Today in Digital Marketing - Special Episode: Can Too Much Ad Frequency Work in Your Favour?

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. Hello, today's episode is a special edition as I'm out of the country for agency business at zensurance.com. Be protected. Be Zen. or our premium newsletter, but we're bringing you four of these during my absence. If you like what you hear and want to learn more about the premium podcast, just tap Go Premium in the show notes. Our regular marketing newscasts return when I'm back on Wednesday. Enjoy. We have a lot of metrics available to us in the various platforms, ad managers, CPMs, clicks, reach.

Starting point is 00:01:02 And one of the numbers marketers often keep an eye on is frequency. Frequency, of course, a measure of how many times an average person in your target group has seen your ad. There's great debate about what that number should be. When I started my career in marketing almost 30 years ago, the going wisdom was that 27 was the magic number. Why? Because people thought it took nine exposures to an ad to make a decision, and that people only actually remember one out of every three ads. Nine times three was 27. These days, of course, that number seems ludicrously high,

Starting point is 00:01:37 and I think most people in our space believe that number should be between two and five, depending on what funnel stage you're at. Any more, and consumers become weary and start to get negative feelings about the ad. There's been a lot of research on how much frequency is too much, but that research to date has focused on near-term results, more immediate effects. Which is odd, considering the whole point of marketing is to build a long-term relationship between the brand and the consumer. Is it possible that if we push through the uncomfortably high frequency, yes, even if that generates short-term negative feelings, that eventually the consumer

Starting point is 00:02:19 will warm up to the brand? That's what Anne Cronrod and her colleagues sought to answer, and they published their findings in a paper called How Time Can Reverse the Negative Effect of Frequent Advertising Repetition on Brand Preference. She's an assistant professor of marketing at the University of Massachusetts and joins me now. Dr. Cronrod, welcome. Hello. Can I just say the title of your paper would make for lousy clickbait considering you give away the result in the title. And truth be told, that was the subtitle.

Starting point is 00:02:50 The main title of your paper was actually Ad Wearout Wearout. So that ad fatigue itself eventually fatigues. Is that right? We're mainly talking about ad annoyance more than ad fatigue, but wear out kind of combines together, I think, annoyance and fatigue. And you tested for three factors, actually, annoyance, memory and brand preference. Can you walk us through each of those? Yes. So in your introduction, you were actually mentioning that people might be very annoyed with ads, especially in the short term. And the question, what happens in the long term with what happens with this annoyance and also with a parallel process, I guess, that is which is memory.

Starting point is 00:03:42 So there are these are two parallel processes that happen when you see an ad or whatever, and both of them get higher the more times you repeat an ad. So if you repeat an ad many times, someone would get more annoyed with it. That makes sense. But also I was just saying, let's not forget, people will also remember it better. So the outcome of this is, okay, so if I repeat the ad many times, then people get annoyed with it and remember it. Well, is it good or bad for business? And what we're trying to see and what we managed to see eventually in field experiments that we ran on campus, but also in more controlled lab experiments, is that over time, there's an interesting thing that happens. Annoyance, which is kind of a negative feeling, tends to fade much faster than memory. So you're kind of over time. And when I say over

Starting point is 00:04:49 time, I mean, about two to three weeks, which is not a lot of time, actually. Over time, you tend to kind of forget that you were annoyed, the annoyance tends to go down but memory actually keeps for a while for a long while it just keeps high at the same level for a longer time which is a very interesting thing that it's kind of it's for any type of it's not only for advertising it would be for any type of of stimulus we call it stimuli um so like for faces if if or for people, if someone is, you know, you see them repeatedly, and they keep annoying you. But then, after a few years, you meet them, you forget that you were annoyed, and you're actually you remember them better. You're asking, so wait, what about the preference? That's the third. That's the outcome. So if I'm measuring, and that's what happens a lot in marketing, as you mentioned, if I'm measuring the outcomes of ad repetition in terms of preference for a brand, if I do that immediately, I will see reduction in brand preference because I'm repeating the ad too much and it's annoying, right?

Starting point is 00:06:09 But then over time, this ad preference would be increased. So people would be more likely to prefer that brand for which the ad retirements were more frequent, which is good news for marketers, of course. Did you get a sense of where that tipping point is? Is it based on time, like after three months of heavy ad rotation or maybe impressions, like once you hit a frequency of 30, then things will eventually in the long term turn around? Did you identify a numerical tipping point? So we did have an experiment that was over time. It didn't last for months. We actually, we didn't have to because what we saw is that, so we had people kind of exposed at the same time.

