social media clutter header

Investigating Social Media Clutter

UX Research

Executive Summary

Digital clutter and digital hoarding have been shown to decrease mental health. However, current research focuses on general digital clutter, and solutions exist to fix the issue instead of preventing it. Our research focuses on social media and looks at the motivations and circumstances around why people save content, and how they interact with it. Digital clutter deeply affects usability and accessibility, and as a result, is a necessary topic to explore.

We started with an observational study with eight participants recruited through DePaul’s CDM Participant Pool. All participants were 18 years or older and willing to screen share their Pinterest, TikTok, or Instagram applications. We gave scenario-based tasks to participants to gather two main insights: how digital clutter builds up on social media, and how users engage with it.

We then interviewed eight participants, also recruited through the Participant Pool. All participants were 18 years or older and active social media users. We inquired about their behaviors and motivations for organizing, saving, retrieving, and trying content.

We created affinity diagrams and categorized our findings and insights into the following main themes:

We found that saving is social for all users. Many users save with the intent to share content with others later. Additionally, some users also send content as a way of saving, which provides them with an additional location to consider during retrieval. Given this, our solutions, such as a search feature, must account for content that is saved in messages as well as folders.

Convenience is highly valued by all users, but manifests in different ways for different user types, regardless of their degree of organization. Participants with high-effort organization systems prioritize easy retrieval, while those with low-effort or no organization systems prioritize immediate convenience. We suggest AI-driven folder systems or chronological filters for low-effort users, and a tagging feature to aid high-effort systems. A search feature and deleting duplicate saved content were highly preferred by both user groups.

We also found that content type affects users’ retrieval behavior. Users save three types of content: promotional (e.g. products, events, and restaurants), tutorial (e.g. recipes, DIYs), and inspirational (e.g. ideas for design, home decor, etc., as well as content they like). We found that users engage with saved content differently, in terms of frequency and length, based on the intent of retrieval and content type. Intent can be passive or active: users with passive intent tend to retrieve inspirational content, while users with active intent retrieve promotional or tutorial content. Promotional content prompted longer and more engaging digital interactions from users compared to other content types. Our recommendation is to introduce a nudge system to motivate users to act upon their promotional saved items. Additionally, an indicator on the content equipping users with relevant information (e.g. the duration of an activity or reviews for a location) can provide an additional push for users to try content. We also recommend a highlight reel for inspirational saved content. This would help resurface buried content and bring up positive emotions for users.

For our next steps, we will prototype solutions from the priority matrix and evaluate them with users to determine the efficacy of these solutions in mitigating digital clutter.

Introduction

Through advancements in technology, users can create and collect digital content, such as videos or files, more efficiently than ever. As a result, the volume of digital content in an individual’s digital collection continues to increase and users are now faced with digital clutter. Unlike physical clutter that might exist in a few physical spaces, digital clutter spans multiple personal tech devices, websites, and even social platforms (Uğur & Caliskan, 2022; Zhao et al., 2013). For example, a user might have pictures on their phone and computer, in iCloud and Google drive, and multiple social platforms such as Instagram and TikTok. Digital spaces are more flexible and have less spatial limitations than physical spaces, thanks to technology like cloud storage (Sweeten et al., 2018). Thus, users are more likely to build up digital clutter as compared to physical clutter.

Digital clutter and digital hoarding have been shown to decrease mental health and well-being, so it is important to explore how to mitigate it (Sweeten et al., 2018; Neave et al., 2020). Digital clutter is a human-computer interaction issue at its core because it affects usability and impacts the cognitive load of using interfaces. Additionally, mitigating digital clutter can enhance accessibility by ensuring organized and clean digital spaces for everyone.

