ABOUT TRANG: LinkedIn  | Atom Finance




ABOUT ROCKETSHIP HQ: Website | LinkedIn  | Twitter | YouTube

KEY HIGHLIGHTS

🔱 Why Facebook shouldn’t be your only channel

🚩 The importance of channel diversification

🌿 There are channel options for every company size 

🎓 Why products which need education are not a great fit for paid social

🏗️ How to rebuild the infrastructure to transition out of Facebook

🧗 Manpower intensive channels take a while to ramp up

🤑 Publishers follow demonstrated success

🚓 Dealing with the potential of fraud with affiliate traffic

💅 You need to counteract churn with product management to balance low initial signals

🎛️ The few but predictable levers for targeting on pre-install ads

💡 How to use mobile web in the face of iOS 14 and ATT changes

KEY QUOTES

How to find channel alternatives for your acquisition mix

Diversifying from Facebook means that you should just look at the channels that have similar structure and advantages as Facebook. Like, channels that have a lot of eyeballs, channels that have a lot of first party audience data, channels that have built really good native digital ad experiences. That tends to mean that you should look to Google and Amazon first.

The way to approach affiliate marketing program

Affiliate programs and CPA programs do take a little while to ramp up because it is a manpower intensive channel, because you do have to build relationships with partners, and you have to convince them to work with you and that sort of thing. It’s really no different than negotiating any sort of business deal. 

There is a ramping up process to onboarding partners

We started out with offering quite favorable payout terms that we would be able to revisit and renegotiate once we were able to prove that we could drive revenue to these partners, and prove that the product that we were offering was something that their users would connect with instantly.

It is surprisingly easy to run CPA payout ad networks

For us, it was really easy to turn on these CPA payout mobile ad networks just because that inventory is there and you just need to be uploading your asset, and then having it run, and making sure that you’re connected through your MMP or whatever.

There is initial effort required for affiliate marketing

The more affiliate websites took just time in back and forth, and negotiation, because that involves content development, training, providing workflow walkthroughs of how the product is being used, and how we can get that particular publisher to speak about the product in a way that resonates with their audience. That’s something that can take a few weeks, maybe a couple of months.

Why commission based traffic has functioning fraud shields

Commission based traffic, I think, inherently comes with built in fraud shields, because by definition, partners are paid out only on genuine sales or conversions. So when we set up our contracts, we do make sure that we have fraud reversal rights in our terms, and those are usually pretty standard. And because of that, since the publishers know they won’t get paid for fraudulent traffic, they have this level of accountability that disincentivizes them to send us fraud traffic.

FULL TRANSCRIPT BELOW

Shamanth: I’m very excited to welcome Trang Dao to the Mobile User Acquisition Show. Trang, welcome to the show.

Trang: Thanks for having me, Shamanth! Really excited to be here, my first time on this show.

Shamanth: Yeah, thrilled to have you Trang, certainly you came very highly recommended. But also, you have a very unique perspective on what we’re going to talk about today. In that, we’re going to talk about how Facebook is not a big part of your acquisition mix. And I’m very excited to dive into what, to me, appears to be a very rare and unconventional acquisition setup. So excited to have you on the show, excited to dive into all of this. So, Trang, what were some of the first signs that Facebook wasn’t likely to be a sustainable channel?

Trang: I can speak to my current and past roles, which have always historically been in the consumer space. So a lot of the times when I’ve joined a company in the B2C consumer space, they are usually only running one channel and spending in the six figure level on that channel, and that channel is Facebook. And so even without the science, there is always a lot of consternation about reliance on the one channel. 

But I think last year in particular was a good lesson for really everyone in this space about the dangers of being concentrated in Facebook. So last year, we had the pandemic, people were staying at home and online more, there are a lot of social injustice happenings occurring that were also causing a lot of different activity on Facebook. All of this resulted in these really wide swings in Facebook’s supply cost and that, depending on what industry you’re in, and what kind of campaign you’re running at the time, either benefited you or really put you at a disadvantage—at least just from a cost and return perspective. 

So I think any time that you’re in a situation where you feel as though you don’t have control, or you don’t really know why, or can’t really do anything about the fact that your spend is getting more inefficient, that’s always a good sign that channel might not be the right channel for you at the time. And then, of course, on top of that, last year with all of Apple’s announcements about what they were doing with IDFA and how Facebook chose to react to that, that was also a large signal to a lot of advertisers that Facebook might not be the nice gravy train that it tends to be for a lot of brands.

