Googling Writing: Google Docs, Draftback, and Researching and Responding to Students’ Writing Process

Contributor: Roger Powell
Affiliation: Graceland University
Email: rlpowell at graceland.edu
Released: 7 December 2021
Updated: 2 February 2022
Published: Issue 26.2 (Spring 2022)


Throughout the 1970s and 80s, studying student writing processes was a central concern of composition scholars (Emig, 1977; Perl, 1979; Sommers, 1980). Donald Murray's (1972) now infamous saying teach “writing as a process, not a product” (p. 11) is deeply embedded into our pedagogies. While these often cited studies formed the central concept of process pedagogy in composition, less research on students’ writing process has appeared in recent years. Furthermore, using digital tools to both develop and study first-year composition (FYC) students’ writing processes is infrequently explored in composition and rhetoric despite the push towards studying the writing process in the now popular “Writing About Writing” curriculums (Wardle & Downs, 2017). By digital tools, I mean any digital or electronic tool, website, or digital program that aids in making processes easier or that aids in making users more aware of said processes. Specifically, the processes I refer to here are writing processes. And while digital tools have been used to give feedback on student writing, they haven’t been used to give feedback on students' actual writing processes. Perhaps part of this is because, as Takayoshi (2018) states: "for a discipline whose key term is composition (whether we call ourselves composition studies or rhetoric and composition), it is odd that over the past two decades we have so completely neglected to examine in any systematic or fine-grained way composition as a process" (p. 551). If composition and rhetoric had not been studying composing processes empirically for over 20 years, then research examining digital tools and how they may develop and/or allow students to study their own writing process invites further exploration. While some new research has emerged on this subject with the intention of shedding light on a previously mystifying practice, it is not always in a FYC context or an academic setting. For example, Takayoshi examined writers composing and writing processes on social media. She did so by having writers use a screen capturing program as they wrote comments, posts, or status updates on Facebook. Students then reflected on this writing via a think aloud process where they narrated their writing, revising, and editing throughout the process of writing on social media. Takayoshi found that some students conceptualized writing as a cognitive, social, and culturally constructed process of meaning making. Students had to think through the intentionality of their words, which was shaped by their intended audience and what cultural expectations there might be for that audience when composing on social media.

But scholars do recognize the potential for using digital tools to enhance writing processes in academic settings. For example, Rowell and Flicks’ (2019) recent work has explored how digital soundscapes (background music, ambient noises, and silence) affected composition students’ academic writing processes. They found that background music and silence were the most effective ways for students to compose because they were able to concentrate better. Ambient noises from online streaming services distracted writers and made it hard for them to compose even a little writing at a time. Other research even examines how online, digital comments that appear at the bottom of websites might be useful in generating ideas and conceptualizing audience. Gallagher (2018) specifically examined how requiring students to read online comments on blogs, websites, etc. could help students imagine authentic audiences and therefore write towards them before and during composing digital texts online. His research suggests that digital texts become enhanced when students imagine the type of audience that presents itself in digital, online comments.

While this research shows that digital tools can be very important to the writing process, there are many more digital tools that have the potential to deepen FYC students' understanding of their writing process, help them develop it, and do this all by studying their individual writing processes. I believe that another emerging technology, Google Chrome Draftback Web Extension (or as I will refer to it in the rest of this webtext "Draftback"), might hold the most potential to develop student writing because it combines and extends the before mentioned digital tools and gives the option for both instructors and students to research student writing processes in ways that make sense for both the expert (instructor) and novice (student). It does this by capturing students' writing processes through a screen capture, developing a video from the start to finish of a project, generating procedural data of the individual's writing processes, and keeping a document history.

James Somers created Draftback by reverse engineering Google Docs and using an add-on extension to Google Chrome (Sommers, 2014). It works in conjunction with Google Docs and can track the revisions that are made in a document by measuring keystrokes and deletions. Whenever a user changes the document at all, Draftback measures it as a revision. Therefore, it offers a useful set of procedural data to discuss with students. Draftback will also track the data and provide figures like the one below:

Screen Shot 2021 04 28 At 3.51.44 PM

Fig. 1. Draftback statistics from a draft of this webtext.

As Figure 1 details, Draftback measures the number of revisions, the time spent working in the document, and the distinct writing sessions, which are measured by how many times the writer completely closes the browser. 

