Social Media Advertising

Do Facebook page likes matter? Probably more than you think

What matters most on Facebook? Is it page likes? Post engagement? Reach? Impressions? It's one of the hottest debate topics in social media, especially in the political field. The truth is that how much each metric matters largely depends on what you're trying to achieve with your Facebook page. Though it's worth keeping in mind that the publicly visible statistics of page aren't the be-all and end-all of how a page might be performing, but they're a useful starting point.

While some digital consultants might be quick to dismiss page likes as a "vanity metric", I'd be just as quick to dismiss them as having just demonstrated that they have no idea what they're talking about. Page likes are useful for a few reasons:

  1. Organic reach: The more page likes you have, the more people are likely to see and engage with your content without you having to spend any money to promote it, and with funding for social media always competing with other  
  2. Viral-multiplier: Other than ensuring a decent background level of visibility for your page, a page with more page likes is far more likely to have a piece of content go viral organically thanks to there being more eyeballs on it.
  3. Community support: Having more page likes means you have a larger community of supporters who are ready to engage with your content, whether that's reacting, commenting, or sharing it. The largely that community is, the more validation it gives to others who are part of that community.

As a rule of thumb I've usually found that page likes usually translate to a page reaching around two to three times that number of unique people each week - and that's without advertising money. So an active page with around 200,000 likes will reach between 400,000 and 600,000 people on a standard week. That's a pretty useful multiplier to get your content out there. And one of the basics of advertising is to get as many eyeballs as possible looking at your ads.

Other than understanding how large your potential organic audience could be, the more page likes you have, the more statistically relevant the demographic, geographic, and interests information you can pull from Facebook's analytics tool will be. Having a better understanding of who your core audience is, and then being able to break that down into more targeted information for advertising latter on, is invaluable to putting relevant content in front of people.

Which brings us to reach and impressions - essentially two sides of the same coin. It's useful to define both of these as they're often misused by people (usually to inflate their own success). Reach should refer to the number of unique users who have seen a piece of content from your page. Impressions should be the total number of times content has been seen from your page (e.g. often the same person seeing multiple pieces of content, or the same piece of content multiple times). This allows you to divide impressions by reach and understand just how often people are likely to be seeing content from you. Reach and impressions are useful for comparing social media with traditional advertising mediums, say TV spots or newspaper ads as you're able to use similar metrics to judge success.

For example, you could run an ad asking for donations simultaneously on social media and in print and TV advertising, and on each channel you could use a slightly different URL. Using that unique URL, as well as knowing how many people were likely to have seen the ad in each channel, you're then able to understand which was the most cost effective in driving donations (hint - it's almost always going to be social).

Like TV and newspaper ads though, reach and impressions don't necessarily mean that someone has actually noticed your content. Just as TV and newspaper metrics assume that because someone has seen a show or picked up a newspaper, and thus seen your ad, Facebook, and indeed other social media channels, assume that someone flicking through their newsfeed and seeing your content has actually noticed it. Often it'll be something they've shot straight past while scrolling down their feed without giving it a second thought. If you ever wonder why pages keep experimenting with differing video and graphic dimensions - it's because they're trying to maximise the amount of real estate they take up on your screen while you're browsing.

This is where engagement metrics come into play, as you can usually assume that someone engaging with a piece of content - whether it's clicking on a post, reacting to it, commenting, or sharing it, means that they A) Have genuinely seen it, and B) It actually held their attention enough for them to do something with it. While to some extent higher engagement rates do drive your post to have greater reach in Facebook's algorithms (though I've seen pieces of content with fantastic engagement reach tiny audiences relative to page size) at the very least it gives you the ability to understand see what content is attention grabbing enough that people stop and look at it.

