In her first #DigiLab blog, Digital Lab Fellow Rachael Williams explains how she’s kept things simple (but pleasantly effective) for her first “scrappy” experiment.
Inspired after webinar number one with the fantastic Rachel Grossman from dog & pony dc and fresh from the first chat with my amazing mentor Sara Devine from the Brooklyn Museum, I wanted to get my first little experiment into action pretty swiftly.
I’ve been working on a social media campaign which highlights what the British Library calls its Treasures Collection — as well as showstoppers like Magna Carta and Jane Austen’s writing desk — these also include the hidden gems; the unexpected treasures perhaps less well known but certainly no less spectacular.
The aim of the campaign? To keep an always-on message present within our social media so that a) people know that we have a free gallery to visit, and b) to drive traffic to our Treasures microsite. Handily, the campaign also ties in with the airing of season two of Treasures Of The British Library on Sky Arts — a programme which sees famous faces pick six ‘treasures’ to capture their life, career and passions.
To start, the social media posts designed drum up a buzz around the #BLTreasures in the series included a combination of glossy portrait images, animated GIFs, trailers and the mentioning of the breadth of the collection. But they just didn’t seem to be capturing our audiences’ imagination as much as we’d hoped.
The copy in the first couple of posts highlighting the TV series and the accompanying website articles took a more generic angle. The first post tried to weave in the many areas presenter Fiona Bruce explored within the collection (‘From Charlotte Bronte to a cake fit for a queen’…), which I thought would have a broader appeal. While the Hanif Kureishi promo perhaps assumed too much prior knowledge about the collection item and presented the content in a rather ‘flat’ way — in hindsight playing on why this manuscript is so important to Hanif, and the significance of E R Braithwaite’s handwritten revisions, would have possibly captured people’s attention much more.
Earlier social media posts promoting #BLTreasures and the Sky Arts series.
The numbers on these posts were healthy (the Facebook post here received 28k+ impressions and 40+ link clicks, while the tweet received 27k+ impressions and over 300 engagements — but ideally we’d like to see at least 100 likes on a tweet/Facebook post). However there was room for improvement and in terms of engagements and we wanted to drive more comments.
Time for a slight change in approach to try and unearth the ‘why’; why should people care?
At the mid-point way in the series, I crafted posts which focused on one showstopper picked by actor Andrew Scott, rather than use the approach of previous posts which tried to tell the whole journey of each celebrity. It seems that drilling down into one chapter rather than attempting to give a whistle-stop tour of the entire story, was a good idea. These posts picked one item of Scott’s to focus on, explained why he’d chosen it and why it was so relevant to him personally, and told a rather intriguing story that explained why this item was so unique. Each celebrity picked such a wonderful array of interesting items with captivating tales behind them, it was tricky to select just one to highlight, but it was a little risk worth taking to see if people’s attention and imagination could be better captured.
Facebook and Instagram posts using more focused content approach.
It may have been the dream team combination of Shakespeare and Andrew Scott, or maybe it was telling people something they didn’t already know; finding that little hook which intrigued and triggered an emotional response in our audience. (Perhaps it was a lucky combination of all of these things — it’s always hard to pinpoint one single reason for any moment of success on social media).
But these posts were received very well by our audience, our second-best performing post of the month in terms of engagement and comments on Facebook, and over 7000 likes and 50 comments on Instagram (our top-performing post of 2018 so far). Deepening engagement with users is something we’re trying to improve — creating posts which drive emotional interactions and comments — so this was a massive step in the right direction. We also saw a nice spike in traffic to the microsite web page.
Snapshot of Instagram account showing difference in reception to posts.
Selection of positive comments from audience across social media channels.
The posts were sent at similar times on a Wednesday or Thursday morning, to the same audiences across Facebook, Twitter and Instagram. It seems that the more specific angle of picking one collection item to focus on with an intriguing and perhaps unknown backstory, rather than giving an overview of an array of collection items, really helped the later posts get picked up and stopped our audience in their tracks while scrolling through their busy feeds.
Was it pot luck or are we closer to better understanding what makes our audience tick? Bring on Scrappy Experiment Number One, Part II*.
*Cool snappy name still a work in progress.
Scrappy Experiment Number One, Part II
Surprised by a leap in response to Part I of the experiment on social, I wanted to try a little A/B testing with our email audience. I created an email that gave a roundup of the TV series and directed people to the new content on our website where they could discover digitised collection items and the stories behind them.
Would the famous faces with an overview of their individual journeys attract more clicks, or would a more single-treasure-focused approach work best as it did on social? We sent two emails with slightly differing narratives and content to test the click-throughs and drive collection items consulted on the website. Would our email and social audiences respond differently? Would a less-traditional approach persuade or put off our audience?
Version 1 (single-treasure-focused) left, Version 2 (overview approach) right.
The whole email list was made up from people signed up to receiving marketing with a ‘What’s On’ preference from us. They like to hear about news from the Library, particularly around events, galleries and exhibitions. To create the two separate lists, this main list was simply cut into two. As a result, the lists were of equal numbers and included people with the same opt-ins. They would be used to receiving emails from us telling them about our latest displays, how they can get involved in the Library, and culturally exciting additions to our website. Both email lists also received the same subject line, it’s only when they opened the email they would see different content. The serving of each email to each list was done randomly, but each version was delivered to the same number of people and audiences of equal ‘warmth’ to our brand.
Cue drum roll…
Okay, a huge celebratory fanfare isn’t quite needed for this. But still there were lessons to be learned.
Somewhat disappointingly there was no clear statistical winner from the testing in terms of opens (25.2% for version 1 vs 25.0% for version 2) and click-throughs (8.6% for version 1 and 8.0% for version 2). Interestingly version 2 (the celebrity/overview-approach version) did best on revenue, with version 1 (collection item version) doing much better when it came to collection items consulted (30% higher rate of consultation). Did the glossy celebrity shots appeal to spenders, while the manuscript images appealed to those who want to delve deeper into the collection? As the lists would have included some people opted into retail marketing and those interested in research and news about our microsites it makes sense that these journeys took place from the email. I think it could be a safe conclusion to suggest that the collection-item-led approach did whet people’s appetite for pursuing the items further (as the stats confirm).
This little experiment has given me a taster of the small iterations and gradual changes that can be made to better understand our audience, what they want to see, how they want content delivered and how they want to digest it.
Maybe I’ll even be a little bit bolder with my next scrappy experiment…
Rachael Williams is Content and Community Officer at the British Library.
Images courtesy of the British Library ©.