, etc… but the compile process will always take the whole document you have flagged in this way. It’s not able to pull out specific content / sentences from within a document in the binder and just comile that.

So unless you literally have each sentence within your manuscript as an individual document in your binder (which would get unwieldy incredibly quickly!), you can’t use compile.

You can, however, use the powerful search functionality within Scrivener in order to achieve this without needing to leave to go to another program. In particular, I’m talking about “Find by Formatting”.

Go through your document and make the dialogue for each character a specific color. You can then used the Find by Formatting function (Ctrl + F3 on the Windows version, not sure on the Mac) to cycle through the dialogue elements in turn. This will at least enable you to optically view the dialogue and check for consistency of tone, language, voice etc.

If you’re doing this, you’ll probably want to double check that the “Remove Text Color” checkbox is ticked in the Transformations tab of the Compile dialogue to make sure you don’t start sending out multi-colored manuscripts.

Now for the drawbacks:

  • This is going to take some time to implement, and still isn’t that clean an approach
  • Coloring dialogue as you go might impede your flow a bit
  • Spending most of your time reading dialogue in different colors might psychologically make you think the characters voices are more distinct than they really are.

In other words… there are ways to achieve what you are after (sort of), but unless you know you have a major problem with a particular character that you need to rectify, you might find it better to trust your ability to spot it on the page in a normal read though of your novel… the same way that a reader or an editor would.

Thanks. This sounds like a workable solution. I suspect the way around the issue of perceiving a character’s dialogue as suddenly more “colourful” :slight_smile: is to just duplicate each document - before changing the font style. The coloured one would only be for that one particular editing pass where I review dialogue consistency (and then I could nuke it). I’ll probably also try adding keywords to each doc identifying the characters (it’s my first novel with Scrivener so may as well try a bunch of techniques and then see which ones really work the best for me


Check this post out for some pointers on using keywords to track characters as well!

This is pretty easy if you own MS Excel. Here is what I do:

  1. Do not use smart quotes.
  2. After each initial quote mark in a paragraph, insert in-line annotation for the character (I use initials). For example:
    “[TS]This the way the world ends,” sez Elliot. “But who gives a Krapp’s last tape about that?” where [] is the inline annotation.

This doesn’t slow down my writing workflow at all

To check consistency, I compile as a .txt file leaving the in-line annotation in place. Then:

  1. Open Excel
  2. Open the txt file with Excel (the order matters). You are presented with an Import Wizard.
    On screen 1, select Delimited
    On screen 2, set the Other delimiter to ". For text qualifier select none
    Then hit finish.

Things enclosed in quotes will appear in columns B,D,F etc. Each of Elliot’s paragraphs will appear starting in col B with [TS]. In the above example, in row 1, I will have:
Col A blank
Col B [TS]This the way the world ends,
Col C sez Elliot.
Col D But who gives a krapp’s last tape about that?

  1. Delete cols A,C,E, etc - i.e., the non speaking columns (BE CAREFUL - I select A,C,E etc and delete all at once. If you do the deletions column by column, things shift so you must pay attention).

  2. Insert a NEW col A and do Edit>fill>series with Step Value 1.
    This allows you to recapture the original sequence of dialog after you’ve sorted things.

  3. Select Cols A-F (or whatever)

  4. Data>Sort on col B.

  5. Select Cols A-F (or whatever) and only the rows associated with ONE speaker.

  6. Data>Sort on Col A.

Repeat steps 7 & 8 for each speaker.

  1. Save as tab-delimited text and open in your favorite text editor.

I can do this in less time than I spent writing this up!!

LDT, I’m trying to do the same thing. The “find” is an ok workaround but without a “find all” it doesn’t allow you to see the dialogue in one spot. you have to dance from location to location. I wish there were a way to collect the dialogue in one location like a collection or something. I’ve used the Exel workaround, but I still need to go back and correct my draft. It would be nice (with the power of Scrivener) to do this “relatively” on the fly.