The other interesting feature is the video capture. Here is an example from my writing:

Video 1: Google Draftback Video Sample

Draftback will play back the revisions that are done in the document in real-time or an expedited time. To see the revisions, writers do need to complete all of their writing in the Google Doc, so this can present logistical issues in a FYC class. If the writer does not complete their revisions in a Google Doc, it will not record their revision statistics. However, writers don’t need Google Draftback installed from the start of a writing project for all the revisions, time, and writing sessions to be tracked. Draftback will still look through the document’s revision history to track the revisions.

While Draftback offers the potential for interesting explorations in FYC, scholarship has just begun to explore this digital tool in developing and studying FYC student writing processes. Chamberlain (2018) discusses how she used Draftback in a similar context to FYC. She had her students in a junior-level Introduction to Writing Studies class record a think aloud watching their Draftback videos to develop deeper metacognitive awareness of their writing process. Students in these classes would watch their video, think-aloud, transcribe the think-aloud, and code the think aloud. The coding revealed that many students in the class suffered from writer’s block and had to begin to work through this cognitive block to finish their responses.

Chamberlain’s article is encouraging of Draftback’s potential, and in this webtext I’ll imagine Draftback’s use in a new context — a First Year Composition (FYC) course. I’ll further extend Chamberlain’s ideas by discussing the ways in which the procedural data generated by Draftback might be used to help students study their writing process and help teachers to respond to students’ writing processes in order to develop processes more effectively. I’d like to note that I also recognize that research has often explored how digital tools can give effective teacher response, and I won’t recount those numerous articles here.  However, to my knowledge, there exists no research on teachers giving feedback on students' actual writing process, specifically students' composing and writing habits. Furthermore, Google Docs has been examined in regards to enhancing peer review and collaborative writing (Neumann & Kopcha, 2019), but the procedural data can be a productive way in further illuminating students writing process and how they write. Ultimately, I fully believe that Draftback, its videos, and its procedural writing data provide useful digital tools to enable FYC teachers to respond to students’ writing processes and offer feedback on developing and studying them.

Important to also note here is a recent call by Holcomb and Buell (2018) to draw on large scale quantitative data to help composition teachers and scholars better understand how students write, and specifically, revise their writing. These two researchers examined revisions using a corpus software. What they found was that students were doing more sentence level revisions, rather than just word and phrase edits, but that students needed to be taught more about how to revise their writing holistically. Draftback provides both quantitative data and allows students to qualitatively examine how they compose via the video capture as well.

In what follows, I’ll begin by briefly discussing how I had students use Draftback. I’ll then give two examples of students, how I responded to their Draftback videos and procedural data, and how we collaboratively grew their writing processes in individualized ways. Next, I’ll examine factors that impacted these two students' writing processes. I’ll then talk about issues and problems with Draftback. Finally, I’ll explore future directions with Draftback.

Context of the Class

I first used Draftback in the Fall of 2016 in a FYC course, which will serve as an example of how others might utilize this approach in their FYC courses. The class was a modified Writing About Writing approach (Wardle & Downs, 2017) that began by building conceptual knowledge of writing thresholds concepts (i.e. rhetorical situation, genre, writing process, literacy being shaped by previous experience, writerly identity, and reflective practice) from Writing Spaces essays. Students then utilized their knowledge of these threshold concepts to write four distinct genres:  Literacy Narrative, Genre Analysis, Genre How-To Guide, and a Discourse Community Research Project. In addition, throughout the semester, I had students keep a writing journal where they would respond to course readings, reflect on course content, and give their thoughts about writing. At the end of the semester they analyzed these writing journals as a set of writing data and made suggestions to themselves about how they might grow as writers after the course was over.  A part of my class had students track their writing process via Draftback. After each major essay, they would write a reflection on their writing process. At this juncture, I gave some feedback on their Draftback stats to help improve their writing process if possible. To increase buy in, I informed students that I only watch their general typing habits to see if I could spot composing patterns that may be problematic. This ensures students that I’m not looking over their shoulder every minute of the videos even if I’m not in the room with them as they compose. Despite the need to overcome the unavoidable Orwellian factor that comes with this tool, it does afford composition teachers the unique option to intervene and help student writers build a deeper awareness of their writing process.