As I said at the start, these publicly available metrics aren't the full story. Much of what goes on with social media advertising, especially Facebook, isn't visible. The notion of "dark posts" for instance (while the name is a bit melodramatic for me - it's like claiming a newspaper advert put in one local paper, but not another, or an ad put in the sports section but not the news section, is a "dark ad") is one area where a page might look relatively quiet from the general public's view, but in actuality there could be a whole host of activity taking place from paid advertising campaigns that aren't readily apparent.

This was largely the case in the United States 2016 election where the day-to-day content you saw posted on the main candidates' pages was only a fraction of their online activity. Even though, for instance, Hillary Clinton may have been posting upwards of six to eight times a day, she would also have dozens, if not hundreds of supporting ad campaigns running in the background with targeted messages for specific voting groups.

There's no doubt a similar approach is being used by all the parties in this election, though given our more restrictive election spending rules, much smaller size, and less advanced Facebook advertising tools, we're unlikely to see the same level of sophistication here.

In the end, Facebook page likes do matter, but a lack of page likes can be circumvented by spending money on advertising. They're not the only page metric you should pay attention to, but they're an easy top-line indicator of a page's health and potential audience reach. Page likes, and page like growth, are also two of the most popular metrics to appear in media reports on political social media, so it doesn't hurt to have both these figures looking healthy.

Mobilise - The end game of getting out the votes

At the time of writing we're five weeks away from the start of advance voting and four weeks from overseas voting, so it's a timely opportunity to look at the end game of political campaign communications - mobilising your supporters and voters to turn up and cast their ballots for you. In political circles this is known as GOTV - Get out the vote.

Mobilise communications fall into two specific activities. The first is Enrolment. As a political campaign you want to make sure your supporters are correctly enrolled to vote, so there's a huge premium on political parties to push this. Given voter behaviour and demographics are the way they are, this has traditionally been less of a concern for right-wing political parties whose voter demographics tend to skew towards older New Zealanders who are more sedentary in their lifestyles so their enrolment details are usually up to date. Left-wing parties, whose voter demographics tend the other way towards younger, more geographically mobile people, thus have more of a challenge on their hands.

This is why over the past couple of election cycles we've seen differing efforts in New Zealand to mobilise these voters. While there's debate over the political leanings of some of these efforts, groups like Rock Enrol or FFS Vote (For Future's Sake) are examples of third-party attempts to mobilise voters.

The parties themselves take slightly different approaches. The Green Party of Aotearoa New Zealand, for example, put a significant amount of effort in trying to engage overseas voters, as is evidenced by their having an "international candidate" whose job is to secure votes from the Kiwi diaspora around the world. Likewise, Labour is showing signs of trying to engage Kiwis living in Australia with a recent Facebook event calling for volunteers in Melbourne.

If this was 2014 you'd see this activity quite visibly ramp up, but 2017 might be a somewhat different beast. With improvements in social media and online ad targeting since 2014, the majority of the Mobilise content that's going to be produced from an online perspective is going to be below-the-line advertising.

The rationale for this is pretty simple and somewhat cynical. Political parties only want to mobilise their own supporters to vote, so the more you can focus your GOTV activities to focus on your core supporters, the better off in theory you'll be. What all the parties will be doing at the moment is figuring out exactly who they need to talk to in those final three weeks of the campaign in order to mobilise them as voters.

To achieve this from a social media perspective they'll look at all the voter data they've collected via the previous two phases of communication - Persuade and Participate - and match that up with online advertising tools. Whether it's plugging carefully curated email lists into Facebook's ad tool, selectively buying Google AdWords, or targeted advertisements on major news sites, you can generally assume that if you see an ad from any of the major political parties (National, Labour, the Greens and NZ First), you're seeing it because you match some type of demographic, behavioural, or geographic filter that suggests you might vote for them.

While the advertising tools on social media here aren't as detailed as are available in the United States and United Kingdom, they're still detailed enough when partnered with your own data to allow campaigners to be pretty confident they're talking to the right people. It won't always be perfect, for instance I've found Facebook's geographic filters here aren't particularly accurate, but it's much more cost effective at reaching a far greater number of people.

