Finding the voice: Planning and evaluation of social media in cultural institutions
AbstractCultural institutions have sought to develop social media as a means of engaging with their audiences. However, there is a lack of guidance about how to effectively evaluate performance. A study was carried out to explore contemporary practices in cultural institutions to better understand the challenges, methods, and strategies. This involved in-depth interviews with social media managers at eight Scottish institutions and a worldwide survey of one hundred organisations. The participants were asked about how they use social media, what their objectives are, and whether use of frameworks, toolkits, and strategies were useful or indeed feasible. Results showed that despite various approaches, many institutions share priorities and challenges. Although some preferred to be flexible, while others more regimented, there is consensus that social media helps deliver on wider institutional goals and strategies. There is also evidence that institutions want to better understand the impact of their social media but often have problems capturing or interpreting relevant data, and all institutions are responsive to tools that might help them do this. The research tested the principles of one tool in particular, the Culture24 Evaluation Framework. Discussions with participants revealed that while its concepts were relevant and valuable, there remained issues with how it could be implemented, given incongruity with existing operations. Consequently, the research proposes recommendations for developments in social media and evaluative frameworks. The research concludes that institutions that have more defined strategy, even if practices remain flexible, are more decisive and effective in their use of social media. Having objectives and cycles allows better implementation of specific tools and frameworks. Furthermore, careful consideration of the functionality of specific social media platforms allows them to more effectively address specific targets and metrics.
Keywords: Social media, evaluation, strategy, digital, engagement, framework
From messaging family back home to documenting a conference, and from tracking the latest travel news to making customer enquiries of hotels and airlines, social media has become a central part of our personal and commercial lives. We expect it—of friends, families, and businesses. Cultural institutions are equally aware of this, and in recent years there have been developments in the application and understanding of social media by museums, archives, libraries, and other cultural institutions. More institutions are becoming present, recognisable, outgoing, analytical, and tactical in their use of social media.
However, like any other businesses, these institutions have to find an approach that works for their own particular circumstances, with few definitive rules and an ever-changing environment. Combine this with trends within the sector towards reduced resources, a marketplace crowded and complicated by technology, and an ethos of universality and inclusivity, and cultural institutions face significant challenges to deploying effective social media.
This paper focuses on devising and planning for strategic development and evaluation of social media. Combining a thorough literature review and existing frameworks for planning and measuring social media impact, we present the results of a survey used to explore the current practices of cultural institutions on social media, validate a proposed tool for cultural institutions (the Culture24 Evaluation Framework), and provide recommendations and best practices. These include having a clear purpose and strategy for implementing social media and analytic tools, planning for the use of functionalities of particular platforms, and sharing knowledge and experience in order to optimise social media performance.
Over the last decade, the literature has slowly developed from early exposition of the merits and functions of social media towards more discursive consideration of planning and best practices: we now know that that social media use is about two-way communication, and that receipt of such user input could be used to help make strategic decisions (Fernandez, 2009, 36). With this, there was the introduction of tools to help plan social media (like SWOT and PEST analysis), with the overall intention of securing stakeholder support and developing loyalty to the institution. Around the same time, social media was seen as increasingly vital for competitiveness in terms of loyalty and technology (Kho, 2011), while people were also becoming more accustomed to participatory learning and entertainment experiences, and expected interactivity from cultural institutions, too (Simon, 2010).
This early focus on justification prompted subsequent attention on strategy and evaluation. Romero (2011) described how branding had been the primary goal of social media, but that institutions now increasingly sought a return on investment (ROI) from all their activities, including social media. As a result, institutions needed to conceive of social media differently, exploring deeper functions of social media beyond simply marketing: there was potential savings in terms of offering user services, and the informal tone of social media meant it was suited for introducing new interactive provisions. Furthermore, social changes in this time had an impact on academic discourse of social media: social media grew because it gave the users themselves more influence (Universal McCann, 2010, 2014). Consequently, discussions about whether and how to use social media evolved towards assessing its effectiveness. Because of the increased possibilities seen in social media, they became like any other institutional function that faced the “age-old challenges of demonstrating value and assessing efforts” (Colburn & Haines, 2012). Social media was now used out of both opportunity and obligation.
In addressing this shift in emphasis, recent literature has also reflected on the challenges in optimising social media performance. Simon (2010, 315) notes how social media are in “perpetual beta,” remaining a work-in-progress in order to respond to user behaviour. Information professionals have always planned for technology, but there exists little formal strategy for social media, because it is both new and changeable (Steiner, 2012). Institutions now believe that technology has changed audience expectations and put pressure on them to participate in social media, at a time when budgets and resources are limited (Thomson et al., 2013). Sometimes the challenge is not whether to use social media, but in deciding which other tasks have to be dropped to accommodate it (Kho, 2011). Information professionals have also experienced problems in adapting due to misconception of the ephemeral and reciprocal nature of social media, which contrasts with traditions of preservation (Solomon, 2011). A lack of understanding and experience extends to planning and evaluation because, without a clear understanding of the aims for social media, it is even more difficult to plan and evaluate appropriately (Steiner, 2012; Cadell, 2013).
Concurrently, there is also a lack of empirical examples of contemporary practice of social media evaluation tactics in cultural institutions; evidence tends towards brief overviews of particular initiatives, as was the case for the New York Public Library and University of Southern California, that highlight a few examples of best practice in social media strategy (Steiner, 2012), or with the Save Ohio Libraries campaign’s harnessing of “social capital” (Solomon, 2011). A recent development of more detailed analysis rather than anecdotal evidence is shown in Showers (2015), which features case studies specific to library analytics, including in Harvard Library and the British Library. This publication shows a more concerted effort to formalise evaluation, towards a “realization that the real power lies in our ability to collect, share and interrogate data at ever greater scales” (Showers, 2015, 20).