Starting point is 00:06:56 So if you're asking about how long should I keep the repetition going, This we had for technical reasons. We had to do that kind of all at once. So people were exposed to the same ads in our experiments kind of within the same like half an hour, which is very frequent, of course. So it was in that sense, maybe less realistic. But then after that, they were not exposed to the ads. So we can talk about the time lag between the exposure. I don't know how long it took and how many exposures you could play with that. And then the next time that you have to make a decision.

Starting point is 00:07:39 And there's something interesting that happens in between these two occasions. So when you're exposed to some information, then you have to make a decision. When you are exposed to the advertisement, that's not necessarily the time when you actually think about purchasing something, right? Like, I don't know, wedding halls or venues. So we're exposed to these advertising, but we don't consider getting married all the time. When time comes, and this happens, it could happen years after you were exposed to that repeating advertisement for a particular venue. When time comes, that's when you actually form your attitudes about the venue. And at that time, it's really important that you are not annoyed anymore, according to our results, at least,

Starting point is 00:08:33 but you remember this particular venue. So it's really good news for marketers, because it means that I can repeat frequently, repeat my advertisement, and then I don't have to insist that people run and rush to purchase a product immediately. Actually, it would be in my benefit to wait. I see. So it's less about the specific time frame and it's more about timing the offer. I'm thinking like maybe a resort in Mexico where people are not shopping for resorts in Mexico in the heat of summer. Usually they want to go in the winter. So maybe you run all the repetition during that summer phase and then stop so that when they are in the winter, enough time has passed for the annoyance to wear off and the memory and brand preference to kick in. Is that more or less right? Yes. Yes. I don't want to insist on the three

Starting point is 00:09:30 weeks that I mentioned before of the time lag, just because the way I achieved these three weeks was experimental. It was in clean conditions. People didn't see any other advertisements during that time. So within the context of my experiments. So it's not as realistic. And therefore, I'm being cautious saying, oh, it's three weeks or three and a half weeks. I don't want to do that. But this is kind of that's the time frame that I find in my experiments. So we can start from that. I thought how you tested this in one of your experiments was both fascinating and a little funny, given that it involved hundreds of rubber spiders.

Starting point is 00:10:17 Oh, yeah. Could you tell me about your testing methodology? Let's start with the posters. Okay. So first, this was for Halloween, which is an American holiday. And it is all about spooky stuff, right? And we ran the experiment in university residence halls in dormitories, where people actually decorate the rooms with different stuff. And they have Halloween parties.

Starting point is 00:10:48 Students, at least in the United States, they really love Halloween as a time to party. And they use lots of spooky things. So we just we bought a whole bunch of like hundreds of those really small spiders as a decoration. And we kept them in a closed box so people don't know what we have for them. But before we sold or we didn't sell them, we actually gave them away for free. But before we did that we advertised halloween and uh the giveaway and you did this through posters in the in the dorms right right so we we designed two posters and kind of it was indeed very much fun to design

Starting point is 00:11:39 the posters to think about the ad copy and all this stuff. As academics, we don't get to design ads that often. So I was really having fun being a copywriter myself. So this was really fun. And then the difference, why did we have more than one poster? Because one of the posters, we kind of manipulated the number of repetitions and the number of times it would be on the wall by just leaving it on the wall for a longer period. So students that go by the board, they would see it over and over again every day and hopefully get annoyed by having it for so many days.

Starting point is 00:12:26 So we had to talk to the administration not to take it off. And then in the other hall, we put another poster, and we kept it for a lesser time online. And we wanted to see if students would remember the poster, and if students would prefer any of the two brands. We kind of designed it as if the spiders were coming from two different brands. So eventually we had people choose between the two brands and hoped. And actually, that's what we also saw that when Halloween actually came and it was time to really make a decision, like I mentioned before, you form a decision, you form your attitude when it's time to do that, not when you see the ad.