Shamanth: Yeah, whenever you’re dependent on a single platform, there certainly can be risks. What were some of the first signs that you saw that: “Look, if we aren’t fully live on Facebook, we can still sustain our growth. There are other channels to be explored.” 

Trang: Yeah, I think most boards of management understand the challenges, or at least the importance of channel diversification. For us, since we’re in the FinTech space, and a lot of us come from a background in finance. If you want to think about allocating your advertising dollars the same way that you would diversify your investment portfolio, it’s pretty straightforward to understand. When I think about how to move away from Facebook, it hasn’t been the same playbook at each place that I’ve been with. Of course it’d be really easy to rinse and repeat the same methodology, but I’ve had to learn through experience that that doesn’t always work. So the playbook really just depends on the stage of the company, how much time that you have to achieve whatever targets you’re trying to hit, the resources you have available to you, and very much the product and the industry that you’re in. 

For example, if you’re a company, your monthly budgets are in the six figure range, you want to move quickly, you’re a consumer brand, and you have a relatively healthy or high potential brand equity and brand identity,

diversifying from Facebook means that you should just look at the channels that have similar structure and advantages as Facebook. Like, channels that have a lot of eyeballs, channels that have a lot of first party audience data, channels that have built really good native digital ad experiences. That tends to mean that you should look to Google and Amazon first.

And then secondly, look into other big media networks and that sort of thing, if you want to stay digital. 

If you have more modest budgets and a relatively young and unknown brand, you might then want to think about allocating your money first towards more performance based tactics, such as paid search and CPA and commission based payout buys, focus on building out an affiliate partner program so you can get the best out of having them generate longer form content and piggyback off of their brand equity and build awareness for your brand while you’re still getting that guaranteed return.

Shamanth: Yeah, right, there certainly are options, there certainly are multiple other channels, even if you aren’t at a humongous scale. Would it be fair to say you guys are not running on Facebook at all, or do you have some proportion of your budgets just now?

Trang: Right now, we actually aren’t running on any paid social. Out of the two playbooks that I just outlined, we definitely went towards the latter one that I highlighted. I didn’t do this the same way that other companies that I’ve been with, but our product is particularly unique. We are in the FinTech space, we do have a relatively complex investment research platform that does require a bit of education and a bit of onboarding, and there’s only so much that I can do to rely on little 250×300 ad spaces. So for us, I did lean quite heavily into trying to build relationships with partners, publishers, websites, content providers, and that sort of thing.

Shamanth: Yeah, if I understand correctly, you guys were running on Facebook when you joined, or at least earlier on. So when you were like, “Look, we are on Facebook, but we really think we need to move away.” How easy was that conversation internally in terms of getting a buy-in?

Trang: Pretty easy, actually, like really easy. Again, really the timing of what was happening in 2020. It always does tend to be the first conversation I have when I join a job, which is: “Why did we choose to go into this one channel? Have we looked into diversifying?” And that kind of thing. But like I mentioned, the folks that I work with are well versed in investment allocation, everyone intuitively understood that concentration of risk was not a good thing. But there always is a bit of hesitation and fear, like: “Oh, if we turn off the spigot, is it going to tank our growth?” and that kind of thing. But in general, if you come from a performance background, you have the reporting, you have the attribution, you’re able to account for every dollar, it’s pretty easy to make the case we can wean ourselves off of this, we can transition easily, there’s a plan for this, and it’s a pretty predictable plan.

Shamanth: Right. Was it an abrupt, overnight thing where you were like: “Right, let’s start testing your channels right away.” What was that transition phase like for you guys?

Trang: For us, it really didn’t take that long, it took maybe a few months or so, but it was a lot of putting infrastructure and best practices in place. Making sure MMP was set up correctly; revamping our entire event logging; and auditing the tracking that we had for the events that we’re posting back to channels; making sure that dashboards and reporting are in place and that sort of thing. But again, depending on what stage company you’re at, and what resources you have, some places can take several months. We moved pretty quickly so it took a couple of months or something like that. But as soon as we have the reporting, logging and data infrastructure in place, it’s pretty easy to turn on other digital channels. And then it’s just a matter of testing and doing your best practices, like allocating the right level budget to certain tasks, setting the targets, making sure you’re optimizing, building your baseline threshold, like CPAs per channel, and then just running your standard media mix.