If you doubt the importance of this, you only need to look at the U.S. election. Donald Trump not only campaigned more in key battleground rust belt states, but ultimately was able to mobilise a greater number of supporters to vote than Hillary Clinton, enabling him to benefit from the quirks of the electoral college.

Participate - Petitions aren't just about supporting an issue anymore

Think by signing your name and email to an online petition that you're showing your support for an issue that's important to you? Think again. While getting you to show your support for something is part of the reason why political parties and lobby groups hit you with an endless stream of petitions and sign up pages, it's no longer the primary driver of this behaviour.

Ultimately it's all about getting your email address and starting you on a pathway to being an active support and participant in their campaign or cause.

Having already covered the extensive lengths political campaigns go to in persuading you to support them, I'm now going to look at how they aim to - and I apologise for borrowing a corporate term here - move you up their value chain.

Generally the value chain that political campaigns, lobby groups, and NGOs use goes something like this:
Voter sees content about an issue > Voter signs up via a petition on that issue > Voter receives request to donate > Voter receives request to join campaign/cause > Voter is asked to volunteer.

The idea is that each step in this path represents a greater commitment from the person they're targeting, and to move people along that value chain you need to be able to speak to them in an increasingly more targeted and personal way.

Traditionally political campaigns have done this through building voter files, and none do it better than the Democrats and the Republicans in the United States. As I've previously recommended, Sasha Issenberg's The Victory Lab does a fantastic job of explaining how this data driven approach has both evolved and is implemented in the U.S.

Where the U.S. and overseas jurisdictions differ from New Zealand is that their privacy and information sharing laws are much more lax than those here. If, like me, you've signed up to receive emails from both the Republicans and the Democrats, you may have noticed that you also eventually receive other emails from affiliated campaigns and causes. That's because they're constantly buying, selling, and refining their voter files to get the best possible information on voters so they can better identify those they want to target, and what messages that want show them.

What's changed in the last decade hasn't necessarily been this approach, it's been the channels it's applied to and the scale on which it can be done. Prior to the internet and social media it was used to micro-target direct mail, newspaper, radio, and TV advertising buying, door knocking, and phone calling. Now the same methodology is being used on social media, email, and online ad platforms, and it's backed up by data on an unprecedented scale.

As I mentioned before, New Zealand's privacy laws means that unlike the U.S. where political parties go and buy massive consumer data sets from commercial providers, that's simply not possible here. The net result has been that New Zealand's political parties, lobby groups, and NGOs have to go out and build these data sets themselves largely from scratch. This is why signing your name, age, email address, and town or postcode to a petition is almost mandatory to be able to show your support for it. Because once they have that data, they can add a line to their voter file with your name, contact details, demographic and geographic information, and issue tags to it.

The idea then is that they begin to target you through their online channels to get you to take the next step. Signed a petition? How about you donate so you can support us as we fight for it? Donated? Why not join our campaign and be part of a community of like-minded people? Joined our campaign? How about getting out and about and volunteering to help us out?

Next, because you've signed a petition on a specific issue, that party, lobby group, or NGO now knows you have a potential interest in that topic, so they can tailor content to serve to you whether it's via sending segmented emails to specific audiences, use it to target Google AdWords or SEO optimise a website, or uploading that email list into Facebook to let it create a broader audience of people who are similar to you via their demographic, geographic, and Facebook activities. On this final note, it's important to keep in mind that Facebook doesn't share that data on who else might look like you with someone who's setting up ads, but it does mean political parties can continually refine their voter files, data analysis, and targeting strategies by serving up similar content to those people.

