Survey Sunday #5: Using and Finding Things at the AO3

About the answers

This week on Survey Sunday, as requested, we’re releasing the first batch of AO3 related questions. These were:

  • Question 15. How useful do you find the Archive’s tagging system?
  • I’d be very interested in the results for Questions 19 to 21, concerning kudos and comments on the Archive.
  • I’m very interested in the answers to 32 and 33 specifically, and to a slightly lesser extent, all of the opinion-based questions about individual AO3 functions. (15, 16, 22, 23, 24, etc.)

Today we’ll be answering questions #10 to #17. These relate generally to how many people use the Archive and how they do so. All questions presented in this post are complete and include final numbers. Future posts will shed light on the ease-of-use of the different Archive features and other general opinions about the Archive.

It’s important to note that everybody who replied in Question #10 that they hadn’t used the Archive would have skipped all the other questions related to the Archive, though this fact will be mentioned when relevant in the corresponding sections.

#10 — Have you ever used the Archive of Our Own (AO3)? (n = 5794)

Answer options Response count Response percentage
Yes 5564 96.0%
No 230 4.0%
Answered question (n) 5794
Skipped question 192

Question #10 asked whether the respondent had ever used the AO3. 3.2% percent of the people who took this survey skipped this question.

Of those who answered, 4% said they didn’t use it and 96% said they did. As the question did not distinguish between frequent users and someone who had been to the Archive once or twice, further questions were designed to shed more light on users’ activity.

One thing to note about this result is that, as AO3 users received emails informing them of the Survey, unlike users from other OTW projects, we were likely to get a very high “Yes” rate on this question.

graph for question 10, description in the text above.

#11 — Do you have an account at the Archive of Our Own? (n = 5555)

Answer options Response count Response percentage
Yes 4.583 82.5%
No 792 14.3%
I’m waiting for an invite 180 3.2%
Answered question (n) 5555
Skipped question 431

Question #11 also testifies to the influence of AO3 users finding the Survey through direct contact. This question asked how many respondents already have an AO3 account. 201 people skipped this question on purpose, and 230 skipped it automatically by answering ‘no’ on question #10.

A clear majority of respondents were account holders — 82.5%. Only 3.2% of those answering were waiting for an invitation. Back in April, during the survey period, the AO3 invite list was around 5000, roughly a sixth of the current waitlist. This is, however, still a small response from that group.

The number of accounts at that time was around 42,000, meaning that at least 10% of AO3 account holders answered the Survey, compared to around 3% of those on the invite list. This difference in response suggests that either account holders are more likely to be aware of OTW news and events, or, more probably, that the proportion of respondents in this category is substantially larger because they were emailed individually about the survey.

graph for question 11, description in the text above.

#12 — How efficient do you find the Archive’s invitation system (invite queue or sending invites)?

Answer options Response count Response percentage
Very efficient 1227 25.9%
Somewhat efficient 1180 24.9%
No particular opinion 2029 42.8%
Inefficient 254 5.4%
Very inefficient 46 1.0%
Answered question (n) 4736
Skipped question 1250

Question #12 showed a growing skip rate as well, with 1,250 people (21%) of the people who took the survey, not offering an opinion about AO3’s invite system — these people encompass those who answered ‘no’ to #10 (230), people who answered they didn’t have an account and weren’t waiting for one in #11 (792 — who also automatically bypassed this question) and people who just decided not to answer (228).

The highest response was the neutral ‘no particular opinion’ (2,029 or 42.8% of the people who responded). However, the invite system was at the time relatively favourably regarded with a 25.9% considering it ‘very efficient’, and another 24.9% considering it “Somewhat efficient.” Only 5.4% felt it was “Inefficient” and a mere 1% considered it “Very Inefficient”.

This opinion might be considerably different if more than 180 people answering the survey were currently in the invite queue or if the survey were run today, when nearly 30,000 invite requests are outstanding.

It may also have been different if the answer scale’s positive and negative sides had been written with the exact same vocabulary — the fact that one could not answer ‘somewhat inefficient’ may have pushed people holding more moderately negative opinions towards neutrality (not wanting to answer ‘inefficient’ outright) while one could answer ‘somewhat efficient’, attenuating the possible positive responses and thus encouraging them. The influence of the phrasing, however, can’t be measured, and this is only one interpretation of its possible effect.

graph for question 12, description in the text above.

#13 — What are you using the Archive for?

Answer options Response count Response percentage
For subscribing to creators/works 2654 47.9%
For viewing fanworks (e.g. fanfiction, fanart, audiofic…) 5279 95.2%
For bookmarking fanworks 2887 52.1%
For posting fanworks 3199 57.7%
For running challenges 110 2.0%
For participating in challenges 1018 18.4%
Other 168 3.0%
Answered question (n) 5543
Skipped question 458

Question #13 looked at what people were doing at the AO3. The number of skips dropped back to 7.4% of all respondents, only 228 of whom skipped this question directly (the other 230 had automatically bypassed the whole section by answering ‘no’ to question #10). 92.6% (5,543) of the people who took this survey answered this question in some manner, either choosing one or more options, choosing one or more options and writing in clarifying notes about them or other choice(s), or just writing in other choice(s).

95.2% of the people who answered said they used the Archive to ‘view fanworks’ — perhaps predictably, by far the most popular choice. Only 57.7% said they post fanworks. 52.1% said they use it to bookmark fanworks, and 47.9% that they use it to subscribe to creators or works. 18.4% said they use it to participate in challenges, while only 2.0% said they use it to run challenges (mods generally being fewer than participants, this is not a surprise).

graph for question 13, description in the text above.