Shamanth: Yeah. You did allude to how having affiliates, partners and publishers work on a CPA payout basis, that has ended up being a big chunk of what you guys do just now. So how easy was it earlier on when you were starting this transition to recruit publishers or partners who could work on a CPA basis?

Trang: Yeah, that is a really good question, because

affiliate programs and CPA programs do take a little while to ramp up because it is a manpower intensive channel, because you do have to build relationships with partners, and you have to convince them to work with you and that sort of thing. It’s really no different than negotiating any sort of business deal. 

We started out with offering quite favorable payout terms that we would be able to revisit and renegotiate once we were able to prove that we could drive revenue to these partners, and prove that the product that we were offering was something that their users would connect with instantly.

Your first phase is always pitching yourself, negotiating terms, and then really within, I’d say, two or three months—because we are able to drive performance for these partners—then the domino effect happens. And then once one major partner starts talking about you, the others will come to you and then it all just really works out. But again, it is all very much based on you having a product that is something that can be easily marketed by these partners.

Shamanth: When you say partners, what might be some examples of the kinds of partners these folks are?

Trang: On the affiliate side we have three categories of partners. We have content partners, the folks who are running the blogs, the how-to sites. Again, we’re an investment research product so there’s a whole host of those types of publishers who have YouTube channels and websites on guidance for investment and finance. 

Then there are other categories of partners like loyalty and rewards partners. For example, credit card rewards portals and that sort of thing, where these are huge audiences of folks who are directed to certain loyalty sites and incentivized to make purchases through the sites because they might get certain rewards or that sort of thing. 

And then there’s always that category more applicable to retail, but there’s always the category of the coupon sites. The CouponCabins, the RetailMeNots, those are always good to, as a best practice, maintain relationships with just because if you have a relationship with those folks, you have more control over how your brand is represented on those channels. You can control promotion leakage and all that sort of thing. 

But on the affiliate side, that’s pretty much how it buckets out. I’m always surprised, there are always actually a lot of networks and agencies that are running some type of either arbitrage strategy or something that just have a lot of inventory that they bought, that they’re then willing to turn around and sell to you on a CPA or payout basis. I do run on quite a few of those in the mobile ad network forum.

Shamanth: Sure. So there were individual categories of publishers that you described, there are like networks. And in terms of the first steps you took, when you started branching into these channels, did you start testing all of these categories? Was there one that you started off with? What did the first steps look like?

Trang: The first steps are really just who can we activate as quickly as possible. And so, again, that will really just depend on what company you’re at, and what your product is.

 

For us, it was really easy to turn on these CPA payout mobile ad networks just because that inventory is there and you just need to be uploading your asset, and then having it run, and making sure that you’re connected through your MMP or whatever.

But

the more affiliate websites took just time in back and forth, and negotiation, because that involves content development, training, providing workflow walkthroughs of how the product is being used, and how we can get that particular publisher to speak about the product in a way that resonates with their audience. That’s something that can take a few weeks, maybe a couple of months.

But again, you’d want to have many pots on the skillet. It’s all about juggling all the different types of partners that you can activate and keeping in mind each of their individual timelines, and just making sure that you’re managing everything so that there’s a steady flow.

Shamanth: Certainly, certainly. And with the affiliate traffic there, I imagine, is the potential of fraud. So how do you address that?

Trang: The thing about affiliate traffic I find is, it’s kind of like one of the oldest digital channels. I don’t even know, it’s probably just as old as paid search, maybe even older. But all that means is with that channel, the idea of fraud in that channel has existed for quite some time. So there are a lot of solutions in place to address that.

Commission based traffic, I think, inherently comes with built in fraud shields, because by definition, partners are paid out only on genuine sales or conversions. So when we set up our contracts, we do make sure that we have fraud reversal rights in our terms, and those are usually pretty standard. And because of that, since the publishers know they won’t get paid for fraudulent traffic, they have this level of accountability that disincentivizes them to send us fraud traffic.

That’s the first part. 

The second part is there are a lot of like real time fraud monitoring and compliance services that we do activate, at least for the mobile ad networks. Those are fraud detection services that can really easily and quickly determine and identify bots. So we do have that in place. 