Across these texts is a shift from strategy being optional or advantageous, to being essential for ongoing management. Steiner (2012) asserts that social media strategy helps institutions be more proactive in using and monitoring available tools. Simon (2010, 16) argues that it is not enough to use social media just because visitors enjoy it, because it “trivialises the mission-relevance” of the interactions. Similarly, Solomon (2011) asserts that a lack of strategy means there is no clear picture of what success is, so no way to progress, while Stuart (2009) argues that, without a specific purpose for social media, it is impossible to construct useful metrics to justify or reappraise its activities.
Strategies help explain social media’s benefits, making it easier to justify evaluation; people do not like to be measured and need to know why it is needed (Stuart, 2014). Strategies are also essential because the key to long-term success is to know where user and institution objectives align (Universal McCann, 2014). Organisations need to match their social media to their users’ methods, and to identify this they require evidence.
Consequently, Web analytics tools is a growing topic. Social media is inherently intangible and difficult to measure; therefore, both qualitative and quantitative measures are needed (Fichter & Wisniewski, 2008). Analytics are a way of substituting the monetary factor of ROI calculations with a different type of revenue, such as number of visitors (Romero, 2011). While surveys can lack accuracy due to relying on self-reporting, analytics retain a degree of neutrality (Colburn & Haines, 2012). In a more recent development, Showers (2015) argues that metrics such as page views and likes are no longer enough, because these features are evolving. He advocates continual questioning of the motivation behind certain measurements, achieved by reviewing a wide range of metrics.
Inherent in this is a desire to define and measure impact. Thompson et al. (2013, 30) found that 56% of institutions say social media has a “major impact” on boosting public profile, while 53% say it improves their engagement with the public. Simon (2010, 172), meanwhile, states that social media’s greatest impact comes from its “spreadability.” Flores (2014, 197) describes impact in terms of influence, information, relationship, and trust—and interestingly found that social media is never among the preferred “touchpoints” for any of these impact dimensions. It is therefore evident that there is no single strategy or framework for measuring impact, because impact can be conceived in various ways. This stands to emphasise the importance of individual institutions planning and evaluating social media in order to ascertain their own conception of impact. This is where frameworks, which can help institutions do this, have begun to feature in the literature.
An early model features in Fichter and Wisniewski (2008, 55). They cite the “Trinity Approach” developed by Avinash Kaushik as a means of measuring success based on defining three factors: behavioural data, the outcome, and the experience. Respectively, this means institutions should collect the data they think is suitable, answer the questions of what this data means, and then use this to explain why this is worthwhile. The writers propose that the approach is a way of combining both quantitative and qualitative measurements through analytics, and thus help “chart our course”: social media can encourage, promote, innovate, learn, adapt, improve customer service, and discover and deliver what users want. While these motivations are not developed further, they are comparable to the “themes” that feature in Culture24’s subsequent framework.
Another proposed framework is the Impact Toolkit designed by the Chartered Institute of Library and Information Professionals (CILIP, 2015). This toolkit provides “courses” on six stages of demonstrating impact that first help identify objectives and then help demonstrate them through appropriate communications. The third course, on “impact and evidence,” uses the Evaluating Impact Model developed by Markless and Streatfield (2013), which prescribes two sides to evaluation: service performance measurements, which will be based on specific activities and more statistical data; and “impact on people,” which is more qualitatively based on criteria and collecting evidence. This toolkit is flexible in how it can be utilised, but maintains an emphasis on having a deliberate approach to social media and use of analytics to justify this.
A third framework has developed in Universal McCann’s annual Web behaviour research. In Wave 7 (Universal McCann, 2014, 37), they introduce the concept that “even the most superficial social media activity is driven by one or more ‘human needs’: learning, relationships, diversion, progression or recognition.” The conclusion is that people value organisations, or “brands,” that help them meet these needs, and therefore organisations should direct their social media accordingly. By Wave 8 (Universal McCann, 2015), this has developed to the more explicit recommendation that content is directed to specific objectives. There is a transition from simply identifying achievements, such as retweets, to considering the underlying purpose behind them. Universal McCann’s model is not formalised into a framework, but it reflects a strategic shift towards evaluating performance against audience motivations.
This study focused on Culture24’s “Let’s Get Real” project, which has focused on helping cultural institutions engage with audiences through digital platforms. Its second publication includes the Evaluation Framework, designed to help institutions define their objectives across six “themes,” each with corresponding targets and metrics. The framework helps institutions decide what to measure, so that they could judge the extent to which their users’ behaviour corresponded with their aims. This could then also be applied to setting targets for improvement—and prompt consideration of what the institution wants out of its social media. The six “themes” could be summarised as follows:
- Community: maximising the institution’s audience
- Brand: improving the institutions image
- Marketing and Communication: increasing attendance or use of the institution’s resources
- Interaction: having more frequent/qualitative interactions with the institution’s audience
- Content: creating resources through digital collaboration
- Visitor Services: providing a digital point of service for the institution’s audience
The framework indicates that institutions focus on the theme that corresponds to their social media’s purpose, and the framework then provides you with suggestions for measuring this objective. For instance, if your priority is to achieve a community on social media, then your target should be the size of your following, your metric will be the number of followers you have, and the tool you use is the relevant social media analytic application, such as Twitter Analytics. A poignant phrase in the study is that: “understanding what success and failure look like is the key to knowing if you have achieved either.”
In this respect, a primary application of the framework is to help facilitate this understanding. Relatedly, the framework gets institutions to “measure what they value, not value what they measure”: in essence, the framework is as much about deciding what matters as it is about evaluating particular statistics. Nevertheless, the challenge of prioritising one theme is therefore a feature of the Framework under scrutiny. Furthermore, there is ambiguity about whether the framework is more valuable as a step-by-step reporting tool or as something more flexible. These were considered in the current study, which explored how institutions might use such a tool.
