Note: We have a growing community of mobile marketers!

 The Mobile Growth Lab Slack: A community that was a part of our workshop series – The Mobile Growth Lab, is now open to the general public. Join over 200 mobile marketers to discuss challenges and share your expertise. More details are available here: https://mobileuseracquisitionshow.com/slack/

If you’re ready to join the growing community, fill this form: https://forms.gle/cRCYM4gT1tdXgg6u5

PANELISTS Grace Ouma-Cabezas | Evan Woods | Matej Lancaric | Shamanth Rao | Anish Shah

ABOUT ROCKETSHIP HQ: Website | LinkedIn  | Twitter | YouTube

KEY HIGHLIGHTS

⚖️ Comparing the system pre and post-ATT.

🎯 How has performance changed over the last year in terms of metrics?

⏰ Why the first 24 hours data is important

💎 How bigger, established companies were at an advantage 

🦴 What kinds of creatives are working well now?

🛍 How to make decisions about creative investment

🪅 How spend and creative strategy are related

🧪 How channel mix has an impact

🎧 The shift to Tiktok and TV

🎲 Web pages seem to bring about more conversions

📈 The evolution of ad networks

📻 Paid partnerships with content creators

🎢 Investing in incrementality solves some of the problems with measurement

KEY QUOTES

How to capture data during the first 24 hours in games

Matej: The first 24 hours in games, especially on iOS, is really important. I had to talk to the product and game teams about how we can grab as many events as possible in the first 24 hours. For example, we had to implement different special offers in the game. Just capture as many purchases as possible in the first 24 hours, and not screw up the whole game economy as well, that was really important.

How web-based strategies can be more productive

Grace: We just leaned in with a much more web based strategy, which I think has been really productive for us because there was some education around who we were what we were and thinking about this onboarding, lead generation, bringing people into the funnel for our CRM and lifecycle to nurture them into subscribers for our products. So not so much a performance change, but it did heavily shape where we invested.

Following Facebook recommended strategies can be beneficial

Evan: I would say we actually grew a lot during these changes, mainly because our strategies were Facebook’s core five type strategies. So a lot of our Facebook structures, for example, are highly consolidated. We were really focused on broader audiences. So I think a lot of what has become best practice, we were already doing. Some were in preparation for, but really that’s where we saw the best performance out of Facebook. 

Deciding on the volume of creatives

Evan: For one of our clients- a 7 to 8 figure a month client, they need a lot, and things saturate quickly. We’ve been working internally on a calculator. If you’re spending as much a day how much creative will you actually need? It really does vary, because there’s another client we have, they got on to Oprah’s favorite things. That creative they rode it for almost five months, way after holiday. So it can vary. But I do think that spend is a big indicator obviously of how quickly your creative will saturate and we do try to triangulate what is that number of video stat that we need to constantly have in rotation in order to combat that saturation.

Emotion and storytelling can be more powerful than just targeting

Grace: What has been really successful for us, we had a video asset that was really emotive and touched on these use cases across the different life stages that we support that perform the best. And we also chopped it up into individual stories. So we had one that was more postpartum one that was more pregnancy, one that was a little bit ambiguous – you’ve just found out, you’re pregnant and you’re not sure if you’re happy or sad, but the sort of the compilation video that showed just how we imagined our user and the instances in which they’d reach for their app and go to Poppy Seed Health, performed the best. And then we found down the funnel that it was actually attracting primarily people in their early pregnancy. 

So even though we were showing all these different use cases, I think, for the user, they’re able to kind of foreshadow how they may apply to them. So it was broad in terms of our offering but actually performed way better than trying to drill in individually. So I don’t know if that’s a function of how targeting is changed but really digging into the emotion and the storytelling and trying to show them where you would be when you would need us, really contextualizing ourselves in an emotional way has been the most powerful thing.

The creative approach that Grace and her team followed

Grace: It’s gotten us temporarily banned from a few platforms. But we show a woman pumping, we show someone changing a dirty diaper, we show a woman in the mirror combing her hair and some of her hair coming out, alluding to postpartum hair loss, we show somebody on the toilet, somebody looking at a pregnancy test. So we really get in there because a part of our brand was eliminating shame and taboo in those situations. So those were really compelling to users. The only individual ad that performed almost as well was where we showed a woman breast pumping. And we always use real people that we actually know. These are not models of professional talent that we use and I feel that was so resonant. We used a plus sized woman and we got so much positive feedback like you’d never see this in maternity advertising. So occasionally we had to contest a few “This is adult content” to get those out. But once they went out, they performed exceptionally well.

Moving away from the traditional MoM & YoY growth

Evan: Diverting those dollars to conversion rate optimization I think is a big thing as well as if we can’t target or optimize as effectively as before, therefore leading to often lower conversion rates, can we actually impact that off of Facebook, or invest in better measurement, etc. 

Why a landing page can have a better conversion rate

Shamanth: We set up a landing page to describe the product, and the landing page has a CTA button that goes to the App Store. The advantage is that the landing page provides a lot more context. So longer copy, which the app store doesn’t provide. So I think that’s one reason why that converts much better now. But again, that’s much more effective outside of gaming.