Our study adds to the current conversation of digital clutter by exploring how it builds up on individuals’ social media and users’ current mitigation strategies. Current research mostly focuses on digital clutter as a whole but does not look at specific digital spaces, like social media. We hope to bridge this gap with our research. Additionally, existing products to deal with digital clutter, such as Unroll.me for email, focus on responding to digital clutter that has already built up. We believe a holistic approach to look at how and why digital clutter accumulates is needed. Our study seeks to understand the motivations and circumstances behind saving content on social media. Specifically, we look at users’ saving behaviors and how they interact with saved content. Through our research, we hope to identify design implications to mitigate digital clutter on social media platforms that could ultimately improve mental health.

Methods

OBSERVATION.

Participants

We recruited eight participants through the CDM Participant Pool. All participants were active users of social media, 18 years or older, and willing to share their mobile screens. Seven of the eight participants were female, and one was male. We informed participants that their screens, which displayed either TikTok, Pinterest, or Instagram, would be recorded for observational purposes. Six participants shared their Instagram content, one shared Pinterest, and one shared TikTok.

Data Collection

We met participants remotely via Zoom and asked them to review and sign a consent form before the study. We then introduced the project and gained further verbal consent from the participants.

We presented participants with two scenario-based tasks, and asked several follow-up questions. For the first task, we were interested in observing how content clutter might build up. For the second, we wanted to observe how they interacted with this saved content. To reduce potential bias, we let participants perform tasks uninterrupted and only spoke when necessary. Finally, we asked follow-up questions to assess their feelings and thoughts on the study. We went back to the study recordings to take notes and ensure we recorded accurate data.

Analysis Methods

To learn more about the participants' behaviors, we applied the AEIOU (Activities, Environment, Interactions, Objects, and Users) framework. The team applied open coding to conduct a qualitative analysis, and used FigJam to create an affinity diagram using our notes.

INTERVIEW.

Participants

We recruited eight participants from the CDM Participant Pool. All participants were adults and active social media users. Their ages ranged from 23 to 38, with five females and three males. Most participants were master's students, with three also working.

Data Collection

All interviews were conducted remotely via Zoom. Before conducting the interview, we obtained participants’ consent and inquired about their preferred social media platforms. The interview was divided into six sections of deep-focus open-ended questions.

We first focused on users' behaviors and motivations for saving, organizing, retrieving, and trying content. Next, we explored the connection between social media, organization, and mental health. Finally, we inquired about users’ awareness of and preferences for technologies to help mitigate digital clutter, and their perspectives on our proposed design solutions. We also collected demographic information before concluding the session.

Analysis Methods

We independently coded our interview transcripts using Atlas.ti and developed a codebook outlining the themes that emerged from our interviews. We used Mural to collaborate on an affinity diagram that revealed our main findings. This data informed our persona spectrums, personas, scenarios, journey maps, and priority matrix.

Findings

We summarized our findings from the observations and interviews into four categories using an affinity diagram:

BEHAVIORS.

Saving Content

All participants saved content to folders on social media applications. Most had personalized organization systems. For example, one interviewee interested in soccer, cooking, and movies had one folder for each category. Participants with such organization systems only saved new content to contextually-relevant folders; if a piece of content did not fit in their existing folders, they created a new folder for it.

All participants also had a catch-all folder, which one participant likened to a junk drawer. Examples of this for social media include the default “all posts” folder on Instagram, the “likes” tab on TikTok, and personally-created catch-all folders. Users who had personalized organization systems used their catch-all less frequently.

Finally, some participants utilized additional methods, such as saving to collaborative folders, downloading content to their devices, screenshotting content, and sending content to friends.

Organizing Content

Three interviewees unsaved or deleted content as a way of organizing their folders, while five moved content to new folders. However, none set aside time specifically to organize; participants did so when they came across content during retrieval that they noticed was misplaced or no longer interesting to them.

Revisiting Content

Most participants’ process of retrieval was trying to remember when they saved it, followed by “a lot of scrolling” through their folders or messages.

After retrieving content, almost all participants viewed the post, comments, or captions again, or sent it to their friends. Only three mentioned taking additional actions on posts, including taking a screenshot, visiting the poster’s profile, or clicking an external link.

