Contents
- Online music collaboration statistics from Songtree’s co-creation database
- Online music collaboration statistics: online music-making during lockdown
- Online music collaboration statistics: asynchronous workflows and platform experiences
- online music collaboration: Online collaboration patterns among working musicians
- online music collaboration: Real-time network music performance and audio delay
- Sources
Online music collaboration statistics from Songtree’s co-creation database
A Songtree database snapshot dated November 13, 2019, contained 295,193 registered users and 57,868 authors. Those counts describe different database categories: the study distinguishes users from authors, so they should not be treated as interchangeable measures of active collaborators. The same original dump contained 263,526 songs. These are historical platform records, not estimates of the number of musicians using online collaboration tools today. Source: Songtree co-creation study, Table 1
The dump’s song categories show why the total needs context. It listed 200,336 new songs and 63,190 overdubs. An overdub is a user-added track on an existing recording, a format that can support contributions by multiple people. The same snapshot also listed 26,945 self-overdubs, which do not necessarily represent work between separate authors. These are counts of database items, not a direct count of collaborative projects or people who collaborated. Source: Songtree co-creation study, Table 1
| Songtree database category | Count in original dump |
|---|---|
| Users | 295,193 |
| Authors | 57,868 |
| Songs | 263,526 |
| New songs | 200,336 |
| Overdubs | 63,190 |
| Self-overdubs | 26,945 |
The snapshot also recorded 1,022 contest songs, 513 closed songs, 47,504 hidden songs, 271 orphan songs and 1,199 remixes. These categories matter when interpreting the database’s scope: contest, closed, orphan and remix entries were among the categories excluded from the study’s final analytic dataset, while hidden songs were not necessarily publicly discoverable. The paper’s final analysis covers activity from April 2015 through October 16, 2019; it is a filtered set with a different scope from the November 13 dump. The category counts therefore should not be added together as distinct collaboration behaviors. Source: Songtree co-creation study, Table 1 and dataset description
Online music collaboration statistics: online music-making during lockdown
A survey on joint music-making collected 234 valid responses between April 20 and May 1, 2020. Its respondents were recruited online through music communities, and 95% lived in Belgium or the Netherlands. The results describe this respondent group during the first lockdown period; they are not a population estimate for musicians everywhere. Source: Frontiers, “Impact of Lockdown Measures on Joint Music Making”
The survey separated music-making with others while physically present from online real-time activity and alternative remote methods. For physically present social music-making, 234 respondents (100%) answered yes before lockdown; during lockdown, 49 (21%) did. For online real-time joint music-making, the reported yes count was 11 (5%) before lockdown and 40 (17%) during it. The two series represent different ways of making music together, so a change in one does not stand in for the other. Source: Frontiers, Tables 3a and 4a
| Joint music-making method | Before lockdown: yes | During lockdown: yes |
|---|---|---|
| Physically present | 234 (100%) | 49 (21%) |
| Online real-time | 11 (5%) | 40 (17%) |
| Alternative remote methods | 67 (29%) | 129 (55%) |
Alternative remote methods were reported separately from online real-time playing. Their yes count rose from 67 respondents (29%) before lockdown to 129 (55%) during lockdown. In this survey, the category included approaches such as recording parts separately and combining them later. The measures capture respondents’ reported activity in the specified periods, not the number of recordings or the quality of finished music. Source: Frontiers, Table 6a and results
Online music collaboration statistics: asynchronous workflows and platform experiences
Among respondents who used alternative methods during lockdown, 101 (78%) said they recorded their parts separately and then pasted the parts together digitally. Another 3 (2%) reported playing along with pre-recorded material, such as music accessed through Spotify or YouTube. A further 23 (18%) used both of those methods. These reported method counts describe the alternative-method user subgroup, rather than all 234 survey respondents. Source: Frontiers, results on alternative remote methods
This pattern distinguishes asynchronous collaboration—participants contribute at different times—from playing together through an online real-time connection. The survey records whether respondents used these approaches; it does not compare their production quality, time demands or artistic outcomes. The method figures should therefore be read as a description of reported workflow choices during lockdown, within a geographically concentrated online survey sample. Source: Frontiers, “Impact of Lockdown Measures on Joint Music Making”
Among respondents who used online real-time platforms during lockdown, 18 reported using Zoom, 17 reported Skype and 14 reported Messenger. The counts can overlap because respondents could report more than one platform; adding them would not give a count of unique people. These figures document use within this survey and period, not the current popularity or technical suitability of those services for music production. Source: Frontiers, results on online real-time methods
online music collaboration: Online collaboration patterns among working musicians
A separate study surveyed 101 working musicians recruited from 4 music-creator communities: Young Entertainment Professionals, Grammy Museum, Audio Engineering Society and ASCAP. Its regression table reports estimated coefficients for survey measures. A coefficient is not a percentage, and these estimates should be interpreted within this sample and model rather than as a direct count of musicians changing their behavior. Source: Frontiers, “Collaborating in Isolation”
For time spent making music online, the estimated coefficient was β = 2.612 for the early pandemic period relative to before the pandemic (p < 0.05). For the most recent pandemic month at survey completion, the corresponding coefficient was β = 3.312 (p < 0.01). Both estimates concern the study’s online music-making measure. They do not specify a percentage change or, by themselves, establish why the measure differed across periods. Source: Frontiers, Table 3, “Time making music: Online”
Other early-period coefficients were negative for collaborative rather than solitary music-making (β = −2.261, p < 0.05), initiative in seeking collaborators (β = −3.045, p < 0.01), and activity with new collaborators (β = −2.561, p < 0.05). Separately, the model estimated a positive coefficient of β = 2.892 (p < 0.01) for social-media usefulness for collaboration as time since shelter-in-place increased. These are model estimates for distinct survey measures, not direct percentages of musicians or causal effects. The study notes that p-values were not corrected for multiple comparisons, another reason to avoid treating individual coefficients as standalone population findings. Source: Frontiers, Table 3 and statistical-analysis notes
online music collaboration: Real-time network music performance and audio delay
A controlled network music-performance experiment involved 22 musicians performing in 11 pairs. It examined real-time performance conditions, which differ from asynchronous file-based production: participants played together over a network, where audio delay and quality can affect the experience. Source: MDPI, “Impact of Audio Delay and Quality in Network Music Performance”
The authors reported that one-way mouth-to-ear (M2E) delays of 40 ms still yielded positive quality-of-experience ratings for many participants. They described performances as synchronized and satisfactory even when one-way M2E delay reached 40 ms. This is a finding from the experiment’s participants and tested conditions, not a universal latency threshold for every instrument, network, or collaboration setup. Source: MDPI, conclusions on delay and quality of experience
In one tested audio-quality condition, the authors reported that perceived audio quality was unaffected when the sampling rate was reduced from 88.2 kHz to 8 kHz. The paper also notes that subsets of participants, notably pianists, were more sensitive. The result belongs to the study’s controlled scenario and should not be read as evidence that every musician or production task will find the same reduction imperceptible. Source: MDPI, conclusions on sampling rate and participant sensitivity
Sources
- How many artists collaborate without ever meeting? – Music Collaboration Survey
- An In-Depth Analysis of Occasional and Recurring Collaborations in Online Music Co-creation
- Impact of Lockdown Measures on Joint Music Making: Playing Online and Physically Together
- Collaborating in Isolation: Assessing the Effects of the Covid-19 Pandemic on Patterns of Collaborative Behavior Among Working Musicians
- Modeling Online Creative Collaborations
- Remote Collaboration Software Survey – The Results
- Impact of Audio Delay and Quality in Network Music Performance