Making use of affiliates to bring users to your app

Anish: If Facebook and Instagram aren’t as knowledgeable that you’re in that life stage, you’re not going to be seeing ads on them. If a customer is on a baby/pregnancy oriented website, you’re getting a better conversion rate, because they’re not being pumped with ads from other companies who are doing similar things. So the customer experience might actually be pretty good there from not being too inundated

Finding the right influencers

Grace: There are people who have relatively robust followings and have never done partnership deals before. They’re still approachable. We worked with somebody who had a million followers and it was still a very economical relationship for us. And it was their first partnership deal and they were excited. I feel like Tiktokers are still excited, but less jaded than the Instagram brand.

Broken measurements are avoided when you have good partnerships

Evan: For so many other publications, they’re incredibly bought in audiences, there is no issue around IDFA. And it is a flat fee. So you’re taking on risk. But if you have the right audience, you’re able to activate them. The ROAS I know for that particular publisher with every one that I’ve worked with is just extraordinarily high. So relying on other brands and having the right placement, hopefully somebody that’s a little bit more editorial and explains the benefits of a product or service.

Investing on data analytics

Evan: Large brands also use really basic tactics, in addition to GA and making sure that you’re getting data from the ad set and the creative and the campaign “How did you hear about us?” I’d say having posts or just a purchase survey, on the confirmation page or an email. Huge advertisers are using it. It’s obviously submitted from an actual customer and you are able to get a much better way through upper funnel channels, have their efficacy, I think it’s become a lot more important. And then yes, I agree that looking at just blended aggregate and trying to isolate how much you’re spending on one particular channel at one given time and seeing, is there marginal efficiency overall for the business, so we’re making these larger changes.

FULL TRANSCRIPT BELOW

Kayci  

Welcome. Thank you for joining us. We have Ruckus and Rocketship HQ coming together to present a conversation on omnichannel strategies for a post IDFA world. We have some incredible panelists here to share their knowledge and wisdom. Anish is going to be moderating today’s conversation.

Anish  

Quick introductions – my background was both in-house and consulting in growth for about a decade and then I started an executive recruiting firm in New York. We have a team of 16.

Grace  

I’m Grace and I’m very excited to be here. I’m the Head of Growth at Poppy Seed Health. We’re an app based telehealth solution that provides 24/7 text access to doulas, midwives and nurses for pregnant women. I’ve been around growth marketing for a long time straddling between DTC and digital media, mostly at startups. So I got to do a lot of fun stuff over the years. So excited to be here.

Evan  

My name is Evan Woods. I’m currently General Manager at DT.Co. We are a growth studio, we focus on D2C businesses, we have about 15 clients right now. Prior to that, I spent most of my time within growth and retention marketing at various startups, including HelloFresh and a couple of others. I started my career on the publisher side at Twitter and LinkedIn doing more marketing operations stuff and I’m very excited for this conversation.

Shamanth 

Hi, everyone. My name is Shamanth Rao, and I’m the founder of the boutique growth marketing firm Rocketship HQ. We work with mobile apps to help them grow. Our portfolio is split between games and consumer subscription apps. I’ve managed multiple millions in spend in the past primarily with games, I was fortunate enough to be a part of three acquisitions and I run the podcast Mobile UA Show. I’ve had some of you on the show as well,

Matej  

Hi guys. My name is Matej Lancaric. I’ve been in the gaming industry for nine years already. Right now I work as a UA consultant & freelancer with multiple gaming companies around the world, and I have my own creative team of 12 motion designers. I’m happy to be here. 

Anish  

What we’re going to be talking about today are the changes that have happened over the last year. About a year ago, Apple deprecated some of the attribution capabilities that marketers previously had. That has thrown a lot of marketers into a swirl. And so what we’re going to talk about is – What have people done over the last year? What have they learned? What are some of the tests that we’ve run? What experiments and how do we all collectively move forward? 

We’ve unearthed some of the more interesting questions. But before we get into those, I know everyone has a certain level of how deeply they’re dealing with attribution on a day to day basis and how the changes have affected them. So it’s pretty helpful for Shamanth to be here, who’s running a really popular mobile acquisition agency and has to deal with this everyday, to explain what was capable previously, what happened and what is capable now.

Shamanth  

At a high level, what changed about a year ago, actually before 2020, there was a unique identifier. Think of it as a phone number, called the IDFA for advertisers. So if you were Facebook, if you were Google, you could see Shamanth’s IDFA, which is basically Shamanth’s phone number, and say Shamanth made a $2 purchase in this word game, he bought this D2C brand, so Facebook and Google and any advertiser would know exactly what Shamanth bought. And that phone number was observable by all these platforms without Shamanth’s consent. 

About a year ago, what Apple did is say, none of the platforms are going to access this phone number, without Shamanth’s consent. That is a huge deal because not many users are going to be giving consent. Now Facebook doesn’t know that I have actually made all these other purchases. So Facebook cannot target me nearly as well as they did a year and a half ago. What that means is as advertisers and marketers, the performance is just far worse than what it used to be just because we cannot target purchasers or high value users in the way that we used to.

