Hey, A huge dust-up took place this month – Facebook on one side, Australia on the other (or its government + media at least) – with most news reports around the world framing it simplistically as Good vs. Evil. Of course, there's a lot more to it. The media has an easy job painting Facebook as the Big Bad thanks to regular stories like one which popped up everywhere this week with the damning headline "Facebook Lied To Advertisers" – endlessly shared on Facebook, naturally. The headline seems pretty clear, right? Facebook lied. You can't trust them, and so on... right? We've Seen This Movie Before For anyone who has been around publishing for the last ten years, you are probably getting a lot of flashbacks right now to 2012/2013, when these same media companies were painting a giant big target on Amazon's back… and lapping up those sweet, sweet clicks that rain down once you whip up enough outrage on a hot topic. Just like last time, the Facebook stories can sometimes look a little different once you read more closely. The story headlined above was framed to hammer home the point that you can't trust the data you get from Facebook. Someone less familiar with the jargon involved might naturally assume that the suspect metrics spoken about are mission-critical. Like how many times your ads were shown. Or what you paid each time someone clicked on them. What Did Facebook Do, Exactly? What Facebook actually stands accused of is something far more trivial: inaccurate estimates of the size of some audiences – numbers which were off for somewhat understandable reasons. In simple terms, Facebook didn't revise audience estimates downward to remove dormant or fake accounts and it also didn't exactly... act with haste once it was raised internally. But to show this is less of a deal than the media have been making out, let's look at the effect in practical terms. If you were running an ad on Facebook, and wanted to target female readers over the age of 45, who owned a Kindle and liked cozy mysteries, Facebook might have told you there were 60,000 people in America you could potentially reach with an ad. (Note: these numbers are completely hypothetical.) Whereas the more accurate estimate once dormant and fake accounts were removed might have been 52,000 people. That's it. Not quite a nothingburger perhaps, I definitely want all the metrics I'm seeing to be clear and accurate, but certainly not the great swindle it has been made out to be. Facebook didn't overcharge, inflate the cost of clicks, or lie about the effectiveness of the ads anyone paid for – things you might assume from this week's headlines. Really, it had no material effect on ad performance. Personally, I was always operating on the assumption that audience estimates like this were extremely rough estimates anyway, and included things like dormant and fake accounts because… why wouldn't they? When Counting Stuff Is Surprisingly Hard Not every Facebook user is the same. Many check-in daily but some might go days or weeks without logging on. Others again might have created a Facebook profile at some point in the past, but only use it to message people or post a status update every few months or perhaps no longer use it at all. And then there are a whole host of fake accounts which are created for reasons good and bad (people protecting their real identity, others like me who are testing various things, and then nefarious types engaged in catfishing or full-time d*ck pic distribution). Here comes the part that makes me really cynical: this is not a new story. It's from 2018. So Good They Newsed It Twice This story about inaccurate audience estimates was reported widely in 2018, although with somewhat less sensationalistic headlines. If you dig a little deeper, all these stories spring from the same source: a lawsuit a few years back in California. Weirdly, none of media articles this week report on the outcome of that lawsuit. Did the plaintiff win? Were these claims substantiated? Nobody reading these articles knows because none of the journalists seemed to care about that part of the story. In fact, the claims weren't substantiated because the case was settled and never went to trial, but you don't see that mentioned anywhere because the media have their headline – Facebook "lied" – and that was enough to run with, and then spawn a hundred more opinion pieces and features breaking down what all this means for humanity… and then do the same thing again three years later because we are all goldfish these days in the constant firehose of outrage and clickbait. And the net effect for anyone advertising on Facebook appears to be quite negative indeed: Facebook is now much more reticent about estimating audience sizes - you may have seen Facebook simply not estimating some audiences at all (especially custom audiences). And all that serves to do it make it harder for advertisers. So instead of helping advertisers, you're actually hurting them. This media process of strip-mining hot topics for anything that can generate clicks is something we previously saw around Amazon. And, once again, this manufacturing of outrage does little to tackle the actual, meaningful, underlying issues surrounding Big Tech. The Real Agenda? But what it does is achieve is the erosion of trust in Facebook and its advertising program, which neatly suits the commercial agenda of most media organizations, who, lest we forget, are giant companies themselves – ones which are in a losing battle for ad dollars. There's a delicious irony in Fake News about Facebook going viral… on Facebook. Hoisted by their own petard, you might think, with some justification. Although why people keep leaving so many of these petards laying around is beyond me. Anyway. Facebook is not an easy company to defend, and I'm not here to fight on Facebook's behalf. My concern – as it was with Amazon in the past – is that authors might make bad marketing or career decisions based on slanted media articles from companies who most certainly have their own agenda, one festooned with delicious dollar bills. If you don't want to advertise on Facebook, if you don't want to use Facebook, if you want to campaign to shut down Facebook, that's completely your call. I'm not here to argue otherwise – I just want everyone to make decisions based on better information. I have had my own issues with Facebook in the past, and I've written things highly critical of them – and I'm sure I'd have penned quite a few more angry missives if I ever veered into the topics of politics or what overarching effects technology has on society, and our happiness. I still choose to use Facebook and I still choose to advertise on Facebook - and find value in both those things. That doesn't mean I'm blind to the negatives, but it does mean I tend to approach them as problems to be solved. How Facebook Fudges The Numbers I'm an experienced Facebook advertiser and come from a tech background so I know how to adjust my Facebook dashboard so I get a more accurate picture of how my ads are performing. For example, two key metrics for any advertiser on Facebook are CPC and CTR – the Cost Per Click and the Clickthrough Rate. It's imperative that we know what clicks to our books are costing (that's your CPC), and it's just as important to know if the ads are hitting the mark generally (that's your CTR) so we know if we need to change up our images or ad text to make them more attractive. But Facebook… fudges those numbers a little. Facebook doesn't lie but it definitely makes itself look good. Your job is to cut through the noise and get a more accurate picture. Specifically, you need to change the defaults in your dashboard to something called Outbound CTR. What's the difference? Regular 'ol CTR includes all clicks on your ad, not just those leading to your website or books on Amazon. Which means clicks on the Like button, on the share button, and clicks to comment – any clicks at all will be counted in that standard CTR measurement. Outbound CTR is far superior because it strictly focuses on outbound clicks – i.e. those which actually lead to Amazon/your website. Sometimes the difference between regular CTR and Outbound CTR is minimal, but on occasion it can be significant. And you need to know that. You need to know when Facebook is being straight-up like Paula Abdul, and when it is using all the filters to present "the best version of itself." How To Stop Facebook Fudging The Numbers Here's how you can start building yourself a more accurate picture of your ad performance.
|
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.