Content Type and Interactions

We noted three primary categories of content saved or retrieved by users: promotional, tutorial, and inspirational. Promotional content includes products, events, and restaurants; tutorials include recipes and other DIYs; and inspirational content includes general ideas for design, home decor, etc., or content that the user simply enjoys.

During the observation, promotional content prompted most actions, with participants clicking on links associated with these posts. Some participants opened external applications to continue browsing about the topic. For example, one user found a restaurant they had saved a video about, and looked up reviews and directions. As such, interactions with promotional content were longer. While interviewees had relatively less digital interaction with tutorial content, they noted that it prompted further real-life action from them.

Inspirational content had less frequent and shorter interactions overall. Participants typically viewed these posts without further engagement.

MOTIVATIONS FOR SAVING AND ORGANIZING.

Content Retrieval

All participants stated that they saved content they aimed to revisit for varying reasons, and saving would allow them to find it more easily.

Sharing

All participants noted a social aspect to their saving habits. Five interviewees stated they saved content to share with friends later, and some sent content to their friends as a way of saving.

Curating the Social Media Experience

Most participants viewed saving and curating as valuable to their social media experience. Organizing saved content helps participants parse through general social media content and control their digital spaces. Half of the participants identified a nostalgic aspect of revisiting content, with one referring to their social media folders as a “personal archive."

EMOTIONS AND ROADBLOCKS.

Content Retrieval

Regardless of their level of organization, all participants found revisiting older saved content challenging. One participant stated the following about their folders: “Sometimes it's hard to remember what all I have in there. Some things just get lost essentially.”

Unsuccessful content retrieval attempts were identified as the primary reason for negative emotions in participants. Three participants also felt guilty or burdened with the amount of content they had saved. Relatedly, most participants associated organization with positive emotions.

Trying Content

Almost all participants noted feelings of satisfaction, motivation, or accomplishment after successfully trying saved content. The main roadblock preventing users from trying content was the perceived time or effort required to do the activity. However, half of the participants noted that “it’s just social media,” and felt no negative emotions towards the amount of saved content they had not tried.

USERS WANT PERSONALIZATION AND EASE OF USE.

When asked about each of our design recommendations, all participants mentioned that they prioritize personalization and accuracy in any solution. Searchability was the most important feature for all participants, and was brought up as the ideal solution for two participants before we asked about it.

Priority Matrix

Feature Feature Category Priority Impact Feasibility
Search Retrieval Highest High High
Nudge to do saved content Trying content Medium Medium Medium-Low
Nudge to clean up content Organization Low High Medium
Highlight reel of old saved content Retrieval Low Medium High
Filter Retrieval High High High
AI Folders Organization Medium High Medium
Time Commitment Indicator for DIY Trying content High Medium Low
Effort Indicator for DIY Trying content High Medium Low
Reviews/Popularity rating for restaurants Trying content Medium Medium Low
Hashtags for saved content Organization High High High
Deleting duplicates Organization Medium Medium High

Discussion

Our goal for this study was to contribute to existing literature about digital clutter. We explored its presence on social media, considered the motivations and behaviors behind it, and examined its potential effects on mental health.

Based on our findings, our team identified five key insights:

USERS SAVE FOR RETRIEVAL.

All of our participants verbalized that they save content intending to reference it later – including those who save to share later, as they must also retrieve it before sharing. Participants with high-effort organizational systems could access their saved content more easily. It is important to highlight this relationship between organization and retrieval. The introduction of a search feature can potentially simplify users’ ability to retrieve saved content.

SAVING HAS A SOCIAL ASPECT.

The majority of participants mentioned that they save content to send to others later. Some users also reported that they view sending content to friends as a way of saving instead of using folders. These participants relied on remembering who they sent specific content to for retrieval. Given the social aspect of saving, solutions aiming to mitigate digital clutter, such as a search feature, must account for content that is saved in messages, in addition to users’ saved folders.

CONVENIENCE IS A PRIORITY.