Anish  

Which is a great segue into the first question, how has your performance changed? Since the policy changes? Some of you with agency businesses probably get to see across a wide variety of clients and customers, is there an average or is it all over the board in terms of how and why the changes have happened, what the metrics are, what is dropped, and by what percentage? Are there certain things that have made certain companies drop more than others? Like is it scale, the types of products they sell?

Matej  

I have a very positive view on the whole IDFA change. Since that happened, all the campaigns I’m running at the moment for iOS for the gaming companies, almost all of them are actually way better than before. Because first of all, a lot of people from the gaming industry, they don’t know how to evaluate the campaigns because the ATT made this pretty hard to see, they don’t know how to evaluate so they dropped spend. When COVID started, all the competition was just gone on the iOS side. So CPMs are down, everything looks pretty good. But in terms of how to run these types of campaigns, I had to go back and forth with the product team and make all the changes in terms of the configuration on the SKadnetwork framework. But in general, everything looked pretty good. So I’m fine. I can run these campaigns pretty profitably. 

Anish  

That’s awesome. That’s good to hear. So what kind of changes helped you get these campaigns? You said you had to dig in with your product and engineering teams, what have you put in?

Matej  

So

the first 24 hours in games, especially on iOS, is really important. I had to talk to the product and game teams about how we can grab as many events as possible in the first 24 hours. For example, we had to implement different special offers in the game. Just capture as many purchases as possible in the first 24 hours, and not screw up the whole game economy as well, that was really important.

So now, when I’m looking at the SKadnetwork schema, I’m able to actually evaluate the ROAS and run even lead value optimization campaigns. But that was a long, long process until I got to where I am at the moment. 

Shamanth  

Just to add to what Matej said, I think how you evaluate campaigns is so critical, because you cannot evaluate as deterministically as you used to. In fact, a number of folks we’ve worked with, have grown, they’ve done well. But they’ve had to embrace the fact that they cannot evaluate campaigns anywhere as clearly. A year and a half ago, you were like, Oh, this is my ROAS, this is exactly how this creative is doing. Folks who are doing well that we work with say, there’s some amount of fuzziness, but we know we’re growing. I imagine we’ll talk about this later. But folks have embraced some amount of incrementality testing, blended metrics evaluation, and of course, like Matej said, really dialed in on the first 24 hours, whenever that’s possible.

Grace  

For us, I’m happy to say we actually launched a year ago today. So it shapes our go to market strategy it in some ways, it was a little bit freeing because we didn’t have a historical performance and channel mix that we were worried about disrupting, but we also had to look at how we’re standing up our paid performance infrastructure, it’s definitely different than it would have been had we launched a year prior to this change. So we knew that the tried and true B2C model of the lookalike audiences on Facebook was not going to work for us. It also had us take a step back in terms of how we wanted to think about privacy holistically, especially with our product, we’re touching people during vulnerable moments along their birthing journey. 

So we actually opted not to do any pixeling at all for app download on Facebook, even though we knew that the quality wouldn’t be as good as it could be.

So we just leaned in with a much more web based strategy, which I think has been really productive for us because there was some education around who we were what we were and thinking about this onboarding, lead generation, bringing people into the funnel for our CRM and lifecycle to nurture them into subscribers for our products. So not so much a performance change, but it did heavily shape where we invested.

And then of course, I feel like Apple Search Ads was a big winner with all these changes, because that became such a fundamental part of our performance mix.

Anish  

Makes a lot of sense. Yeah, I remember I feel like over a year ago, when Facebook and Google had been evolving towards kind of removing deeper segmentation options, a lot of the conversation was around, they’re trying to level the playing field so that it’s not the best marketers, or the best acquisition people who win, it’s sort of like you could launch a product and just because you don’t have a super senior or experienced acquisition team who knows the ins and outs of these platforms, you could still win.

That was where UAC came from which I think removed some segmentation options from Google and then Facebook was really pushing their automation and AI and machine learning to make the decisions for you on which campaigns to go. So I’m wondering whether a lot of these changes are reverting back to where some of the companies who have access to the best kind of product engineering data, and acquisition marketing teams can now actually gain some level of higher leverage where the platforms previously were trying to level the playing field.

Shamanth  

I agree with what you said, and I think maybe like Matej said,a lot of his work was with the product team, and like I said, a lot of our work has been just making sense of the data, because two years ago, the data was all there in the dashboard. Now, it’s really understanding and making intelligent assumptions about what the data means. And that requires resources and expertise. Folks that don’t have that expertise are not going to be able to manage as easily. So definitely, I think there’s a skew in favor of the more established companies, I would say,

Anish  

Makes a lot of sense. Evan is working at an agency as a general manager, he has a tonne of D2C and ecommerce brands. Have you noticed differences between some of the brands you’re working across in terms of being affected more than others? And if so, are there some raw percentages or just something you notice, companies who do this maybe have gotten affected more? So whether it’s the size of budgets, or being at a certain scale, or lots of different things like acquisition, offline, online, etc.

Evan  

I’d say that it does vary across the board. And most of our clients are B2C. Some of them at a higher price point products and somewhat immediate conversion as well, a big change was the attribution windows going from 28 days to 7 days. So if there is a longer buy cycle, then that makes it a bit more difficult for brands to attribute directly to Facebook.