Steiner (2012, 1) encapsulates the current challenges suggesting that, no longer the “gatekeepers” to information, cultural institutions are faced with having to create and evaluate their social media content, or else users will do this for them. However, participatory engagement often requires different outcomes and evaluative methods than those to which cultural institutions are accustomed (Simon, 2010).
This study focused on the need for more explicit guidance on how to evaluate and assess social media use. The main research question that emerged from the literature review is how cultural can institutions better understand social media and its evaluation. This subsequently produced three sub-objectives for the research:
- Explore the current understanding and practices of cultural institutions using and evaluating social media, particularly with regard to challenges and impediments to developing these practices
- Test an existing framework as a way of exploring these challenges; namely, compare the principles in the Culture24 Evaluation Framework to assess its efficacy and potential improvements
- Through this research, provide recommendations for best practice and practical solutions for cultural institutions looking to explore the use of frameworks
In fulfilling these objectives, it was hoped that the study would demonstrate the overall context of cultural institutions’ behaviours and outlook on the topic. As a byproduct of surveying cultural institutions, it would also be possible to collect quantitative details of the platforms, experience, and reach of cultural institutions, and potentially demonstrate whether there was correlation between the use of strategy and the level of success on social media.
In seeking to demonstrate contemporary context, the study sought both qualitative and quantitative evidence from cultural institutions. This paper will focus on the results obtained through an online survey aimed at capturing the experience of institutions from a range of sectors and locations. The survey was designed to complement separate interviews with eight Scottish cultural institutions. The interviews with local institutions explored their current culture of planning and measuring social media, and their reaction to the Culture24 framework. Survey questions followed a similar structure, covering current practices, perceptions of strategy and evaluation, and overall objectives.
On this last point, one survey question was a veiled testing of the Evaluation Framework. Interviews indicated that rather than focus on one theme, each institution considered all themes important to some extent. In the survey, we therefore sought to quantify the extent that each theme was important, on a five-point scale from “Not Important” to “Top Priority.” The results could then be used to show the importance of each theme and whether any were more commonly higher priority.
No limitations or targets were set on participant numbers, and recruitment sought to attract as many libraries, museums, archives, and similar institutions using different communication channels (personal contacts, JISC message boards, forums dedicated to social media). Culture24 were also engaged to utilise their connections with the relevant institutions.
More successful, however, was recruitment through a Twitter profile, @memoryorgstudy, that communicated directly with institutions. Tweets were used to advertise the survey and send direct invitations, while also encouraging institutions to share the survey with others via retweets. The rationale for Twitter advertising was identified during interviews, which indicated the ubiquity of Twitter in relevant institutions, as well as its nature as a source of networking amongst institutions. The account engaged museums, libraries, and archives, as well as interest groups, conferences, and high-profile personnel. Using Twitter also allowed a clearer picture of success rates: institutions could more easily indicate their response by retweeting, favouriting, or replying to the invitation.
Analysis was initially carried out on the results of each question individually. Secondly, filters were applied to explore comparative results by institution type and geographical location. Thirdly, each question was reviewed in more granular detail, though the application of filters. This was done for key features from each question, to gain insight into potential effects of certain criteria. Following initial analysis of the survey results as above, there was also comparative analysis against the interview results; both interviews and the online survey had a similar structure and topics, in particular when testing the Evaluation Framework’s themes and principles. This meant that interview testimony could be set against a wider context, and survey results could be compared to more empirical examples from interviews.
There were a total of 101 responses to the survey. Of these, 31 were incomplete, including 17 that only completed the first question. The drop-off rate after this was less pronounced, except for the final question, which saw five participants leave. There was a total of 70 completed question sets. 33 respondents (49%) were from the United Kingdom, and 17 respondents (25%) were from the United States. 29 respondents (43%) identified themselves as museums. 11 respondents (16%) identified themselves as libraries. 9 respondents (13%) identified themselves as archives. A further 25 respondents (37%) fell into other categories or could not be discerned: this included universities, heritage sites, arts and cultural organisations, and heritage projects.
Twitter is almost universally used by cultural institutions, with 96% of respondents using this media. Facebook follows at 84%, then YouTube (57%), and an official blog (47%). Perhaps most notable is that Instagram (45%) is now used almost as widely as blogging, while Pinterest and Flickr are other relatively novel platforms with significant usage.
The study found that institutions use between four and five platforms on average. Furthermore, the range indicates that institutions use a few primary platforms, most commonly Twitter and Facebook, and then a couple of supplementary platforms. It might also be concluded that there is a slight tendency towards visual or image-based platforms over text. Instagram, Pinterest, Flickr, and YouTube were more common than Google+, Tumblr, and wikis.
The second question asked the extent to which the institution thought strategically about social media, with no consistent pattern among responses. 40% of institutions have a social media strategy document, compared to 45% that do not. Social media policy documents were slightly more common, with 49% of institutions possessing this, compared to 38% that do not. Institutions’ use of specific targets and goals is also evenly split: 42% of respondents do not set targets compared to 39% that do. These results indicates there is a roughly even split between formal and informal approaches to social media.