While the Green Party of Aotearoa New Zealand have, in my mind, historically been the best at this, largely borrowing techniques from Greenpeace, WWF, and Forest & Bird, as far as I'm aware the Labour Party have had the single most successful sign up campaign with their "What number baby are you?" My understanding is that Labour gained somewhere in the order of 40,000 to 50,000 new email addresses to it, and while there was a larger than normal level of backlash to it when people realised they were also signing up to Labour's email list, I'd wager that the churn of unsubscribes they were getting were far outweighed by the gains they made.

It's also worth pointing out that contrary to the headline in the above article that Labour was tricking people into signing up, from my experience there's usually been text (and I believe it's a legal requirement) to the effect that signing a petition means you're agreeing to receive communications from that political party and its candidates.

As someone who's worked in this field, the notion that this is somehow a political dark art I think isn't really fair. Rather, it's part and parcel of the continuing evolution of not just communications practice, but what we're seeing more generally too across society. Other than long-standing commercial use of consumer data to target people with advertising, this approach also has the potential to offer massive benefits from a civic and social perspective too.

Whether it's getting people to engage with our democracy at all its levels, being able to understand how to communicate with people in a crisis via targeted alerts across mobile phones, social channels, and physical alerts, or being able to more accurately predict and intervene early to prevent health or law and order issues, data like this can and is used in a responsible and productive way. We're hugely fortunate in New Zealand that we do have very robust structures in place for governing its use responsibly, and a culture that does err on the side of caution about its use.

Next time we'll look at how political campaigns aim to Mobilise their supporters to get out and vote for them, a phase of political campaigning that now spans the two weeks prior to polling day with the opening of overseas and advance voting.

Persuade - Values, policies, and attack communications

Last week I introduced the concept of there being three phases of political communications - Persuade, Participate, and Mobilise. Today I'm going to drill down into the first of these - Persuade - which in New Zealand's election cycle makes up the bulk of what you've seen over the past three years. Just the weekend been you will have seen these played out around the country too with hoardings going up around the country from midnight on 22 July.

The primary purpose of Persuade communications is, as the name suggests, to persuade you to vote for a given candidate or party. They're all about building the narrative around why choice A is better than choice B, and convincing you to make that choice, or to put you off the other choice. Persuade captures the who, why, and what of politics:

  • Who is a candidate or party? (Identity)
  • Why are they running? (Values)
  • What are they proposing to do? (Policies)

Broadly speaking, Persuade communications are made up of three themes. These are, in order of importance:

  1. Values and identity
  2. Policies
  3. Attack

I've listed their priority in these way as in my experience, and supported by some academic insights, these tend to be the factors that sway a voter when they're deciding who they want to vote for. There's a reason why personality politics have become the norm in our highly media saturated world; it's because values and identity are the primary drivers of voter behaviour.

Values and identity

Voters want to know if a candidate or party shares their values and whether they can identify personally with them. This is one thing that Sir John Key did phenomenally well across a long period of time. Political commentators often talk about about "Middle New Zealand", what they're referring to is a large swathe of the country who generally share the same values and personal identity. It was this group of New Zealanders who Sir John Key, and the National Party, have largely monopolised the vote of since 2008. I think the whole point about values and identity being more important than policies was nicely summed up by the brilliant NZ Swing Voters Against Dogmatic Party Affiliated Memes Facebook page recently:

19884154_1417066014997849_2050326632142056717_n.jpg

If you look at a politician like Winston Peters, it's doubtful whether his supporters know what NZ First's policies are other than the Super Gold Card. However, they believe that his anti-immigrant and economic nationalist rhetoric aligns with their own values that hark back to a largely imagined "golden era" for New Zealand from their youth. They identify with Winston Peters as someone similar to them, and so vote for him.

There's been three pieces of digital content recently that are perfect examples of values and identity communications. The first was the Green Party of Aotearoa New Zealand's Great Greens campaign in April which was all about highlighting the diverse range of values, issues, and lifestyles that Green Party supporters have in common. National followed with their own piece squarely focused at linking Bill English's value and identity with that of Middle New Zealand with Let's Get Together. Labour has been the most recent with their first campaign ad which attempts to soften Andrew Little's often harsh public image by showing him in family and relaxing settings, as well as throwing in a bit of policy and leveraging Jacinda Ardern's popularity for good measure.