I would say we actually grew a lot during these changes, mainly because our strategies were Facebook’s core five type strategies. So a lot of our Facebook structures, for example, are highly consolidated. We were really focused on broader audiences. So I think a lot of what has become best practice, we were already doing. Some were in preparation for, but really that’s where we saw the best performance out of Facebook. 

And as you said, it’s become much more about machine learning AI. I think where the brands have succeeded, is creative. So if you are able to invest heavily, and it’s really costly, I mean, if you’re working with Narrative or Tube Science or many of those other agencies, they’re charging two or three full time employees potentially to give you creative, but that is ultimately one of the most important numbers. Now, as we all know, with targeting being less important, and having a tonne of ad sets, our campaigns are being less important. So I think the advertisers that have done well have been ones that have been able to invest in creative and to do so iteratively, and have the resources and see the importance of it I think are the ones that have had the biggest benefit.

I’d say also ones where they’ve also invested in other things outside of Facebook. So they’ve invested in their website, they’ve invested in collecting email addresses and understanding the impact, they’ve invested in analytics, even if it’s, as simple as GA and last click, they’re looking at Facebook as part of the picture, but they know they need to optimize what they own in order to compete. So those are some of the things that I’ve seen in terms of trends.

Anish  

Yeah, that’s a great point about creatives. Have there been any shifts in the types of creatives that are doing well? And also the quantity of creatives that are needed on a regular basis, as marketers are thinking about how to go about creative work for the next quarter. Is it investing in a couple of really slick pieces that are well produced and drive a lot of  brand value, or just getting a lot of quantity for a lower cost? 

Evan  

I think it varies. I think a lot of the trends that we’ve seen over the past few years are the same, which is UGC & native creative obviously works well. Of course, there’s an anatomy of how to make an ad from the hook value propositions, testing different value props, and the order of those. That’s pretty much the same and has been for a while. I’d say that a lot more creative testing is important, creatives that do work for broader audiences are also really important. So not being too niche, because that’s really where I think a lot of advertisers have moved. More volume is important – we have some brands where the founders are just making a bunch of UGC and we’re running it across Facebook and Tiktok, and it absolutely wins! It’s super low budget and just works really well. Other brands are partnering with large agencies and are getting batches of 15 to 20 new creatives every week to two weeks, sometimes new production, sometimes iteration. So I guess the strategy of creatives isn’t necessarily different. But I would say the volume and investment is different.

Anish  

The investment in volume has gone up. 15 to 20 creatives, that’s across just Facebook or is that a lot? So if someone was trying to forecast how many creatives they should be doing on a regular basis, does that feel high for someone to do 15 to 20 creatives a week.

Evan  

I think that is pretty high.

For one of our clients- a 7 to 8 figure a month client, they need a lot, and things saturate quickly. We’ve been working internally on a calculator. If you’re spending as much a day how much creative will you actually need? It really does vary, because there’s another client we have, they got on to Oprah’s favorite things. That creative they rode it for almost five months, way after holiday. So it can vary. But I do think that spend is a big indicator obviously of how quickly your creative will saturate and we do try to triangulate what is that number of video stat that we need to constantly have in rotation in order to combat that saturation.

Anish  

External validation is always really great. And in terms of creative I feel like a lot of brands have seen that those creatives tend to do a lot better than just showcasing your product and trying to explain it so I feel like that’s a good clue for a lot of people to go out and try to get some of those high level affiliates even if you can’t get an Oprah kind of hit. Even though Oprah’s new things can be a little bit more of a high level affiliate programme the way it’s built out so going and getting Oprah or Ellen if anyone can go figure out how to do that, helps a lot. 

Grace obviously your marketing a really high consideration product. Health care is so important to people and they have to be really trusting of it. Does your creative strategy change in that? With high consideration products you often have to have multiple different ads and multiple different looks to get someone to trust you. But with attribution not as great as before, does that change a little bit? Or do you have to feel like you have to shrink a lot of messages into one ad? 

Grace  

We’ve tried both approaches, actually, because our product is a little bit unique in that it’s a health based subscription. But it also relates back to an acute need, since the pregnant person of the birthing person may have a question or concern at 3am right? That may be the driver, or like, do I need this in my second trimester? How do I want to deal with this pregnancy etc. So we sat in an in between space when we first launched creatively, I’m just doubling down on the creative point. We’ve done a lot of investment, we have an amazing in-house creative team. It’s a small but mighty team. We produce everything in-house, we invested a lot in that. And that has been a big help in performance with all the money you can imagine all the visibility that we could have had in terms of targeting new parents, expectant parents, etc. That really isn’t as robust as it is today. So creative has been huge.

What has been really successful for us, we had a video asset that was really emotive and touched on these use cases across the different life stages that we support that perform the best. And we also chopped it up into individual stories. So we had one that was more postpartum one that was more pregnancy, one that was a little bit ambiguous – you’ve just found out, you’re pregnant and you’re not sure if you’re happy or sad, but the sort of the compilation video that showed just how we imagined our user and the instances in which they’d reach for their app and go to Poppy Seed Health, performed the best. And then we found down the funnel that it was actually attracting primarily people in their early pregnancy. 

