Facebook, Facebook, Facebook. Murdoch Stevens details how to use and get around the Zuckerberg Empire in his attempts to spread public awareness about increasing the quota of refugees into New Zealand.

I was curious about what it would actually take to get the refugee quota increased. I knew I could draw on friends across the arts, academic and activist communities. I hoped to be able to count on them to share a post or two on social media and spread news of the campaign through their networks. But how would it escalate from there?

I spent a weekend syncing email addresses with Facebook pages, and invited enough people to be Facebook friends that my connections doubled and I was banned from sending more invites for a week. My girlfriend at the time sent me a horrified message that her Facebook live feed was full of notifications that I had become friends with someone, including – to her shock – the finance minister Bill English. While I had met some of the politicians who I friend requested, my real aim was for our content to appear in their newsfeeds and notifications.

I wrote a long and detailed group email, as well as numerous personal messages, to ask friends if they would sign up to be a supporter or amplifier of the campaign. Those roles were for people who were keen to receive an email from us “just five times” in the coming campaign, when we wanted them to use their social media networks to spread a very important message. I specifically asked for help on five occasions to show that this was to be a short-lived campaign with a very explicit ask, rather than some amorphous, long-term commitment. From those who responded, we created our first email list, which included a few “super helpers” who were keen to be as involved as they could. While very close friends signed up to help due to friendship and loyalty, others who I might not have expected to be interested put their names forward because they cared about the issue.

Close friends could help launch the campaign, but that would only buoy us for the first few months.

Beyond my personal page, I also created a “Doing our bit – Double NZ’s refugee quota” Facebook page. A first look at the analytics and targeted advertising available to these pages can be a shock for people who haven’t seen the kinds of tools available to reach an audience. These pages are where Facebook’s advertising revenue comes from, and they offer a shortcut for individuals to quickly understand that while the site is free to use, there is a product being sold: the attention of regular users. Facebook’s newsfeed functions just like most digital news sources, prioritising content that gets people to click, or in any way interact with it.

Most people know that popular posts – as measured by all manner of interaction from clicks, shares, comments and likes – get shown to more people by Facebook. A picture of yourself sharing positive news will generally get more engagement than sad news. The focus on positive stories is a difficulty for campaigns that are highlighting social injustices. Most stories about refugee issues are distressing: for every human-interest story about successful community projects, there are a dozen reports of war, drownings and horrific persecution. It is only the most egregious of these reports that incense people enough to share this kind of story. Facebook’s addition of multiple reaction buttons such as angry, sad, and love has helped balance this lopsidedness, but positive stories still receive far and away the most attention.

While a few Facebook pages catch a public mood and grow both organically and rapidly, most languish unless they pay for promotion. Back in the earliest stages of the campaign, Facebook users still expected their pages – rather than personal profiles – to be seen by most people who liked those pages. That ‘organic’ audience continues to atrophy. Spending money is necessary to achieving a good reach for posts, though getting great reach requires a good post as well as advertising. We experimented with a range of tones and posts, measuring analytics of what brought in different kinds of engagement, new followers and more reach. Just when we thought we were getting the hang of mastering the algorithm, Facebook would change it. Years later, when the Cambridge Analytica scandal erupted, we saw just how far other organisations with far deeper pockets had gone in collecting and analysing user data to target very different political messages.

We had all sorts of little techniques developed to get around Facebook’s rules: I would Photoshop a filter over text-heavy images so that they would bypass the Facebook rule that you couldn’t post an image with more than 20 per cent text. Handwritten text was usually able to bypass the same monitoring. Before the 2016 US election, pages also had the ability to change the text and picture of a link that they shared. While many newspapers knew this, and were able to write headlines and use pictures that would perform well on social media, others did not. When those sources “buried the lede” or used a stock image one too many times, we would rewrite the headline, or substitute a more suitable image cadged from elsewhere on the web. But the real strength of the page lay in the singular nature of the campaign and the lack of any ulterior motive – even one as common as fundraising.

Online articles from Stuff and the New Zealand Herald, as well as a few cartoons drawn by Hayden Currie, formed the bulk of our requests to supporters to share content across Facebook. Each time this happened our number of supporters grew, especially when friends included kind words showing how much they were personally invested in the campaign. This organic reach, combined with advertising, boosted the number of people who followed our page, and surpassed our goal.

Extract taken from  Doing Our Bit: The Campaign to Double the Refugee Quota by Murdoch Stephens (Bridget Williams Books, $14.99), available from Unity Books.