All of these pieces are about highlighting the values that a party or candidate stands for, and promoting them as being someone that voters can identify with as empathising with them and their concerns.

Policies

While policy communications do often sadly play second fiddle, they're still important. Think about 2005 when a resurgent Don Brash nearly led National to victory over Helen Clark's Labour. Labour's interest free student loan policy was credited with ensuring a significant youth turn out for them and seeing off that challenge and putting Labour in the box seat for post-election coalition negotiations.

The party in government and opposition usually approach policy communications differently over our Parliamentary term. A party in government needs to be constantly promoting the good things it's doing with its time occupying the Treasury benches. Whereas the opposition are creating their own narrative over that period about why those measures aren't working so that come election campaign season, they're ready to announce solutions to a receptive public.

Getting policy communications wrong can also undermine your values and identity messaging. Nothing illustrates this better than Theresa May and the Conservative's disastrous u-turn on the dementia tax. In an instant May lost her credibility as a decisive leader which drastically undermined all the values and identity work the Conservatives had done.

Where you'll see policy communications really ramp up is from early August through to the start of advance voting as parties announce policies every other day. While the values and identity content will continue in the background, you should notice a switch to promoting the policies they're announcing, especially as they attempt to sell their ideas directly to voters rather than relying on traditional media participants to relay their announcements for them.

Attack

Despite some holier-than-thou claims out there, attack communications do work to influence who people are going to vote for. That being said, they need to be used carefully and sparingly. In New Zealand we've typically opted for a humour based approach rather than the more outright negative stuff seen overseas. National recently did this using a clip from the AM Show where Paula Bennett made light of the fact that Labour's billboards feature both Andrew Little and Jacinda Ardern. Back in 2014 National's theme was the idea of Labour, the Greens, and the Internet Party rowing in opposite directions in their boat.

Our left wing parties have taken a more indirect approach, with the Labour Party linked Backing the Kiwi Meme churning out regular meme based attack pieces, as does the Freshly Picked Green Memes, both of which have much largely Facebook audiences than their equivalent right wing pages.

Attack doesn't always work though. Again the United Kingdom's election is a good example of this, where the Conservative's relentless attacks on Jeremy Corbyn's links to the I.R.A. and Hamas were to little effect, especially as Labour turned out a massive youth vote for them. The reason why these weren't successful I think is largely due to the fact that the people who this line of attack was most likely to resonate with were already voting Conservative, whereas those aged 18 to 32 largely didn't live through the Troubles in Northern Ireland or Palestinian linked terror attacks, and so they didn't put much stock in them, instead judging the Jeremy Corbyn who was in front of them then and there.

An equivalent example in New Zealand has been Labour and the Greens endless use of the neo-liberal attack line, or National and ACT's use of Fortress New Zealand. The negative experiences of Fortress New Zealand and the upheavals that came with liberalising the economy were issues that largely played out through the 1980s and early 1990s. They're upwards of 25 years ago and there's a new generation coming to the polling booths this year who aren't moved by attacks based on conjuring up ghosts of the past.

The impact of social media

At a fundamental level social media hasn't changed the core themes of Persuade content. All the things that have underpinned political messaging Where it has had an impact is in the ability to reach and target more people so that you can get that messaging in front of a larger audience. When I talked about Middle New Zealand earlier, it's remarkably easier to identify and advertise to that audience on Facebook based off what our most popular interests are. These are usually brands that you'd typically identify as being integral to New Zealand identity. It's things like Whittaker's Chocolates, Air New Zealand, the All Blacks, even imported brands like McDonald's, KFC, and other household names all feature. Keep in mind that this interest targeting is on top of using geographic and demographic targeting too.