So even though we were showing all these different use cases, I think, for the user, they’re able to kind of foreshadow how they may apply to them. So it was broad in terms of our offering but actually performed way better than trying to drill in individually. So I don’t know if that’s a function of how targeting is changed but really digging into the emotion and the storytelling and trying to show them where you would be when you would need us, really contextualizing ourselves in an emotional way has been the most powerful thing.

And I think it’s helped us overcome some of these challenges of targeting for sure.

Anish  

Can you explain what is in one of those creatives that has done really well? Like what would the audience see?

Grace  

Sure, I’m happy to.

It’s gotten us temporarily banned from a few platforms. But we show a woman pumping, we show someone changing a dirty diaper, we show a woman in the mirror combing her hair and some of her hair coming out, alluding to postpartum hair loss, we show somebody on the toilet, somebody looking at a pregnancy test. So we really get in there because a part of our brand was eliminating shame and taboo in those situations. So those were really compelling to users. The only individual ad that performed almost as well was where we showed a woman breast pumping. And we always use real people that we actually know. These are not models of professional talent that we use and I feel that was so resonant. We used a plus sized woman and we got so much positive feedback like you’d never see this in maternity advertising. So occasionally we had to contest a few “This is adult content” to get those out. But once they went out, they performed exceptionally well.

Anish  

Thank you for those examples. I think that’s super helpful and everyone in the audience can try to do their best to guess which one of those got shut down by Facebook. Do you feel like there’s that creative idea, how many new creatives you’re looking for in a particular week, month, etc.

Grace  

I would just double down on the point around spend. I think it really does heavily depend on your spend, having been at different brands and at different spend levels. We’re earlier stage on the smaller end spend wise, we’re able to get more life because there’s still so much that we haven’t penetrated but I’ve been at other places with a heavier spend where you really couldn’t run one piece of creative more than 4 to 6 weeks and we usually had 10 to 15 in place. I think it has been audience dependent too. Our life stage is dynamic, one moment you’re in it one moment you’re out of it. So I think we’re able to get a little bit more life out of our creatives for that reason as well.

Anish  

Because the spend is low there’s a lot more audience for you to touch especially if you know it’s working. Makes a lot of sense.

Matej  

From the creative point of view and gaming, especially if you’re a niche game, then you need to frequently refresh the creatives because obviously the audience is smaller than if you have a wide appeal game. And then you can just tap into almost unlimited audiences. Then there’s also the trends. And from my point of view, we usually try to refresh the creatives on a weekly basis. Ideally, at least one or two new creative concepts a week. This also depends on the spend, but anywhere between 100k and almost half a million spend a month, then you will just need four a week, because that’s just getting really interesting.

Okay, that’s great for a baseline. So maybe with some of the stuff you’re working on, 2 to 4 creatives per week, if you’re spending roughly half a million to a million in that timeframe.

Yeah, I think 2 to 4 below half a million and above half a million you will just need way more than five creatives a week at least that’s what I’ve seen working well.

Anish  

Okay, great. So to revise that: half a million and under, maybe four or less creatives. And then if you’re looking at half a million or more that you need five or more creatives per week. If someone is trying to create a baseline for how you think about this.

Shamanth  

We also work on a weekly cadence basis. We also find it helpful because we know there’s a weekly production cycle that goes on. A creative director and strategist may say, every Monday we have to consolidate concepts and gather the results. In terms of analysis cadence, it helps us have that discipline in having a structured testing process. Obviously, the quantity depends on the spend, but irrespective of the quantity, we still try to do it weekly, just to make sure we have that rhythm going.

Anish  

The original answer of how people were trying to approach the deprecation was spread across multiple different channels, especially companies who were incredibly Facebook and Instagram focused, which was really, almost the majority of D2C companies that I had seen. Have you collectively thought about expanding channel mix? I think the factors have been pretty straightforward, but how have you thought about channel mix? 

Evan  

I’d love to say we’ve successfully diversified away from Facebook, it’s a small amount of all of our clients’ ad spend, it’s just not true and unfortunately, we’re all just still relying on Facebook and Instagram. So I would say overall, budgets have often decreased and the idea of seeing this month on month, quarter over quarter, year over year growth, that sort of element I think has gone down and more smart media planning around, seasonality, promotions, other things that are happening and really capitalizing on those moments. Sometimes

diverting those dollars to conversion rate optimization I think is a big thing as well as if we can’t target or optimize as effectively as before, therefore leading to often lower conversion rates, can we actually impact that off of Facebook, or invest in better measurement, etc. 

I’ve seen that happen and in terms of actual dollars, I would say the two big channels I’ve seen are Tiktok & TV. Tiktok has obviously exploded and as an ad platform is becoming better and better. CPMs are just lower. I’ve seen a lot of advertisers have almost six, seven figure budgets on Tiktok a month and it’s extremely creative driven, very different from Facebook, but we’ve seen a lot of success there. 