Starting point is 00:13:15 So when they saw the ad, they were just, you know, they remembered it and they were annoyed by it, both. And the more frequently or the more days it was up the more annoyed and the better they remembered it so later they kind of forgot about the annoyance the annoyance went down and they chose more they preferred to get the present from the the ad that the brand that was represented in the more frequent ad and and can you explain sort of how that actually sort of worked? Did you have like a table set up with two boxes, one from brand A, one from brand B, and then they just had to pick which one?

Starting point is 00:13:54 That's exactly what we did. Yes. Yeah, that's what we did. The interesting thing that happens is, and it has a name, but it doesn't matter, that over time, it's a name, but it doesn't matter. It's a survival mechanism that over time we tend to kind of decay and put aside everything that is negative in our experiences. And we just keep going with only the positive ones. So that's how we explain why annoyance goes away so fast compared to memory.

Starting point is 00:14:28 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 at $19 per month at zensurance.com. Be protected. Be Zen. Were they ever told that they had been part of an experiment with those posters? Well, yes, because actually, after, so if you're asking, you know, in terms of ethical conduct, if you run a field experiment and you need people to not know that they're in an experiment and behave naturally, one of the things that you often would do is it's called debriefing. You would tell them later when they're done, when you are done with them, you would tell them that now we're ready. Now you have actually been part of an experiment and there's no risks to you. And here's what happened. And if you're interested to learn more, here's my email, etc.

Starting point is 00:15:41 I guess one of the limitations of that approach, though, is that people were getting a free gift rather than making a purchase. Do you think the results would have been different if you'd have tested purchase intent? Yeah, this is definitely one of the limitations of anything that involves an experiment versus an actual, you know, actually selling something. We in academia, we're often limited because we can't have any financial interest. So I, we are often not allowed to take money for anything. And sometimes experimenters even take money and then return it back to people, if they're able, in our case, when we weren't able to track back the students, you know, and get them back. I wasn't testing people's ability or likelihood to part with

Starting point is 00:16:38 cents or dollars. I was testing preference. So this is just kind of to remind that this was my dependent variable. This was what I was looking at and not so much about willingness to pay or would I buy it or not. But I agree with you that a nice extension of this research would be to actually try it with purchase. And the rubber spiders thing was an in-person thing. I felt that your second test, the second experiment was a little closer to our world in terms of digital marketing. Can you talk about the travel mug ads? Sure. This was the long-term kind of experiment where people saw one of, they saw two ads, they were reading an article. And, you know, sometimes when you read an online article, there are ads that pop up. And I was

Starting point is 00:17:32 trying to create this sort of experience when you're reading an article online, and some ads pop up, and you can just get rid of them by clicking the X, right? It's not too much, but if they keep coming, then it's ad repetition that happens. From my perspective, that's what I wanted, right? So the ads for these two, just travel mugs, regular mugs, that again, it was a little bit of fun for me because I designed the lousy ads for those mugs. The ads weren't very good because I'm not a very good copywriter, I guess.

Starting point is 00:18:12 But so the ads were not amazing by themselves. So that's another kind of it's a more, we would say, more conservative test. So the ads weren't that good. They were annoying by themselves. And then they also repeat during your reading of the article. So people kept reading the article, they didn't notice necessarily which ad appeared and when, which is also important, because it's kind of it's not a very conscious process. And then when they finished reading the article, we gave them a lag of time. So some people completed the next stage where they were saying which of the mugs they would

Starting point is 00:18:57 prefer. One of the ads repeated more often than the other. That was the difference between the two conditions, I guess, if you would say this way. So some people responded the next day. Some people responded after a week, two weeks, three weeks, and so on. And we were able to actually see a continuum not just two spots we were able to see a continuum of what happened to their annoyance towards the ads the and compare the more frequent and less frequent one and what happened to their memory of the ads so we were able to see not only preference but also the why why preferred. So this was a very nice explanation for what is going on, that annoyance goes down very sharply, but then memory kind of sticks and keeps going for many

Starting point is 00:19:56 weeks after you saw the ad. Are there any instances where this play for long-term results, even though you are initially sacrificing a bit of negative sentiment against your brand, knowing that that will wear off, so that that sort of play for the long-term brand preference, are there any instances where that would backfire? the mistake of trying to remind people. So you're trying to introduce or include elements from the ad, not only the product, but elements from the ad, like the jingle or the copy elements from the ad, and you're reminding of the ad itself. That might be a mistake to do because at this point, you're reminding of the annoyance that was long forgotten. Oh, I see. Right. So because this is a short-term annoyance, you want to kind of leave an actual real gap without any reminding. Yeah. I'm pausing because it seems counter to everything we are taught in marketing, right? Which is set up your offer and then every week or so or every day or every X period, like constantly sort of be in there.