Sample Response to Google Draftback Videos/Statistics

To demonstrate how Draftback can be a tool for students to study their writing process and for their instructor to give good feedback on that process, I’ll turn to some sample responses I gave in my classroom. These sample responses are derived from a collaborative IRB-approved study with Marissa McKinley, Kelsey Hixson-Bowles, and Dana Driscoll. As a research team, we collected data and presented initial findings at the 2017 Conference on College Composition and Communication. Since then, I performed a deeper analysis that significantly expanded and revised that original project to what I am presenting here.

The sample responses I’ll highlight are for two students—Cherie and Tom (these are pseudonyms to protect participants’ identities), and I’ll additionally show their Draftback usage using tables. I show the tables instead of screenshots of the Draftback statistics because each student had a Google Doc for each draft of their paper. For Cherie, it was three, and for Tom, it was two. As a result, there are multiple sets of statistics, and a table will easily show a sum of all the statistics across these drafts. Cherie had the following statistics on the Draftback of all drafts of her Literacy Narrative:

Writing Time Number of Revisions Number of Sessions
3 hours, 20 minutes 8,269 18
Table 1. Cherie's Draftback statistics for her Literacy Narrative

The statistics demonstrate that Cherie has a strong writing process that we would hope that students would employ. That is, spending time on her writing across drafts, making revisions, and doing the writing over several days. In all, she spent over 3 hours on a 3-4 page paper, had over 8,000 revisions, and spent over a week on her writing process with 18 distinctive writing sessions. Her statistics tell me that Cherie put substantial effort and time on this project. Additionally, because of this process, Cherie wrote a first draft that definitely met the expectations of the genre and the course.  However, I encouraged her to continue working on the paper for the final draft by cutting down repetitive phrases, combining sentences, and developing key moments with further “show don’t tell” strategies of description, dialogue, and the use of figurative language. Because of this time and taking feedback seriously, Cherie wrote a quality final draft of her Literacy Narrative and therefore, earned an “A.” I had this to say about her writing process in my comments on Draftback: "Your Draftback video showed that you spent a lot of time working on this narrative over a period of time and that you had quite a few revisions. I think this is a writing process that will continue to serve you well, and I would suggest doing this process with the next paper in the class and other classes where you write." Cherie continued her trend of using ample time and making significant revisions on her genre analysis, which is demonstrated in the following stats from multiple Google Documents: 

Writing Time Number of Revisions Number of Sessions
3 hours, 38 minutes 6,000 18
Table 2. Cherie's Draftback statistics for her Genre Analysis

In comparison, Tom had a different writing process, which we see in his Google Draftback statistics across all his drafts:

Writing Time Number of Revisions Number of Sessions
2 hours, 49 minutes 5,110 7
Table 3. Tom's Draftback statistics for his Literacy Narrative

Obviously, Tom doesn’t spend as much time as Cherie, has fewer revisions, and not as many sessions. Based on the data from Draftback, it appears that Tom procrastinates on this paper and he ends up earning a lower grade than Cherie. It should be noted here that Tom also was battling personal issues, which likely caused him to put off his writing. I surmise this from the statistics, observations in class, and one-on-one meetings with him.  Therefore, I gave him the following comments on his writing process in Google Docs: "In regards to your Draftback video, I think you could maybe add a writing session or two to your overall process. I realize you were struggling through some things, but a little bit more time never hurt. It is something that will be useful in future classes and writing you might do in your professional life." When I commented on Tom’s writing process, I made a conscious decision to start a bit smaller so that he wasn’t overwhelmed. But overall, more time is what it appears that Tom needs to grow his overall process. He didn’t quite do this for his second paper, the genre analysis, but I gave him very similar comments on his second Google Draftback video with some suggestions of how to maximize his time.  During the final project, a discourse community research project, I saw a significant difference in how much time Tom spent on his writing across all his drafts:

Writing Time Number of Revisions Number of Sessions
3 hours, 49 minutes 8,303 10
Table 4. Tom's Draftback statistics for his Discourse Community Research Project

Tom added two extra days to his writing process, 3,000 more revisions, and nearly double the time on writing sessions. His grade reflects this as he earned his highest grade of the semester. Overall, these two examples demonstrate that this technology has the ability to help teachers productively intervene with students' writing processes in the form of comments. 