 TV has been I’d say a bigger investment for clients who can afford it whereas Tiktok is a lot better for smaller brands who still have to invest let’s say 30k but the creative is cheaper. So those are some of the biggest ones but unfortunately, this isn’t saying we diversified away to affiliate channels and Facebook is becoming a significantly lower portion of the pie. I have not seen that materialize for anyone that I’ve worked with.

Anish  

That’s interesting. It’s not easy to just say, “Oh, this channel that was performing great at scale with numbers that we were happy about, another channel is just gonna do the same thing even if that’s not nearly as advanced in their technology, doesn’t have as big an audience with the same demographics. It’s just going to replace it.” Yeah, there is some level of wishful thinking within that. So thanks for eliminating that wishful thinking.

Shamanth  

Yeah, I would say we’ve seen a little bit of spend away from Facebook, definitely and it’s very vertical dependent. On the subscription or consumer apps we’ve seen the shift to Tiktok to be much stronger than gaming. We’ve also seen a shift to web based flows for apps. So the apps for which we build web pages or landing pages and direct traffic to that, that’s been hugely effective, even if we just use Facebook. 

Anish  

Can you explain that a little bit?

Shamanth  

We set up a landing page to describe the product, and the landing page has a CTA button that goes to the App Store. The advantage is that the landing page provides a lot more context. So longer copy, which the app store doesn’t provide. So I think that’s one reason why that converts much better now. But again, that’s much more effective outside of gaming. 

Matej  

How did you manage the friction between the App Store and the landing page, because there is the extra stripe step that people need to take. 

Shamanth  

That is true. But at least the metrics we’re seeing seem to suggest that having that extra step is still better than what we would do with the broken tracking with the app install campaigns and with SKAN.

Matej  

I can imagine it works well. Because it uses different optimization methods, rather than just MAI that all other companies are running at the moment. And then this creates way lower CPMs. 

Grace  

We’ve had a lot of success with that strategy at Poppy with that web-based strategy, it’s been awesome for us. We did some incrementality testing just to validate that growth, that we wouldn’t have otherwise seen. So that’s essentially how we’re using Facebook primarily just to drive to our landing pages, and then off to the App Store.

Anish  

That’s interesting. Maybe previously you didn’t do as much of that would have been a strategy always because it is getting people to convert a little bit more and more

Shamanth  

We didn’t need to do that in the past, because Facebook would just target past purchasers anyway, because of IDFA. And if you have made a purchase in a different subscription app, it didn’t matter if you didn’t have the long copy to convert, you are a pregnant person, Facebook knows if you’re a pregnant person you’re going to convert, irrespective of how much information you give them.

Anish  

Specifically with mobile, has there been a return to a lot of these random mobile ad exchanges and networks and trading desks and all of that, where maybe people stopped really using that stuff? When Facebook and Google were doing a better job of pumping these folks or not so much. 

Shamanth 

Some of the bigger ones, yes. In gaming we’ve seen this in the bigger ad networks. And I think that’s primarily because some of these bigger ad networks still use device fingerprinting, which Apple very explicitly forbids, but has not cracked down on.

Matej  

But until that happens, we can see better results on those networks. So it sounds leveraged.

Shamanth 

Exactly. I don’t think that smaller ad networks are anywhere as prominent. But the bigger ones, that’s definitely a place where we diversified spend away from Facebook, because even if Facebook does well, tracking is broken, whereas with fingerprinting I know tracking feels somewhat more deterministic than Facebook is.

Matej  

I moved a lot of spend from Facebook to Google, Unity ads and other ad networks because of the fingerprinting and actually the results I get from these networks. The quality decreased so much in the last year but even before it even happened. Since then I started slowly moving the budgets elsewhere and while it started well, even though I said there are still very profitable channels, Facebook’s iOS campaigns are still very profitable for me. But everything around Android traffic and other stuff, it’s like trying to diversify even more.

Anish  

Yeah, I feel like when I meet people who run these ad networks or work at them, or even publisher side companies who would be selling ads directly, the one consistent thing I’d hear was, well, it’s a really bad business model competing with Facebook or Google, you’re just not going to win. So the ad network model is now really difficult to get anybody excited about and so I’d imagine your inboxes getting flooded with a lot more with some of these folks who are like, by the way, we’re now a better option than we used to be.

Matej  

Thankfully, I have UA consultant on my LinkedIn. So I don’t include all the companies I work with. So in that case, I have zero messages from these guys, which is a blessing. 

Grace  

I was going to say yes, inboxes. Some of the calls I’m taking, which is a big change for the last few years, where probably the last seven, eight years I wouldn’t even consider it, but direct buying actually is something that’s become more interesting to us even at a relatively lower spend level. We have a specific person looking at media companies or content platforms that aren’t super huge, because we don’t need to be huge. We just need to have an audience of pregnant and postpartum people. So I think we’ve been more bullish about paid partnerships. Tiktok when I think we probably would have still experimented, but not looking at it as an experiment but looking at it like this could potentially be a big chunk of our effort and spend, I think that was a change in mindset with our go to market strategy based on this. 

Anish  

That’s fascinating. I also wonder if from the customer side, let’s say a product that you’d sell Grace, which obviously needs to be in a certain life stage to matter to you.

