Turn on, tune in, tweet out – experiments in engagement

This time of year is often one of reflection – what went well, what could be improved and so on. In this spirit we are putting up here an assessment of the livestreaming aspect of our outreach programme over the past couple of years.

This blog asks what was successful? What flopped? Where did we get bang for buck? Read on and find out…

Lofty goals

The OSC works towards collaborative engagement with the research community and relevant stakeholders – amongst other things, this helps us to communicate policies, promote our services and identify needs and knowledge gaps within the communities we work with.  

It will come as no surprise, therefore, that the words ‘open’ and ‘transparent’ crop up frequently when we are planning our communications. In the context of events and outreach, we usually start from the position of wanting to invite as many guests to the table as possible – not just those within the University but across the whole scholarly communication community. Given the international span of this group of people, one obvious solution is to take the party on the road – virtually speaking, at least.  

Starting in October 2016 with the ambition to make the most of technological solutions to achieve this (whilst taking into account the limits of our A/V expertise and resources), we experimented over 18 months with various platforms and approaches in order to live-stream, record and share footage from the events we hosted. Evaluating the returns on these efforts has led to some useful lessons: whilst we’d like to share our events as widely as possible, we have had to make some strategic choices to make the venture worthwhile.  

In terms of evaluating the impact of online sharing, we acknowledge that social media marketing is one small part of our Communications remit – whilst the scope for digging down into statistics on YouTube and Twitter engagement is almost unlimited, the time available to devote to this activity is not.

Stepping into the stream: live broadcasting and video recording 

Livestreaming allows viewers to remotely attend events, and we hoped to find a method of broadcasting that would adequately capture all sound and visuals (including slide presentations) whilst allowing viewers to simultaneously contribute their questions and comments. We found these goals something of a challenge!  

1. Adobe Connect

When organising one-day workshops, we initially managed the streaming, recording and processing ourselves using the Adobe Connect package (which came with the advantage that we could use the University’s subscription without any additional costs for us).

However, this method required a stable connection to the wired Local Area Network (LAN), plus high-intensity input from our team members, neither of which were factors that could always be guaranteed – many of the University’s lecture rooms are in old buildings with minimal A/V infrastructure at best, and it was not always possible to plug into sound systems or connect to the ethernet.

After the events, we made recordings of the live-stream available via our YouTube channel, despite some of them falling short of our expectations in terms of sound quality and uninterrupted broadcasting. We concluded that whilst Adobe Connect was excellent for hosting webinars in a controlled environment (where the room was quiet and we were familiar with the available technological capacity) it was not suitable for livestreaming large events. 

2. Calling in the professionals

We took a different approach when organising higher profile events such as the Engaging Researchers in Good Data Management event in November 2017, hiring an external company to take care of both the livestream and video recordings. The difference in quality was remarkable – of course, you get what you pay for!

We also trialled the approach of making video recordings of one-day workshops, without live-streaming. Hiring professional recording equipment from the University Information Service to do this and having a quick in-house tutorial on how to use it again required high-intensity input from our team, although it produced higher quality results than filming through Adobe Connect  

Was it worth it? 

After 18 months of trying out these different methods, we needed to establish if the investment of time and money was reaping rewards, particularly given that the hire of professional equipment and services accounted for the largest single expense for an event. We needed to decide how much priority to give recording and streaming events for sharing in our Communications Strategy.  

A summary of the statistics showed: 

  • maximum livestream engagement reached 50 participants (for the Engaging Researchers event)  
  • engagement with our content on YouTube at the time of dissemination (through advertising in our newsletters, emails and Twitter accounts) varied from ten clicks to 600 clicks. 

The engagement statistics at the time of the event were moderate, and the audience for the livestream did not exceed the audience in the room. We therefore concluded that we would reserve the option of livestreaming for events where sharing on-the-spot footage was of significant benefit to the wider scholarly communication and research data community – for instance for high-profile conferences or politically urgent discussions.