The effect that having a strategy document has on other responses is also significant. The study found that, when an institution has a social media strategy, it is overwhelmingly more likely to have a policy, set specific goals, use analytics, and expect to spend more time on social media in future.
|Total||Have a social media policy||Set specific social media targets/ goals||Use analytic tools to monitor web/ social media||Expect to spend more time on social media|
|Yes||No||NA/ don’t know||Yes||No||NA/ don’t know||Yes||No||NA/ don’t know||Yes||No||NA/ don’t know|
|Have a social media strategy||34||29||4||1||18||9||7||32||2||0||25||1||8|
|Do not have a social media strategy||38||9||28||1||13||24||1||25||11||2||24||7||7|
|NA/Don’t know if have a social media policy||12||3||0||9||2||2||8||6||0||6||5||0||7|
Table 1: effect of strategy document on other strategic activities
There is a degree of consensus regarding both the use of analytic tools and plans to spend more time on social media in future. 75% of respondents use analytic tools, compared to 15% that do not. This indicates that analytics are becoming fairly standard. However, results point towards a variety of inconsistent approaches across respondents. While it is important to point out that these results do not reveal whether the impact of a strategy is positive or negative, they do perhaps indicate that a strategy provides more decisiveness regarding social media activities.
Question 3 asked the extent to which a variety of objectives based on the themes of the Culture24 Evaluation Framework were relevant to the institution. The first impression is that all themes are considered important to some extent. Only six respondents chose any of the themes, listed above, as “Not Important.”
By contrast, the most common response for five of the objectives was “Very Important”; only the creation of resources received more responses for “Important to some extent.” The weighting towards high importance validates each theme as appropriate, but also shows that no single theme is predominant. Relatedly, there is significance among results for “Top Priority” themes. Only 24 of the 77 respondents (31%) selected any of the themes as a top priority. Furthermore, 19 of the 24 who gave a “Top Priority” selected more than one theme as such. This indicates that, for a large majority of organisations, social media serves multiple purposes.
|#||Question||Not important||Slightly important||Important to some extent||Very important||Top priority||Total Responses||Mean|
|1||Maximize our audience||1||4||19||42||11||77||3.75|
|2||Improve our image and be more noticeable||0||2||9||47||19||77||4.08|
|3||Increase attendance, revenue, or use of our resources||3||7||23||31||13||77||3.57|
|4||Have more engaging conversations with our audience||0||5||18||37||17||77||3.86|
|5||Create resources and collaborate with others||2||16||28||21||10||77||3.27|
|6||Provide a digital point of service and improve our users’ experience||2||9||23||32||11||77||3.53|
Table 2: “To what extent are these objectives important for your organisation’s social media strategy?”
The overwhelming challenge to using and evaluating social media was time and resources issues, cited by 74% of respondents. Elsewhere there was an interesting discovery regarding audience engagement. Of the 26 institutions with problems engaging, only 8 indicated problems establishing an audience; likewise only 8 of the 24 institutions with problems establishing followers have problems engaging with them. This suggests that these issues do not go hand in hand.
In terms of strategy, 25% cited the setting of targets and objectives as a challenge, while 33% said defining appropriate metrics was challenging. However, as with audience issues, there was little correlation between these two problems. This suggests that some may be able to define their objectives, but do not know the metrics to achieve them; likewise, some organisations will know what it is they are counting, but do not know their purpose. This indicates that there is a clear market for the Evaluation Framework in terms of its function in helping organisations frame their intentions.
Results here suggest a lack of critical depth in many institutions, due to lower figures for metrics that indicate engagement. For instance, only 52% track the number of comments, only 33% track the bounce-rate from their site, and 32% track the use of their marketing messages. Furthermore, only 15% identify whether the sentiment behind comments is positive or negative, and only 13% follow the length of comments; this suggests that few organisations measure social media qualitatively. Perhaps the clearest indication of the superficiality of current measurements is that while 95% track their immediate followers, only 25% go to the next stage and track their potential audience reach.
|#||Answer||Responses (75 total)||%|
|1||Number of followers/friends||71||95%|
|2||Number of comments||39||52%|
|3||Number of re-posts (e.g., shares, likes, retweets)||57||76%|
|4||Sentiment (i.e., Positive/Negative)||11||15%|
|5||Follower demographics (e.g., age, gender)||19||25%|
|6||Use of marketing messages (e.g., hashtags, keywords, mentions)||24||32%|
|7||Reputation of followers (e.g., famous/influential followers)||13||17%|
|8||Potential audience reach (e.g., followers of followers)||19||25%|
|9||Bounce rate (i.e., how many website visitors arrive and then leave without exploring site further)||25||33%|
|10||Direction (i.e., where users have come from, such as from a Google search)||22||29%|
|11||Response rate (i.e., how long it takes for your institution to reply to a message on social media)||9||12%|
|14||Length of comments/ conversation threads||10||13%|
Table 3: “Which of the following do you currently track/measure?”
A further question sought the motivation behind collection of statistics. The most common frequency was monthly (40%), while measurement “when required” was almost as common (37%). In total, therefore, there were 20 of 75 institutions (27%) that did not have a defined interval for collecting statistics. Only 13 of 75 institutions (17%) currently document social media figures annually. This could be due to the relative novelty of social media or because it is possibly seen as redundant due to more frequent measurements. However, it also indicates lack of long-term analysis, perhaps consequential of lack of strategy. Overall, the metrics-based results indicate that contemporary cultural institutions remain in the early stages of social media development.
Results indicate that the use of analytics programmes is now widespread. As well as 85% of institutions using social media analytics, such as Facebook Insights and Twitter Analytics, 71% use Web analytics; 45 of the 75 respondents (60%) use both of these. However, when combined with the results of the previous questions, which indicate a lack of depth, it might be argued that while most use analytics tools, they do not fully utilise their capabilities.
Results showed that most institutions use multiple analytics tools, with a mean of just below three per institution. Facebook Insights (63%), Google Analytics (63%), and Twitter Analytics (56%) are the most common tools. Results also suggest that use of commercial analytics tools is more sporadic (e.g., Hootsuite, with 26% respondents currently using it). It is not clear whether this is due to lack of resources or if it signifies a lack of satisfaction with service; however, it does indicate that these platforms are waning in popularity.