If Facebook and Instagram aren’t as knowledgeable that you’re in that life stage, you’re not going to be seeing ads on them. If a customer is on a baby/pregnancy oriented website, you’re getting a better conversion rate, because they’re not being pumped with ads from other companies who are doing similar things. So the customer experience might actually be pretty good there from not being too inundated.

That’s fascinating. Have you gone through and done some of these buys, or you’re looking into them now.

Grace  

We’ve done at least two and explored some further, and I’ve had some success there. I think especially now where a lot of content platforms are willing to work and they’ve had to get creative in terms of how they’re going to sell media. So now, in this stage where they are a more compelling option, it’s been a nice intersection point where they’re able to buy in innovative ways, because they had to try and remain competitive. And now it’s more interesting to use part of the budget and tests there to be in front of that audience.

Anish  

You probably get access to some of their email lists if you really wanted to, or even content that you can repurpose, otherwise if they’re doing interviews or advertise or advertorial all kinds of things.

Grace  

Exactly. There’s always a package around it that has some legs. So that’s been really valuable for us.

Anish  

So it looks like everyone has somewhat expanded channels, but really is still investing quite a bit in the core paid search, paid social worlds. It’s like there’s a little bit more open mindedness towards ad networks and a little bit more open mindedness to buying directly from publishers, but are there any hidden channels that you’ve not hidden? Maybe like, this is a random area, and maybe it doesn’t have scale, but it performs really well with the smaller audiences that you hear about, micro influencers, smaller affiliate networks etc., that might not give you the scale, but the numbers look good, and you can blend it into your overall broader CAC.

Grace  

We’ve been working with content creators on Tiktok, but that was a good target for us because of the price. A lot of those folks are still emerging. And

there are people who have relatively robust followings and have never done partnership deals before. They’re still approachable. We worked with somebody who had a million followers and it was still a very economical relationship for us. And it was their first partnership deal and they were excited. I feel like Tiktokers are still excited, but less jaded than the Instagram brand.

Influencers that are gonna bid in beauty and all this other stuff before so maybe I’m the jaded one, but it’s been great for reach, impressions and then you can just take that asset and make an ad with it. So we’re kind of getting some value on the asset to that. We’re not having to produce in-house so you look at the cost blended, it’s pretty efficient. We’re getting the impressions from being on their platform that we have an ad we can use and that cost is sort of all blended together. So that’s been done. I think it’s not hidden but it’s been like a flywheel that’s worked really well, for us.

Shamanth  

We do something very similar. We hire actors and not influencers, which is a lot lot more affordable than hiring somebody that has a million plus followers. And because we are able to recruit these actors fairly and cost effectively it helps us build volumes of creatives. We could test different appearances, different ethnicities depending on the product. So definitely, I would say for volumes of UGC/influencer like ads going with actors or micro-influencers has definitely been a huge deal for us.

Evan  

I think similarly. Even just whitelisted pages. We’ve created pages, we’ve had about 1000 content pages that are hyper relevant to whatever the product is and then whitelisted those and those have done extremely well. So sometimes it is the influencer and sort of how well known they are other times it’s just having a different handle right and getting those CPM and or CTR gains. And then the last thing I would say too, is just working on the partnership side, I’m sure Grace can attest

for so many other publications, you’re just buying, they’re incredibly bought in audiences, there is no issue around IDFA. And it is a flat fee. So you’re taking on risk. But if you have the right audience, you’re able to activate them. The ROAS I know for that particular publisher with every one that I’ve worked with is just extraordinarily high. So relying on other brands and having the right placement, hopefully somebody that’s a little bit more editorial and explains the benefits of a product or service.

Anish  

What’s really cool is Grace sounds like she reaches out directly to someone that you found would be interesting for your audience who has a good following who your audience would probably be like this is potentially an authoritative person, is that priced at a pretty flat fee as well? Based on how many followers they have, or it’s all based on other things like likes or follows or anything like that?

Grace  

Content creator pricing is all over the place. So many dependencies, where I’m in right now we can afford most people, even up to a million followers on Tiktok, or even half a million, we can still wrap our arms around it because of the space we’re in. It’s so specific, if we’ve moved out into kids 5+, we would probably completely get priced out. Because we’re able to touch people, if they’re in that life stage, that may not even be the primary thing they talk about, I mean, it may be a part of their story or part of their narrative. So we’re able to have some savings there. But yeah, the sky’s the limit when it comes to content creator feeds.

Anish  

Sounds like you were able to access this person, they’re really thankful and they really were a great partner for you because they were thankful. But when you look at the big YouTubers and the big Instagramers, they have professional Hollywood agents that are negotiating all of that for them. So that’s why I was kind of curious.

Grace  

We work directly. We are being strategic and trying to find folks who haven’t done a lot of partnerships before, that have shared their personal story, connect to the product, and kind of fits our brand ethos, and that we think would be excited about Poppy. They are truly excited about what Poppy is bringing to the world, they’re excited about having maybe their first or second brand partnership, and it’s kind of a win-win for both of us. So that’s how we’ve kind of been able to save on the cost side. So we’ll take advantage of that while it lasts.