Our participants consistently prioritized convenience, regardless of their degree of organization. Participants with high-effort organization systems, such as distinct categorical folders, achieved long-term convenience by using those systems to maintain organization and easy retrieval. Alternatively, users with low-effort or no organization systems received immediate convenience by saving content without organizing it. Adding an AI-driven organizational system has the potential to simplify the content organization process, increasing convenience for users. Another recommendation that can improve convenience is deleting duplicate saved content.

CONVENIENCE OPTIMIZATION VARIES BY USER TYPE WHILE SAVING.

Our participants used one of two organization systems: high-effort organization by creating folders and organizing while saving, and low-effort organization by saving content into a single catch-all folder. These behaviors highlight the different approaches our participants took to achieve convenience.

All users relied on chronology to parse through saved content. Additionally, unsuccessful content retrieval attempts (e.g. with older saved content) were the primary cause for negative emotions. As such, we recommend the ability to filter by date so that users can easily reference content based on when they remember saving it.

High-effort organizational systems

Users with high-effort organizational systems prioritized long-term convenience, striving to make the process of retrieving saved content easier by exerting more effort during the saving process. The introduction of a tagging feature has the potential to help these users effectively manage their saved digital clutter. Additionally, by tagging content during the saving process, users could later search these up during retrieval, quickly accessing the content. Easy and successful content retrieval encourages users to engage with or organize their saved content.

Low-effort organizational systems

Users with low-effort organizational systems prioritized short-term convenience, making the saving process as passive as possible. As a result, during retrieval, some participants ended up feeling overwhelmed by the amount of content they had saved. The introduction of AI-driven suggested folders would allow for immediate convenience by removing the need for them to manually create folders, while improving organization. This would facilitate easy retrieval and organization in the future.

CONTENT TYPE AFFECTS BEHAVIOR.

Participants engaged with saved content differently, in terms of frequency and length, depending on whether the content was promotional, tutorial, or inspirational. Despite varied engagement, almost all users associated trying content with positive feelings. Introducing a technology-based nudge system has the potential to motivate users to act on their saved promotional or tutorial items, such as by visiting a saved restaurant or cooking a saved recipe. After completing the action, the system could prompt users to unsave the post, reducing the buildup of saved content. Furthermore, an indicator on the content with relevant information (e.g. the duration of an activity, or reviews for a location) can provide an additional push for users to do the activity.

Active intent

Participants who retrieved saved content with a goal actively engaged with the content; they either shared it, interacted with the post, or took real-life action. This type of content was typically promotional or tutorial.

Passive intent

Participants who lacked a goal for retrieving saved content engaged with it passively, without taking any actions. This content was often inspirational. Users also noted feeling nostalgic when looking at older saved content they enjoyed. Based on this, we recommend adding a highlight reel to resurface older, buried content.

LIMITATIONS.

Although our research offers valuable insights, it is important to address its limitations. In our study, participants were able to share from either Instagram, Pinterest, or TikTok. However, there was an unequal distribution of users across these platforms with the majority favoring Instagram and fewer selecting TikTok or Pinterest. Each of these social media platforms provides different ways for users to browse, save, and revisit content. Given our uneven distribution of participants across platforms, we recognize that our data is primarily influenced by user experiences with Instagram and thus may not accurately reflect user behaviors, attitudes, and experiences across all major social media platforms.

Furthermore, we recruited through the DePaul CDM participant pool, limiting our demographics in terms of age and education. This may not accurately reflect the diverse profiles of many social media users who struggle with digital clutter. Therefore, our findings may have low external validity, making them less generalizable to broader populations.

Next Steps

Moving forward, we plan to conduct additional interviews to further explore digital clutter on social media, including the motivations and behaviors behind it. It would be beneficial to test a larger sample size of participants with experiences across a wider range of social media platforms. Through these interviews, we hope to discover insights that can inform further implications for design. Furthermore, we aim to conduct usability tests to evaluate the effectiveness of our current design recommendations in mitigating saved content clutter on social media.