It is perhaps worth comparing earlier results that indicated institutions use on average between four and five different social media, against an average of about three analytics tools, most likely to be Facebook, Twitter, and Google. This suggests the possibility that analytics tools for other social media platforms are not as well developed or are not considered as relevant by institutions.
Firstly, a majority of institutions confirmed that evaluation was important: 71% agreed that they have a culture of evaluation, compared to just 11% who disagreed. Furthermore, 83% indicated that that social media was vital to achieving institutional goals, as opposed to just 3% (2) who disagreed. However, while 43% indicated that evaluating social media was a priority, 27% indicated that it was not a priority. This indicates that, while social media might help achieve organisations’ objectives, this is not translated into prioritising assessment.
Tables 4 and 5 demonstrate how evaluating social media impacts other evaluative practices. Significantly, the less important evaluation of social media was, the less likely it was that evaluation was part of the organisation as a whole; that there were regimented approaches to social media and evaluation; that specific social media objectives were used; and that social media was vital to achieving organisations’ goals. The survey also found a clear preference for flexibility. By and large, institutions consider social media and its evaluation important, but want to pursue them in a flexible manner. These tendencies have implications for the potential to implement the Evaluation Framework.
|Strongly Disagree||Disagree||Neither Agree Nor Disagree||Agree||Strongly Agree|
|Other statements||Proportion who Agree||Proportion who Agree||Proportion who Agree||Proportion who Agree||Proportion who Agree|
|Evaluation important part of organisation||100%||90%||67%||39%||NA*|
|Regimented approach to social media||40%||20%||19%||11%||NA*|
|Flexible approach to social media||60%||85%||52%||78%||NA*|
|Regimented approach to evaluating social media||30%||25%||5%||11%||NA*|
|Flexible approach to evaluating social media||50%||70%||52%||44%||NA*|
|Organisation has specific objectives for social media||90%||70%||43%||17%||NA*|
|Social media is vital to achieve organisation’s goals||100%||85%||76%||78%||NA*|
Table 4: level of agreement that evaluating social media is not a priority to organisation, and corresponding proportion that agree with other statements
*NB: The figures for “Strongly Agree” have been disregarded, as the sample is too small. There is also doubt over the intention of the respondent, who may have been meaning to “strongly disagree,” based on responses elsewhere.
|Strongly Disagree||Disagree||Neither Agree Nor Disagree||Agree||Strongly Agree|
|Other statements||Proportion who Agree||Proportion who Agree||Proportion who Agree||Proportion who Agree||Proportion who Agree|
|Evaluation important part of organisation||78%||77%||50%||0%||NA|
|Evaluating social media is a priority||61%||37%||20%||50%||NA|
|Regimented approach to social media||26%||14%||10%||0%||NA|
|Flexible approach to social media||74%||69%||70%||50%||NA|
|Regimented approach to evaluating social media||22%||14%||10%||0%||NA|
|Flexible approach to evaluating social media||56%||54%||50%||50%||NA|
|Organisation has specific objectives for social media||70%||46%||30%||0%||NA|
Table 5: level of agreement that social media is vital to organisational goals, and corresponding proportion that agree with other statements
Results from North America showed more divergence from overall results than the UK results. Firstly, American institutions used more social media: most notably all American respondents used Facebook as well as Twitter; 60% have a blog compared to 47% overall; 60% use Instagram compared to 45% overall; and 75% use YouTube compared to 57% overall. Conclusively, American institutions averaged 5.9 social media platforms, compared to 4.68 overall. American institutions were also were more likely to set targets, use analytics, and note figures more regularly. These results indicate North American institutions are much more heavily involved in social media.
|Feature||N. America Proportion||Overall Proportion|
|Establishing an audience is a challenge||20%||32%|
|Engaging with an audience is a challenge||45%||34%|
|Setting targets and objectives is a challenge||15%||25%|
|Defining appropriate metrics is a challenge||45%||33%|
|Take note of metrics/figures weekly||50%||27%|
|Take note of metrics/figures monthly||50%||40%|
|Take note of metrics/figures annually||30%||17%|
|Set specific targets for social media||65%||39%|
|Use analytics tools to monitor social media||90%||75%|
|Use user survey to track metrics||30%||15%|
Table 6: main differences between North American and overall results
Furthermore, Table 7 also shows that North American institutions had notably different priorities regarding objectives. Scores for interacting with audiences, increasing use of resources, and providing digital services were all significantly increased, while the mean score for maximising the audience fell drastically. This perhaps indicates that North American institutions are more interested in marketing and engagements, rather than reaching as many people as possible. On the other hand, their perspectives on evaluation, their methods of tracking metrics, and the particular analytics tools they use are much the same as the overall results. This indicates that while they may use more platforms and have different objectives, the ways they assess social media are much the same.
|Strategic Objective||UK Mean Score||N. America Mean Score||Overall Mean Score||UK Ranking||N. America Ranking||Overall Ranking|
|Maximise our audience||3.85||3.60||3.75||2=||5||3|
|Improve our image and be more noticeable||4.00||4.00||4.08||1||2||1|
|Increase attendance, revenue or use of our resources||3.70||3.95||3.57||4||3||4|
|Have more engaging conversations with our audience||3.85||4.10||3.86||2=||1||2|
|Create resources and collaborate with others||3.24||3.15||3.27||6||6||6|
|Provide a digital point of service and improve user experience||3.55||3.75||3.53||5||4||5|
Table 7: mean scores and rankings for strategic objectives: UK, North America, and overall (mean scores on scale between 1 – Not Important and 5 – Top Priority)
Similar to results for North America, museums also showed slightly higher proportions actively using each social media platform, indicating a higher level of involvement. Museums use an average of 5.72 platforms per institution and also appear more likely to have defined strategies: 55% have a social media strategy document and set specific social media targets, while only 7% do not use analytics.