Anish  

Not everyone has the energy to go and find these wonderful needles in the haystack on their own. It’s why you go through agencies, platforms, et cetera. But when you do it, it really pays off. And I think that’s advice that I often give to people for anything. If you’re hiring an agency to run your ads, if you’re doing anything partnership oriented, get someone who’s hungry and on the way up and needs to make a few things successful before they can start bumping up their prices and scaling. So, you found someone at the right time. And it’s beneficial for hopefully that person as well because they can start building up their portfolio.  

Matej, you dove into that you’ve invested quite a bit in redoing your infrastructure. Are there any specific details that would be helpful for someone to understand if they were to just start going from relying so heavily on Facebook and Google attribution to now building some of their tech?

Matej  

What’s changed quite heavily for me, it wasn’t about the infrastructure that much, but about trying to think about how to evaluate the activities properly. But in terms of the infrastructure, I work with teams of different sizes. So with small teams, they don’t have the data team. So they rely on mobile measurement partners like Appsflyer, Adjust, Singular etc. With that in mind, they just need to set up the SKAd network framework. 

I worked with different companies before we had several data analyses done just to try to think about how we should actually set it and we went through different conversion schemas. I tested it millions of times and then eventually came up with just a revenue schema. We used the purchase that I mentioned before in the first 24 hours, that was really important. Afterwards, the most important part for me is just trying to think about the evaluation. So the MMP is set up, you just need to think what to set up from my side, the revenue and the purchases for the gaming companies work the best, or the customer conversion schema with some of the funnel events for example, use tutorial completion, then level achieved and whatever else that you can see that the players are actually doing in order to achieve the purchase in the first 24 hours. And then afterwards, you just use the purchase as an event with different conversions or revenue brackets. So you can actually measure the ROAS afterwards.

Anish  

Okay, thanks for that. Anyone else have any kind of insights into different things you’ve done within your infrastructure? It can help some folks who are just making the shift to becoming a little bit more advanced.

Shamanth  

I think incrementality and really investing in analytics to understand the incremental impact of channels. That is definitely an area where a number of folks we’ve worked with have invested. It is something that’s increasingly critical.

Anish  

What kind of analytics tools do you think would be best in class to start looking at?

Shamanth  

There are folks that basically run spreadsheet based models, or they work with third party incrementality tools, or have their internal data science teams build out incrementality models. There’s different directions they could go in, but at the very least starting to think about what’s the incremental impact of a channel, which we didn’t need to do. So it’s certainly something we’re seeing more and more.

Anish  

Cool. We’ll jump to the next one. How has measurement changed since ATT? Do you feel like there’s any kind of broad sense of answering this question that we’re going in a direction we haven’t really dived into just yet? Or you’re looking at different metrics? Are you looking at anything differently?

Matej  

I can elaborate a little bit more about the evaluation. Different sizes of the companies vary their measurement approaches. For example, the smaller size companies, they don’t have the data analytics team, so right now, what we are doing with the small teams is just trying to see what kind of level of spend we can have, and what kind of level of revenue we can get. So we take the blended approach, because well, there is not anything else that we can really do. And then you just stack the UA channels, and you see what is the difference between the spend versus the revenue and then try to assume this is a good channel and is producing some revenues or is not. And then with the bigger teams, you have the incrementality tests and the models that you can use, which is obviously way better or the media mix modeling, which is the really advanced way of measuring the ATT or the iOS traffic at the moment.

Anish  

Well, we’re going to be closing up here. If anyone else wants to get in some thoughts on how they’ve upgraded or evolved their measurement tactics over the last year, we’d love to hear about it. 

Evan  

Most brands that we work with, some of them use it if they can afford it. It is a lot more advanced and also expensive.

But large brands also use really basic tactics, in addition to GA and making sure that you’re getting data from the ad set and the creative and the campaign “How did you hear about us?” I’d say having posts or just a purchase survey, on the confirmation page or an email. Huge advertisers are using it. It’s obviously submitted from an actual customer and you are able to get a much better way through upper funnel channels, have their efficacy, I think it’s become a lot more important. And then yes, I agree that looking at just blended aggregate and trying to isolate how much you’re spending on one particular channel at one given time and seeing, is there marginal efficiency overall for the business, so we’re making these larger changes.

Just that alone. I think that anyone could have a really big impact and have it to scale.

Anish  

Thanks for that. Well, this was an awesome chat. I feel like I learned a tonne. And I’ll throw my email address here. If anyone wants to follow up or have any questions, feel free to throw your email address in the chat.

Thanks, everybody. 

A REQUEST BEFORE YOU GO

I have a very important favor to ask, which as those of you who know me know I don’t do often. If you get any pleasure or inspiration from this episode, could you PLEASE leave a review on your favorite podcasting platform – be it iTunes, Overcast, Spotify, Google Podcasts or wherever you get your podcast fix. This podcast is very much a labor of love – and each episode takes many many hours to put together. When you write a review, it will not only be a great deal of encouragement to us, but it will also support getting the word out about the Mobile User Acquisition Show.

Constructive criticism and suggestions for improvement are welcome, whether on podcasting platforms – or by email to shamanth at rocketshiphq.com. We read all reviews & I want to make this podcast better.

Thank you – and I look forward to seeing you with the next episode!