Studying Composing Habits and Writing Process Collaboratively

Besides these written comments that I give students, I also talk to students during individual conferences about their composing habits. This discussion happens after I’ve given the above comments on their first paper and while I’m giving them feedback on their second paper. In this meeting, I like to watch the video of one draft with students because I encourage them to try to spot issues with composing on their own. We also work together to identify how they might compose their writing better. Because these are often freshman and sophomore writers, this guided, collaborative study of their writing process is very useful. With Cherie’s video, I didn’t have much to say and really just encouraged her to compose and write the way she had been writing. For Tom, we noticed he started typing and stopped frequently. It then took him a while to start back up. I thought this was initially because he was thinking of what he might write next, but he said it was because he was distracted by the alluring XBOX 360 in his room. We decided together that he should change this up by starting a timer and composing for brief intervals with intermittent breaks. He tried this out and the next time we watched his video he would spend longer amounts of time typing between breaks. This seemed to also help develop his writing better due to my feedback in a verbal form, which might have been aided by the video as well.  Through oral feedback, written comments, and the viewing of the video, I had the opportunity to engage multiple learning styles and further individualize both feedback and the study of Tom’s writing process.  

Overall, Draftback and giving feedback on it highlights that our students’ writing processes will be different. Likewise, their writerly development will vary as well. Cherie and Tom both grew as writers, but Tom had more to grow upon at first, and it took him nearly the whole semester to spend more time on writing and revision. Draftback allows composition teachers to chart that growth and make suggestions along the way. As both Cherie and Tom are first-time freshman students, this feedback and eventual growth are vital towards their overall development both in my class and in future classes as well because this allows them to develop better writing habits.

Factors that Impact Writing Processes

It’s important to note that several factors might be at play with the students above. The fact that there are three distinctively different genres of writing in this course might be the reason that writing processes change. Also, perhaps both students have factors outside of the classroom that shape what they are doing for writing in the classroom. Cherie had a typical first semester of college. She also had a strong desire to become an English teacher someday (and now teaches high school English in her home state) and, therefore, wants to be the best writer she can be. Tom, on the other hand, was diagnosed with mental illness in the semester we were working together and sought treatment. Along with new composing habits, there’s a notable uptick in time spent on his writing after he began going to counseling. The point is that our students’ writing processes may be shaped by a variety of aspects, some which they can't control. Therefore, instructor responses must consider these factors as well and ultimately help FYC students develop that critical self reflection about their writing.  

Problems with Google Draftback

While Draftback has immense potential, it is certainly not without problems and does present logistical issues. As mentioned above, if students forget to write in a Google Doc, Draftback does not track their revisions. Additionally, students sometimes feel the need to start each draft in a new Google Doc. Doing this makes it tougher to look at Draftback statistics and videos together, makes it a problem when trying to do their reflective writings, and increases the difficulty with teacher response to their processes. Chamberlain (2018) notes that Google Docs is not the most secure program either.  In fact, I had to make sure not to put any grades within the Google Doc itself because it would violate FERPA guidelines; this has to be circumvented by posting grades elsewhere. Lastly, when developing the video capture, the videos must “render,” or load. Sometimes in this process the video rendering can buffer and then ultimately fail, which causes the user to re-render the video. This happens despite quality internet connections. While this does not happen frequently, it is frustrating when it does happen, which can deter the already reluctant writers that sometimes come into our classrooms. A disclaimer is necessary when loading the videos for the first time.

Future Directions with Draftback

I would like to continue working with Draftback in future classes because of its success and the unique potential it has to help productively intervene in students’ writing processes. One consideration I would recommend is to have students think more about how the different genres of writing might change students writing process, and therefore, Draftback statistics. With a new genre comes a new rhetorical situation, which of course may change how an assignment is written. I think students can benefit from a continual examination of their writing process via Draftback statistics, with each new type of writing they encounter. Instructor responses on a student's writing process would naturally need to change to fit this new genre and process as well.