As Table 8 shows, museums also had slightly different strategic priorities. Their highest-ranked strategic theme was Interaction, and Marketing also had a substantively higher mean score. Overall, results for museums indicate that they may be slightly more involved in social media. Elsewhere, results suggested that archives are less interested in the size of their followings than museums or libraries. Also, while museums vary in their degree of strategic planning, libraries and archives generally prefer to be more flexible in their approach and focus more on evaluation and audience engagement than museums, which think more about image and attendance.
|Strategic Objective||Library Mean Score||Museum Mean Score||Archive Mean Score||Overall Mean Score||Library Ranking||Museum Ranking||Archive Ranking||Overall Ranking|
|Maximise audience (Community)||3.64||3.79||3.11||3.75||4=||4||5||3|
|Improve image (Branding)||4.27||4.00||3.56||4.08||1||2||1=||1|
|Increase attendance (Marketing)||4.00||3.83||3.56||3.57||2=||3||1=||4|
|More engaging conversations (Interaction)||4.00||4.03||3.33||3.86||2=||1||3||2|
|Create resources (Content)||3.45||3.34||2.78||3.27||6||6||6||6|
|Digital point of service (Visitor Serv)||3.64||3.69||3.22||3.53||4=||5||4||5|
Table 8: mean scores for strategic objectives: Libraries, museums, archives, and overall (scale between 1 – Not Important to 5 – Top Priority)
Based on responses, it was calculated that 43% of respondents indicated a culture of evaluation regarding social media, in expressing that this was a priority for their institution. Such institutions were more likely to have key strategic features, as shown in Table 9.
On one hand, institutions with a culture of evaluation use the same tools, in much the same ways, and towards similar objectives. However, the impact of prioritising evaluation is perhaps indicated in the challenges they indicate: most are reported at reduced rates. For instance, results suggests that those with a culture of evaluation have less problem engaging with audiences and have more staff buy-in. Overall, results suggest that prioritising evaluation of social media results in more developed or confident use of social media. Such institutions have more formal strategies and experience fewer challenges. Furthermore, they use more platforms and measure their impact more often.
|Feature||Culture of evaluation||Less culture of evaluation of social media||Overall|
|Have a social media strategy document||57%||37%||40%|
|Have a social media policy document||60%||42%||49%|
|Set specific social media targets/goals||67%||16%||39%|
|Use analytic tools to monitor activity||90%||58%||75%|
|Take note of metrics weekly||40%||0%||27%|
|Take note of metrics monthly||47%||32%||40%|
|Take note of metrics whenever possible (regularly)||13%||16%||21%|
|Take note of metrics whenever possible (irregularly)||3%||32%||19%|
|Establishing an audience is a challenge||27%||16%||32%|
|Engaging with an audience is a challenge||23%||37%
|Setting targets and objectives is a challenge||20%||37%||25%|
|Defining appropriate metrics is a challenge||40%||32%||33%|
|Staff cooperation or buy-in is a challenge||23%||42%||29%|
Table 9: key differences between those who prioritise evaluation and those who do not
Conversely, 27% of respondents might be said to have less culture of evaluation, having indicated that measuring social media is not a priority. In this case, the previous results are flipped: they are less likely to approach social media strategically, and they experience more challenges. Most notably, this means that just 16% set targets for social media, and only 58% use analytics—significantly lower than in overall results. Overall, institutions that have less social media strategy and evaluative culture show less appetite to improve. While many of these institutions may argue that this is purely because social media is not as important to them, it is worth considering the earlier statistics that indicated consensus that social media is growing in importance.
As mentioned, this survey was designed following a series of in-depth interviews with community managers in eight institutions. These interviews explored the variety of ways that cultural organisations use and measure their social media presence, and received explicit feedback on the feasibility and efficacy of the Evaluation Framework.
Among common themes in the interviews were notions that the most important thing in practice and evaluation of social media was that the tone of the output matched the organisation and suitable for the medium in use. Staff had to be engaged carrying out activities, and some platforms worked differently than others. Relatedly, institutions increasingly saw social media as not just a marketing tool, but also something that provided a more specific or functional service. The greatest impediment to social media was lack of resources, but this actually meant an increased focus on evaluation to justify time spent and increase efficiency. On the other hand, the institutions generally seemed to be resigned to sustaining current provisions rather than expanding.
Regarding the use of frameworks, interviewees were open-minded but expressed concerns over introducing tools to established procedures and limited resources. The Culture24 Evaluation Framework was appreciated by all those interviewed, but with the caveat that it would probably need to be adapted. First, there were some who wished for more detail on how to implement, while other felt that it was overly proscriptive. The themes were all agreed as relevant, but there was wariness over whether any single objective could be prioritised over others. There was also uncertainty over whether it would work better as a step-by-step measuring device or a more flexible guide for general strategic planning.
Overall, the interview results had more commonalities than differences with the survey results. The key points identified from the interviews were that success on social media often revolved around staff involvement; that many institutions had benefited from a strong evaluative structure; that many institutions were exploring different functionality on social media; and that institutions face difficulty staying on top of evaluation and taking social media to the next level. In this respect, findings in the survey help reinforce this: they showed a correlation between having more robust strategies and staff enthusiasm, and in achieving more detail and development on social media.
Primarily, the survey demonstrates that there is a clear market for strategic tools: a significant number of institutions need help defining statistics and metrics. There is also evidence that cultural institutions lack a consistent approach for reporting and would benefit from strategies and tools that help them concentrate and justify their efforts. In these respects, the Evaluation Framework fills a potential need. It could benefit institutions in giving a purpose to measurements and regular reporting, and could help offer definition to current practices.
