Today in Digital Marketing - 173: Which of These Two Lines is Longer: I or |

Starting point is 00:00:00 Do you have business insurance? If not, how would you pay to recover from a cyber attack, fire damage, theft, or a lawsuit? No business or profession is risk-free. Without insurance, your assets are at risk from major financial losses, data breaches, and natural disasters. Get customized coverage today,

Starting point is 00:00:18 starting at $19 per month at zensurance.com. Be protected. Be Zen. It is perhaps one of the worst kinds of mistakes at zensurance.com. Be protected. Be Zen. many of the most important metrics we rely on. And that is causing a lot of brand managers to re-review their campaign reports today, Monday, June 15th. Happy National Book Day, Argentina. I'm Todd Maffin, and here is what you missed today in Digital Marketing. Facebook says that they are recovering from a nasty outage that almost certainly affects your ad campaigns.

Starting point is 00:01:04 If you were running ads on Facebook or Instagram last week, turns out they accidentally double-counted certain metrics, not only on their web platform, but also through their API. So even third-party reporting tools had the bad data. And the metrics they overestimated were the very ones that we digital marketers rely on to determine whether our campaigns were a success. Conversion metrics, including leads, link clicks, page and post engagement metrics, all double counted. This is a big deal. If you reported to your boss or client that your campaign had a big lift in conversions, you might be in for an uncomfortable conversation

Starting point is 00:01:44 because those results are actually half of what Facebook told you they were. There is some good news here though. That glitch happened from 9am Pacific time on Thursday until noon Pacific time on Friday. So if you were not running ads during this 27 hour window, your numbers should be fine. Also, some metrics were completely unaffected by this unique metrics for one. So things like unique link clicks or unique purchases. Also, they say the bug did not impact delivery or billing, but they're not going to actually backfill with accurate data. So quoting Facebook, please take this timeframe and metrics impact into account when assessing your ad performance.

Starting point is 00:02:36 Somewhat embarrassingly for Facebook, this big ad glitch happened the same week that Facebook published a new overview on how their ad platform works. So you know the first step. You as an advertiser create the ad creative and then tell Facebook the audience who you want to see the ad. But from there, it hasn't always been entirely transparent what goes on behind the curtain. Now we have a little bit more information. For every user of Facebook in a given day, there are thousands of ads that they could be shown. How Facebook determines which of those ads they'll see comes down to a formula. Essentially, out of all the ads a user potentially can see, Facebook's engine will choose a winning ad, and they say they determine that winner based on three factors.

Starting point is 00:03:11 First, your bid. How much money you've allocated to the ad. Remember, Facebook here is talking about the bid price, not the total amount of budget for your campaign. The higher your bid, the more reach you'll get. Second, something Facebook calls the estimated action rate, which is the likelihood that each user will take action on your ad based on their past behavior, both on Facebook and off. And finally, ad quality. This starts out assuming that your ad is of maximum quality, and then your score gets degraded for every bad element.

Starting point is 00:03:45 Bad elements are like when people hide your ad, or report it as spam, or how much text is in your image, or whether you specifically ask for engagement. You know those like and comment and share for your chance to win posts? Those. Also, Facebook said again that ads with the highest bid don't always win the auction. Ads with lower bids often win, they said,

Starting point is 00:04:06 if our system predicts a person is more likely to respond to them or finds that they are of higher quality. There's a link in the notes if you'd like to read their full breakdown. The date is 1951, and a psychologist named Solomon Ash is standing at the front of a conference room. In the room with him, six men, seated at a table. They are here to participate in an experiment. The experiment you'll be taking part in today involves the perception of lengths of lines. He's showing them a big card with one line on the left that's about three inches tall,

Starting point is 00:04:43 and then three lines on the right of varying lengths. One the same size as the line on the left that's about three inches tall, and then three lines on the right of varying lengths. One the same size as the one on the left, one shorter, and one taller. Your task is a very simple one. You're to look at the line on the left and determine which of the three lines on the right is equal to it in length. And, like all good psychology experiments, this one is rigged. In fact, only one of the six men at the table is an actual participant. The other five, they're plants. In on it with the psychologist. They have been told to deliberately

Starting point is 00:05:14 give the wrong answer. In this case, line number two is clearly the wrong answer. It is visibly shorter than the control line. Nobody would get this wrong, right? Two. Two. Two. And yet, as they go down the line, with the plants giving the wrong answer, the real participant at the end of the table starts to fidget nervously. Two. Two.

Starting point is 00:05:41 And then they get to him. He stares at the card, sees that number two is clearly the wrong answer, and says... Two. And it wasn't just him. 75% of real participants intentionally gave the wrong answer. This study is known today in academic circles as the Ash Conformity Experiment. And a great new read up at marketingland.com cites the study in a piece about how neuroscience can help capture customer sentiment and predict future behavior. Quoting from the article, market researchers are aware of the biases associated with group dynamics. Even with the best safeguards in place, group responses don't always reflect real-world experiences.

Starting point is 00:06:28 As you explore different approaches to capturing customer sentiment to estimate behavior in the future, you might want to consider how your offering, or the creative work associated with your offering, will activate certain parts of the brain. One study they talk about in the piece found that measuring activation in a certain part of the brain, a part of the brain that I'm not even going to try to pronounce, was one way of predicting which new songs would become hits years later. Anyway, it's a great read if you are interested in the intersection of marketing and neuroscience. And there's a link in this episode's notes if you'd like to read it.

Starting point is 00:07:10 We have seen forecasts of next year's digital ad markets for a while now, but only recently are we starting to see those whose measurement period take the COVID crisis into account. And one new study from Magna says total ad revenues in the US will, as expected, decline, but not by the catastrophic drop that you might be expecting. Magna forecasts 2020 will end up only 7% down when all is said and done. Part of that mitigated by a rebound they're forecasting for the second half of this year. Next year, they say, ad revenue should grow over 6%, but even with that growth, total revenue will still be $9 billion less than pre-pandemic levels. Part of what's keeping that overall decline in check, the American elections later this year, and our world, digital advertising, is expected to tick up,

Starting point is 00:07:58 if barely, by 1%. But that's an average, of course. When you drill deeper into that digital marketing space, search ads are expected to actually drop by 1% this year. Banner ads will drop by 11%. But social media and other digital ad formats up 8% by the end of the year. And two short things to wrap up. First, Google is now apparently sending out emails to Google My Business admins when a review has been removed from your profile. Nice touch. And if you use Microsoft Teams and have been jealous that those of your colleagues on Zoom have those fun virtual backgrounds, well, now you can have them too. They are in the new version of the app, so be sure to update the app. Classified ads are available on this show. 20 bucks will let you reach a thousand listeners in the digital marketing space. There's a link in this episode's description if you'd like more information. Our theme was composed by Mark Blevis.

Starting point is 00:08:56 Ad sales by Podcorn. I'm Todd Maffin. Talk to you tomorrow.

Today in Digital Marketing - 173: Which of These Two Lines is Longer: I or |

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.