Specifically, in terms of practical suggestions,  I’d encourage teachers and students to go through the “Changes” feature at the top of Google Docs which shows the history of the document. Here is an example from an earlier draft of this project:

Screen Shot 2021 04 28 At 3.58.46 PM
Fig. 2. History of Google changes from this web text

This highlights how we can easily see the changes that a writer makes to a document, which are shown in green above. That way, instructors are able to see the distinctly different drafts, and also have all the Draftback statistics in one location.

In addition, Draftback might lend itself to interesting multimodal examinations of students’ writing processes as well. A common assignment I have FYC students engage in is a “Writing Process Map.” Students will draw their writing process from start to finish with a particular writing assignment. Here’s an example from Cherie:

Fig. 3. Example of a student's writing process map

Then, students use these maps to make specific writing process goals. One goal is to highlight a strength of their writing process and make a plan for continuing that strength. Another goal is to examine a part of the writing process that could be improved upon with a plan to improve that goal. I find myself wondering: What interesting possibilities might there be with writing process maps and Draftback? Could students use the statistics to create stronger, richer maps? I also wonder—what other visual metaphors might they drum up after watching their Draftback videos? And how might this enhance their learning in interesting ways? It seems that Draftback allows for creative and innovative ways for students to visually map their writing process and allows different kinds of learners to think through their writing. 

Ultimately,  I’d also suggest using the combination of writing process maps, Google Doc Changes, Draftback statistics, and Draftback videos to make a “Writing Process Portfolio” that might be given to students at the end of the semester. Draftback might even hold interesting longitudinal possibilities across students' entire writing programs and careers. What if universities required students to track their Draftback statistics through all the writing classes in their college careers? Though there would be logistical issues with Draftback, using it throughout college, students could understand the way they write and compositionists would have a deeper awareness of how they help students achieve success with writing. Therefore, this digital tool could be a good resource for any FYC teacher or student if the logistical limitations and surveillance aspects are dealt with properly via framing and explanation.


Chamberlain, Elizabeth. (2018). Draftback to the future: A tool for writing process analysis. The Journal of Interactive Technology and Pedagogy. Retrieved from https://jitp.commons.gc.cuny.edu/draftback-to-the-future-a-tool-for-writing-process-analysis/

Emig, Janet. (1977). Writing as a mode of learning. College Composition and Communication, 28(2), 122–128.

Gallagher, John R. (2018). Considering the comments: Theorizing online audiences as emergent process. Computers and Composition, 48, 34–48. https://doi.org/10.1016/j.compcom.2018.03.002 

Holcomb, Chris & Buell, Duncan A. (2018). First-year composition as “big data”: Towards examining student revisions at scale. Computers and Composition, 48, 49–66. https://doi.org/10.1016/j.compcom.2018.03.003

Hixon-Bowles, Kelsey, McKinley, Marissa, & Powell, Roger. (2017). Google classroom hacks: How technology can help writing instructors cultivate students’ reflection, metacognition, and transfer of learning. Conference on College Composition and Communication, Portland, OR.

Murray, Donald. (1972). Teach writing as a process not a product. The Leaflet, 11–14.

Neumann, Kalianne L. & Kopcha, Theodore J. (2019). Using Google docs for peer-then-teacher review on middle school students’ writing. Computers and Composition, 54, 102524. https://doi.org/10.1016/j.compcom.2019.102524

Perl, Sondra. (1979). The composing processes of unskilled college writers. Research in the Teaching of English 13(4), 317–366.

Rowell, Christina & Flick, Amy. (2019) “I could get lit to Madonna.” Soundscapes & the first-year writer’s composing process. Computers and Composition, 54, 102513. https://doi.org/10.1016/j.compcom.2019.102513

Sommers, James. (5 November 2014). How I reverse engineered Google docs to play back any document’s keystrokes. JSomers. Retrieved from http://features.jsomers.net/how-i-reverse-engineered-google-docs/ 

Sommers, Nancy. (1980). The revision processes student writers and experienced adult writers. College Composition and Communication, 31(4), 378–388.

Takayoshi, Pamela. (2018). Writing in social worlds: An argument for researching composing processes. College Composition and Communication, 69(4) 550–580.

Wardle, Elizabeth & Downs, Douglas P. (2017). Writing about writing: A college reader. Bedford St Martin’s.


Created by matthew. Last Modification: Wednesday February 2, 2022 19:31:26 GMT-0000 by cheryl.