The survey also shows that institutions show clear signs of experimentation. Technical issues and staff buy-in were not considered an obstacle by most respondents, and with interview testimony, there are indications that most organisations are prepared and accustomed to trying new things that might improve their performance. On the other hand, the survey also emphasised the suggestion during interviews that flexibility is important to the introduction of new tools, for institutions both with and without defined strategies. This has potential implications for the application of the framework. Half will want to adapt it to their embedded practices. Alternately, the other half likely have neither the structure on which to add the framework, nor the desire to introduce such a strict approach. In either case, the application of the framework is likely to be heavily adapted.
Overall, what the survey indicates is that, while many institutions would welcome a tool that fills gaps in their knowledge, there is less scope for each institution to implement this in the same way. The inconsistency of results shows a level of individuality within evaluative practices that no single framework can universally accommodate. On the other hand, the consistency in the challenges and objectives of institutions shows that a framework that comprehensively addresses these features would serve a real strategic purpose.
The survey results showed that the principles behind the Evaluation Framework were valid and there was a clear market for the tool. Evidence suggested that many cultural institutions are still in the developmental stages of planning their social media and implementing appropriate strategies, and that the strategic themes outlined in the framework were all highly relevant across diverse groups. Furthermore, there were clear indications that objective-setting could prove tricky for some, and that there was a lack of experience, confidence, and consistency in the use of appropriate metrics, analytics tools, and timeframes—all principles the toolkit helps implement.
However, the survey results—set against the context of interview testimony—also alluded to difficulties with the framework in terms of the practicalities of its implementation. Notably, institutions have reduced resources, various institutional pressures and procedures, and a widespread preference for flexibility. With the above strengths and weaknesses of the Evaluation Framework in mind, the following recommendations are suggested for developing and implementing the framework.
Be clear in how and why the framework should be used
Is it to measure your definition of performance or define your performance measures? The original intent is to help direct measurements of audience behaviour related to the institution’s social media goals. However, the interviews and survey results indicated that no single objective or “strategic theme” was priority. Decisions should be taken on whether the framework will be used as an ongoing procedure or as part of a periodic review; as a step-by-step process or as scaffolding for general planning.
Offer institutions templates and examples
A related problem with the framework was that, because institutions were likely to need to monitor more than one strategic theme, the work required became unrealistic. Furthermore, different institutions have different reporting priorities, frequencies, and personnel. More guidance is required on different methods for implementing the framework in a variety of circumstances: for instance, templates for individual projects, group projects, or periodic reviews.
Collate resources on metrics and tools
Survey results demonstrated a lack of critical depth in evaluation, despite institutions indicating desire to know the impact of their pursuits. Interviews provided context to this, saying that the metrics and tools were often quite difficult to interpret, and that for the sake of simplicity and speed they would stick to those most familiar. The Culture24 website reveals a variety of documents and links, such as the metrics toolkit, the tools comparison spreadsheet, and the Digital Change website. However, these are dispersed and disconnected from the framework. Combining these features would enrich the framework.
Overall, the survey results showed that cultural institutions do not have a consistent approach to strategy and evaluation, and that these concepts are still relatively unexplored by many. There are also indications that some are turning away from tools; interviews offered that one reason for this is that past the basics of analytics, there is little guidance and some institutions therefore seek simply to preserve, with little motivation to explore further developments. Evidence of other opportunities, and the impact they can have, would help reinforce the value of implementing an additional tool like the Evaluation Framework.
Survey results indicated that, with institutions typically using several platforms and analytic tools, the topic of social media evaluation is progressing. On the other hand, results showed that this development is still in early stages: there are even splits between those with and without formal practices, and there is a predominance of mainly superficial measurements. There is a degree of individuality in terms of reporting strategies and an overwhelming preference for flexibility. Subsequently, the following recommendations are made for institutions looking to develop their social media.
Consider the purpose of your social media, and build it into your content and strategy
The literature review noted that many cultural institutions jumped into social media without thinking about how they would use it. Consequently, they faced difficulty in defining their impact and where they could develop. On the evidence of both interviews and survey, institutions are now beginning to think more strategically. However, framing social media was a novel idea, and few institutions indicated that they had specific priorities or objectives. However, there were signs that more strategic institutions were more likely to be active in more areas and more decisive in identifying what worked for them, and would face less challenges.
Utilise the strengths and functionality of each resource
The survey results showed that institutions typically stick to the tools, features, and metrics with which they are most familiar, possibly due to a lack of experience and understanding of technical aspects. However, there is also growing evidence of experimentation with different tools and platforms. Testimony from interviews suggested this was because institutions were now looking for social media platforms to perform more functional roles, like cataloguing or visitor services, and that such diversification was giving them increased confidence in technical aspects. Institutions should think about what each platform offers them, how this contributes to their purpose, and how this can be incorporated into evaluation.
Evaluation is measurement with a purpose
Evidence suggests that some institutions collect statistics without a purpose, similar to getting involved on social media without a clear intention. While many institutions made measurements and evaluated social media, this did not correspond with a specific strategy or plan. This perhaps explains why many do not develop more in-depth analyses or technologies. Furthermore, evidence showed that institutions are often restricted in how they report, sometimes only doing it when time permits. Having a clear and regular structure for measurements can help institutions stay on top of things, so that when evaluation does take place, it can focus on what matters.
Share the load, share the knowledge
The survey demonstrated that resource issues were the top impediment to developing social media, while there were other indications throughout the research that there was also a lack of understanding or guidance. Those that spent more time on social media were more likely to have plans and policies and to benefit from the increased structure and collaboration that these encourage.