Starting point is 00:21:14 In fact, there are retargeting campaigns which are evergreen, which are on kind of set it and forget it, where someone will spend time on a website or they will engage with a post and suddenly they are in a bucket of people who are just going to see this ad constantly over the next X number of months. Whereas your research has found actually you should stop after a heavy cycle of repetition and let it marinate. Yes, I agree. I agree. This is kind of, in that sense, what I'm suggesting is different from the mainstream that would say just keep advertising. And the main difference would be that I'm looking into the future. I'm saying, okay, so frequency of repetition is something that has been explored in the short run. So what would happen? And in academia, people explore and they kind of they repeat the ads and immediately test

Starting point is 00:22:13 what happens after that. And obviously, memory is great, but annoyance as well. Yeah, I don't know of any other study that studied or identified like a long-term reversal of the effect of ad repetition. Was this a new finding to the world of marketing science? I hope so. Yes, but there was another paper that we referenced in our article that found something kind of similar. But we are explaining – I'm not sure I want to go into it. It's a little bit complex. There was one experiment that was actually a field experiment in, I think it was a Super Bowl, or some sports game, and I'm sorry, if I'm blanking on the exact sports game, where they found that over time, ads that were repeated

Starting point is 00:23:09 during the game, later people, the more frequently they were repeated, people tend to purchase them over time, they tend to purchase them more, which is, that's an actual marketing outcome, of course. Your testing was done in 2017. Since then, ad platforms have matured. The products that they offer, in terms of the actual ad products, have gotten smarter. Even entirely new platforms like TikTok have come along and dominated. Do you think there'd be a difference in your results

Starting point is 00:23:42 if you were to test this today? Yeah, so it depends what part of the change over time that we're looking at. So, for example, I think one change that has occurred is that ads today are more personalized due to the collection of data when you browse. So browsing data and you get, you know, if you're browsing and looking for, I don't know, handbags, suddenly on Facebook, you get ads or handbags, right? And so in that sense, since the ads are more relevant to you in the first place, now that they are personalized, there will be a reduction of the annoyance. I would expect it to be less annoying, even if it repeats, it would be, you would feel that it's less annoying because it's relevant to you. All of this up to the point when you

Starting point is 00:24:39 made the purchase. Once you made the purchase, it's not relevant to you anymore. And now it gets annoying again. And hopefully people who are marketing, who are setting up those retargeting ads are excluding people who have purchased or maybe even putting them in a bucket for targeting for a different product, like an upsell or something like that. What made you want to study this? Well, it started actually from the relevance, from the last point we just talked about. And the earlier stages of this research were specifically about relevance. I find the concept of relevance very interesting because I think it can move mountains metaphorically speaking it can change everything as for example I showed in previous research that has not got

Starting point is 00:25:36 published into in in this article that if you advertise frequently about stuff that is relevant to people, they don't get as annoyed or actually they don't even get annoyed at all. But if you advertise about something that is not relevant, that's when you get to their, you know, to their bad side and you don't want to get there. Well, I thought it was a very interesting study. I thought, you know, I always like trying to find studies which present a bit of a counter to the best practices of our world, which, you know, I think in digital marketing practitioners, I think often we fall into a mindset or an approach where we kind of lock in and perceive that to be the best practice and only the best practice. And so it's often I find it very refreshing to find someone who who presents a counter or an alternative way of handling it. And Cronod is an assistant professor of marketing at the University of Massachusetts. Dr. Cronod, thank you so much for your time. Thank you very much. Thank you.

Starting point is 00:26:41 By the way, Dr. Cronod is looking for partners in the industry to help with other research she's working on specifically around linguistics. So if you have any data sets of user generated content, like reviews or a whole bunch of social media comments that you'd be willing to share in exchange for some intelligence about what your customers really think about your product or service, she would like to hear from you. Her email address is ann, that's A-N-N, underscore Kronrod, which is K-R-O-N-R-O-D at U-M-L dot E-D-U.

Today in Digital Marketing - Special Episode: Can Too Much Ad Frequency Work in Your Favour?

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