‘Other’ answers add up to 169 (some respondents wrote in more than one, and each action or feature mentioned is counted separately, so these are actually from 166 people). Of these other answers, 35.3% were ‘download fanworks’, 21.0% ‘leave feedback of some kind’ (comments or kudos or both), 15.6% ‘to find fanworks’, and 7.8% ‘to calculate stats’, either personal or general for both academic and personal enjoyment reasons. Still 20.4% (34) of these answers are grouped in a general ‘other’ category — none were common enough to generate significant patterns. Some of the options include using the Reading History or Mark to Read Later features, using Collections rather than Challenges, following feeds, participating in the community, and tag wrangling.

graph for breakdown of the 'other' option of question 13, description in the text above.

#14 — Have you ever used the Archive’s search function?

Answer options Response count Response percentage
Yes 5161 93.3%
No 368 6.7%
Answered question (n) 5529
Skipped question 457

In Question #14 457 people (7.6% of the people who took this survey) skipped answering whether they had used the search function at AO3, only about 227 of whom skipped this question inentionally rather than automatically by answering ‘no’ to question #10.

Of the 5,529 who did answer, 93.3% said they had used the Archive’s search function and 6.7% said they hadn’t. Those who had were asked the following 3 questions — #15, #16 and #17 — and those who hadn’t were made to skip directly to the questions about leaving feedback on the Archive (those questions will be analysed in a following post).

graph for question 14, description in the text above.

#15 — How useful do you find the Archive’s tagging system?

Answer options Response count Response percentage
Very useful 2569 49.8%
Somewhat useful 2004 38.9%
No particular opinion 189 3.7%
Of limited use 318 6.2%
Not useful 77 1.5%
Answered question (n) 5157
Skipped question 829

Question #15 asked about tags. Almost double (839 or 14%) the number of respondents skipped this question as the last — these people include the 230 avoiding the entire set of questions by answering ‘no’ to question #10, the 368 people who said they hadn’t ‘used the Archive’s search function’ in #14, and 231 who simply decided not to answer.

The majority of people who did answer this question, 2,569 (49.8%), found tags very useful, with another 2,004 (38.9%) respondents finding them somewhat useful. This indicates that there is an overall positive opinion of the tag system in place, since 87.7% of the respondents find the tag system useful or very useful.

A 3.7% felt neutrally about it, while 318 (6.2%) found them ‘of limited use’ and 77 (1.5%) found them ‘not useful’. Negative opinions thus add up to a 7.7% of respondents.

graph for question 15, description in the text above.

#16 — How efficient is it to search and find works on the Archive?

Answer options Response count Response percentage
Very efficient 1260 24.4%
Somewhat efficient 2678 51.9%
No particular opinion 284 5.5%
Inefficient 717 13.9%
Very inefficient 216 4.2%
Answered question (n) 5155
Skipped question 831

Question #16 asked about the efficiency of finding works on the AO3. Around the same number of people skipped this question as the last, 831 (14%) — which again can be divided into the 230 people who are bypassing the whole Archive section, the 368 who answered they hadn’t ‘used the Archive’s search function’ and 233 who skipped this question intentionally.

The majority of respondents found the search “Somewhat efficient” (51.9%) or “Very efficient” (24.4%), giving us a total of 89,9% of respondents who have a positive opinion of the Search function. On the other hand, 13.9% found it “Inefficient” and a 4.2% found it “Very inefficient”, adding up to 18,1% who have a negative opinion. This is more than double the percentage of people who had a negative opinion of the tagging system.

5.5% of respondents expressed neutrality by choosing ‘no particular opinion’.

As we’ve noted in the analysis of the answers to question #12, responses may have been different if the answer scale used the exact same vocabulary for its positive and negative sides — the fact that one could not answer ‘somewhat inefficient’ may have pushed people holding more moderately negative opinions towards neutrality (not wanting to answer ‘inefficient’ outright) while one could, in fact, answer ‘somewhat efficient’, perhaps attenuating the possible positive responses and thus encouraging them. The influence of the phrasing, however, can’t be measured, and this is only one interpretation of its possible effects.

graph for question 16, description in the text above.

#17 — Did you find content in your fandom(s) on the Archive?

Answer options Response count Response percentage
Yes, always or most of the time 4201 81.5%
Only sometimes 634 12.3%
Same volume as anywhere else 250 4.9%
Rarely 67 1.3%
No, never 0 0.0%
Answered question (n) 5152
Skipped question 834

Question #17 addresses AO3 content, asking if users’ fandoms were represented in the Archive’s content. 834 people skipped this question (14%) — this again gives us 236 people skipping this question intentionally, with the others skipping it automatically as a consequence of previous answers.

Not surprisingly, most people said ‘Yes’, with 4,201 (81.5%) finding content most of the time and only 634 (12.3%) finding it only sometimes. No one said they never found such content, and 250 (4.9%) said they found it in the same quantities as elsewhere. Only 67 (1.3%) said they rarely found their fandoms at the AO3.

These answers are logical if one considers that it would be unusual to keep using an archive when one was unable to use it for one of its main purposes. One important aspect we did not take into account, however, is multi-fannishness. That is, respondents might find many works in one of their fandoms, but few or none in their other fandom(s), and the question design did not allow them to differentiate this. Judging from the overall positive slant, we assume that multifannish people tended towards answering with their well-represented fandom(s) in mind rather than their underrepresented one(s).

graph for question 17, description in the text above.

As a final note for this section, we note that the number of people choosing not to answer (or ‘skip’) a question has mostly stayed constant through the analysed questions, being around 200.