Work with what you enjoy, and enjoy what works
It was clear that every institution has different approaches and universality is elusive. For most survey respondents, flexibility was preferable in social media; but some of these also sought more defined evaluative practices. Social media is an ongoing experiment, and not everything will always succeed, but what was learned from interviews was that success was built on staff doing what they love. The research indicated that although evaluation of social media can identify success, it should not end up taking away the fun of social media that helped achieve that success. To do so, institutions, perhaps with the help of frameworks or toolkits, need to find the right evaluative formula that suits the institution, its personnel, and its audience: the “voice” of the institution.
Overall, the research helped validate the place of social media evaluation and tools that facilitate this. However, it also demonstrated that this need cannot be met in straightforward or universal ways: each institution works differently under its own set of constraints and expectations. Institutions need something flexible enough to adapt to their needs, but also something that has clear justification and proven efficacy, to justify the time and effort.
The research noted with interest how commonly institutions refer to maintaining the “voice” of the institution or allusion to remaining true to their overall mission and values. Social media is something that must complement this, and is therefore unique to every institution—and therefore there is necessarily no single method, objective, or framework for how to approach social media. Despite this, this study has shown that it does not mean that having some sort of theoretical or strategic outlook for social media is undoubtedly beneficial in providing the structure, decisiveness, and confidence that allow the institution to continually develop this individual voice. Strategy and evaluation do not constrain the voice of the institution; they are part of how the institution finds its voice.
Such theoretical basis, however, is predicated on access to practical guidelines and evidence such as relevant, feasible frameworks. Ultimately, cultural institutions need more guidance in this topic, with access to a range of options from which they can select, adapt, and draw inspiration. This research has validated the principles of the Culture24 Evaluation Framework as a good starting point and provided recommendations for both this framework and for institutions themselves, to continue to develop evaluative best practices—and thereby improve performance on social media.
The research described in this paper was conducted within the EU-funded Material Encounters with Digital Cultural Heritage (meSch) project, ICT Call 9: FP7-ICT-2011-9, GA 600851.
The authors would like to thank all institutions who took part in the survey, and the following institutions who were interview participants:
- Glasgow Museums
- Glasgow University Archives Service
- Glasgow Women’s Library
- Lost Glasgow
- National Library of Scotland
- Royal College of Physicians and Surgeons Glasgow Library and Archives
- Scottish Political Archive
- Scottish Screen Archive
- University of Stirling Art Collection
Cadell, L. (2013). “Socially practical or practically unsociable? A study into social media policy experiences in Queensland cultural heritage institutions.” Australian Academic & Research Libraries 44(1): 3–13.
Cahill, K. (2011). “Going social at Vancouver Public Library: what the virtual branch did next.” Program: electronic library and information systems 45(3): 259–278.
CILIP. (2015a). “Impact Toolkit” [website]. Chartered Institute of Library and Information Professionals. Last accessed January 31, 2016. Available http://vle.cilip.org.uk/course/index.php?categoryid=30
Colburn, S., & L. Haines. (2012). “Measuring libraries’ use of YouTube as a promotional tool: An exploratory study and proposed best practices.” Journal of Web Librarianship 6(1): 5–31.
Culture24. (2014). “Let’s Get Real 2: A journey towards understanding and measuring digital engagement.” Last accessed January 31, 2016. Available http://weareculture24.org.uk/projects/action-research/phase-2-digital-engagement/
Fernandez, J. (2009). “A SWOT Analysis for social media in libraries.” Online 33(5): 35–37.
Fichter, D., & J. Wisniewski. (2008). “Social media metrics: making the case for making the effort.” Online 32(6): 54–57.
Flores, L. (2014). How to Measure Digital Marketing: Metrics for Assessing Impact and Designing Success. London: Palgrave Macmillan.
Kho, N.D. (2011). “Social Media in Libraries: Keys to Deeper Engagement.” Information Today 28(6): 1, 31–32.
Markless, S., & D. Streatfield. (2013). Evaluating the Impact of Your Library (Second Edition). London: Facet Publishing.
Romero, N.L. (2011). “ROI. Measuring the social media return on investment in a library.” The Bottom Line: Managing Library Finances 24(2): 145–152.
Showers, B. (ed.) (2015). Library Analytics and Metrics: Using data to drive decisions and services. London: Facet Publishing.
Simon, N. (2010). The Participatory Museum. Santa Cruz, CA: Museum.
Solomon, L. (2011). Doing Social Media So It Matters: A Librarian’s Guide. Chicago, IL: American Library Association.
Steiner, S.K. (2012). Strategic Planning for Social Media in Libraries. London: Facet Publishing.
Stuart, D. (2009). “Social media metrics.” Online 33(6): 22–24.
Stuart, D. (2014). Web Metrics for Library and Information Professionals. London: Facet Publishing.
Thomson, K., K. Purcel, & L. Rainie. (2013). “Arts organisations and digital technologies.” Last accessed January 31, 2016. Available http://www.pewinternet.org/files/old-media//Files/Reports/2013/PIP_ArtsandTechnology_PDF.pdf
Universal McCann. (2010). “Wave 5: The Socialisation of Brands (Social media tracker – 2010)” [presentation]. Last accessed January 31, 2016. Available http://www.slideshare.net/tlinz8612/wave-5-the-socialisation-of-brands-report-5465006
Universal McCann. (2014). “Wave 7: Cracking the Social Code – The Story of Why” [presentation]. Last accessed January 31, 2016. Available http://www.slideshare.net/tlinz8612/wave-7-crackingthesocialcode?related=1
Universal McCann. (2015). “Wave 8: The Language of Content.” Last Accessed January 28, 2016. Available http://wave.umww.com/assets/pdf/wave_8-the-language-of-content.pdf
Fleming, Rod and Areti Damala. "Finding the voice: Planning and evaluation of social media in cultural institutions." MW2016: Museums and the Web 2016. Published January 31, 2016. Consulted .