We would continue to hire professional AV services to video ‘headline’ events that were of interest to the community, but would not make recording standard practice for every event.  

A last experiment… 

We realised there was another aspect to this question: after the initial promotion of the recordings, they sat dormant in the YouTube playlist and embedded on our websites, relying on users discovering them by serendipity. We needed to think about continuing to maximise returns on the investment.   

In order to address this additional concern, between March and June 2018, we used our twitter accounts @CamOpenData and @CamOpenAccess to re-promote 33 and 54 videos respectively. We monitored the viewings on YouTube and looked at various metrics in the Twitter analytics. 

During that period we saw an average 16% increase in the YouTube video clicks, with some videos attracting far more attention than others. These viewing figures were less than we had anticipated, and there were various hypotheses as to why: 

  • We were re-promoting the videos to an audience that may well have seen the videos the first time around, so were not offering anything new. 
  • Some videos were specialised in subject and therefore appealed to a limited audience. 
  • Some videos were lengthy and likely to hold the attention only of the most dedicated viewers. 

It was notable that the professional videos we’d commissioned performed better in YouTube as well as in Twitter in terms of engagement rates and impressions. Perhaps due to our confidence in the quality of these videos, we invested extra time in promoting them (for instance by adding images to our tweets), and engagement was indeed higher. However, the fact that many of these videos were short recordings of single presentations may also have added to their relative appeal. 

What we learnt 

There were lots of positive outcomes of this final experiment. The re-promotion campaign helped to maintain the presence of our brand on Twitter and YouTube, resulting in almost 600 clicks on existing YouTube content over three months. It added diversity to the content of our tweets and increased tweet impressions as a whole. It contributed to our strategic aim to disseminate professional knowledge, maintained contact with our community, and influenced the acquisition of new followers. 

In addition, we observed the most popular themes amongst our Twitter followers:  

  • Open access monograph publishing  
  • How to spot a predatory publisher 
  • Peer review and the benefits of openness 
  • Copyright 
  • Text & Data Mining 
  • Data management needs for different disciplines and different institutions 
  • Standard practices for managing and sharing code 
  • How to make data publications first class research outputs 

These are insights that will inform our planning for future engagement activity. 

Looking ahead 

Our re-promotion experiment has given us a handy list of priorities that will allow us to keep using our film resources even when staff time is scarce, and will inform our event planning from the outset if we know we want to record the occasion.  Our top take-away tips: 

1. Less but better – Resources are limited: livestream important events only but don’t compromise on quality.  Short videos are better received: take into account the length of talks, panel discussions and workshops. Can longer talks be naturally broken into shorter segments? 

2. Specific, practical, catchy – Take time to create engaging and specific titles for videos, and emphasise their practical focus, for example by starting with “How to”. These items are instantly more appealing to the browsing viewer, and also appear higher on search rankings when the subject is Googled.    

 3. Re-use and repurpose – Use short clips from older videos on social media when their content complements news or trends. Routinely reference videos when writing content, for example blogs or training slides. 

Want to know more? 

You can explore these recordings of past events on the OSC’s website, and subscribe to our YouTube Channel 

For an alternative perspective on using video to engage with the research and scholarly communications communities, join our Research Skills Support Coordinator, Claire Sewell, with an expert panel for the MmIT webinar, Using Video in your library and information service2pm Wednesday 12 December, and look out for her upcoming blog on preparing online training.   

 Published 10 December 2018
Written by Hannah Haines and Maria Angelaki
Creative Commons License

One thought on “Turn on, tune in, tweet out – experiments in engagement

Leave a Reply

Your email address will not be published, but will be held by us together with your comment. If you would prefer us not to hold this information please do not submit a comment. For more information about how we handle your personal information, please see http://www.lib.cam.ac.uk/privacy-policy

This site uses Akismet to reduce spam. Learn how your comment data is processed.