While Facebook's audience targeting in New Zealand isn't as refined as it is overseas, this ability to micro-target audiences on a massive scale at a very cost-effective rate has meant that candidates and parties are able to tell their story and persuade people to vote for them on a much larger scale than could ever be achieved through face-to-face, direct mail, phone calls, or even email.

As I alluded to in the Attack section, social media has also transformed the way attack communications are carried out too. The ability for parties and their supporters to create Facebook pages, Twitter handles, or their own websites, is now easier than ever, especially when you think of memes as a form of content. With all the meme generating websites available, this type of content is no longer the preserve of those with graphic design nous and tools, rather anyone with an idea for a meme can turn it into reality in a matter of minutes, and see if it finds legs in the public arena. If it fails, they can always rinse and repeat as often as they like.

Thankfully New Zealand hasn't yet seen anything on the scale of professional trolling as detailed in the New York Time's report "The Agency" which reveals the covert operations of Russian state-sponsored trolls. That's not to say that it isn't happening here, rather if it is it's on a small scale and doesn't make an impact on the mainstream discourse as similar events have done in the United States. In fact the NZ Herald's Kirsty Johnston recently did a very insightful piece on the world of New Zealand's alt-right and their plans to influence the election, which included ideas around using social media.

In the next blog I'll look at the Participate phase of political messaging, and how campaigns are using social media and digital channels to get donations, grow their membership, and crucially get volunteers to help get their messages out.

Would rather donate to charity than spend on political ads? Yeah right!

If you're going to say you'd rather donate $1 million to charity than spend it on political advertising you'd think that's a pretty easy thing to stick to, wouldn't you? Not for Gareth Morgan's The Opportunities Party (TOP) it seems.

TOP got a lot of attention last month for announcing that, in protest over the advertising money allocated to political parties, that it would donate $1 million to charity because they'd rather do that than spend it on political advertising. Rightly, they received a reasonable amount of press coverage for it and while many saw it as a cynical ploy for Gareth Morgan to get himself back in the news, it was at least going to do some good.

So far, so good. Except you'd think if you've just announced you'd rather donate to charity rather than pay for political advertising then you'd not run paid advertisements. Seems pretty simple?

Except that it's not that simple for TOP. To be fair, you can almost forgive them for spending money on promoting their ads about how they don't want to spend money on ads and donate it to charity. The net effect of it all is that some very worth charities are going to get badly needed funds.

The problem is, TOP has now gone on something of a political advertising spree including with print and Facebook ads.

IMG_20170620_140835.jpg
Ad on Water.png

The above are all clearly examples of political advertising. The first one teasing the release of TOP's health policy, the next on meeting the leader of TOP, another on their deputy leader on what can be done to bring down grocery prices, and the final one on water policy. They're all examples of paid political advertising. So what happened when I challenged TOP on this blindingly obvious hypocrisy on their part?

 

Screenshot_20170615-185719.png

This is where TOP is really getting disingenuous about their advertising. Being very familiar with Facebook's advertising platform, and given the frequency that I've seen their ads, they're clearly putting a significant amount of funding behind them to get that type of frequency that I've been seeing. So it's not just $100 on a single Facebook promotion, but a very broad and well backed political advertising campaign across a range of content and channels, including that ad in the Dominion Post too.

I'd also wager that TOP simply doesn't care about their hypocrisy either. If you've ever witnessed either their leader or deputy leader on Twitter, they both appear, in my opinion, to be rather immune to any criticism or contrary views whatsoever.

UPDATE 5 July 2017 10:40am: Contrary to TOP's Facebook comment posted above, apparently they do have a billboard on the corner of Willis and Boulcott Streets in Wellington. Gareth Morgan's left and right hands evidently aren't talking to each other, or they just don't care...

UPDATE 5 July 2017 3:55pm: Have now been supplied a photo of the billboard. Looks like TOP is treating everyone as fools.

19832484_10155583729317474_839854495_n.jpg