I also do a month in audit, every month, of every single conversion that we get. We actually have our own internal fraud script that we run across all of our users, as well. So there’s that third layer that I’ll do every month to still catch anyone who the prior two methods didn’t catch, and then we reverse those payments.

Shamanth: Certainly, certainly. Another risk, I would imagine, would be, getting this from a CPA basis, you’re paying for the first action but not necessarily for subsequent actions, like you’re not getting to retain users, you’re just getting that first transaction. So do you think about the risk that the downstream retention might not be there, this person might just make the first transaction just never come back? 

Trang: Yeah, that is definitely a risk. And going back to the answer I keep saying, which is, I think I’m just really lucky, because this strategy, I don’t think would have worked anywhere else that I would have been. But we are a subscription product. So that in itself means it’s actually pretty easy for me to optimize for the finality of a subscription signup event. we just have to make sure that we’re doing the lifecycle marketing and management work to keep that user continuing every single month beyond the initial subscription. We do have that in place. 

But for the new channels that I’m ramping up, that might not have enough signal, might not have that event being sent passed back to them. Again, I do benefit from having a product that offers a huge wealth of features, which means they’ll have a lot of user actions and events that I can post back to channels and use as predictive and indicative of a high value user. We have a pretty big, really good data team in place and so I do make sure to just regularly refresh with them our predictive user models. If I can’t optimize off of the subscription signup event, I always have a pretty big list of other events I can use that I know that I can optimize towards to get the type of user that I want.

Shamanth: Sure. When it comes to scaling publisher based traffic, affiliate traffic, how do you think about that? What are some of the levers that one might employ?

Trang: Like I mentioned before, it can take a while to ramp up, but you just have to be thoughtful about making sure that you’ve covered all the different types of partners and how they operate and that sort of thing. As long as you’re keeping that in mind and once you start to hit some traction with a select few partners, the rest of them will come a lot more easily. And that’s very much the stage that we’re at right now. 

Right now, my thinking is more, with this channel, it’s not like it can just scale limitlessly. At some point, we do hit a wall. Thankfully, we’re not there yet. But I think for me, that’s really more of the medium term challenge that I’m trying to think about right now, which is how do I measure when I know this has fully scaled out, at which point I will need to think about redefining my channel mix completely.

Shamanth: Yeah, certainly. And you did mention the last time we spoke that one of the channels that is attractive for is pre-install or pre-load ads. So it would appear that there isn’t a lot of control over targeting of these ads. So how do you think about optimizing this? So what, if any, are some of the levers that may be available there?

Trang: Surprisingly enough that there aren’t a lot of targeting levers, but they actually are enough for me to actually work off and build successful campaigns. Most of them right now are where you can target off of the device type—like the actual device model—and then demo, age and gender, and that sort of thing. Some folks are actually even offering interests, too. But those three I found, at least again, for our product, have been pretty predictable.

I cannot stress enough, it’s been a lot of fun for me to market for Atom because I’ve just encountered all these things that I’ve never encountered at any other roles that I’ve been at. But it makes sense, we are an investment research product. So it does correlate quite highly to the type of device that you’re using, the age that you are and the gender you are. And so those are three things that we definitely found within data science that I’ve been able to use.

Shamanth: Certainly. Also with the publisher or affiliate based traffic that we were talking about earlier, I imagine a lot of that are web based, right? They could be on mobile web but web still. So in that case, how, if at all, are they getting impacted by the upcoming ATT changes in iOS 14.5 and beyond?

Trang: You’re right, I do benefit because we do have a web platform. And I have been, over the course of the last several months, moving more towards promoting our web product just to mitigate the impacts from the iOS 14 privacy and ATT changes. Again, unique to this product in particular, I am not going to be as affected as the mobile-only brands and companies. 

But like for everything for mobile, I’ve been trying to do what’s been communicated as best practice over the last seven months or so, like updating all of our SDKs to make sure they will be compatible with SKAdNetwork and making sure that the campaign structures and onboarding events will be compatible and that kind of thing. But again, I am very lucky that I do still have web to rely on—although who knows what’s happening to cookies and browsers there. Even just this morning, I was scrolling through your list of shows, just so that I can get as much knowledge as possible to make sure I was doing everything that I needed to be doing on the mobile side to prepare for this thing that apparently is happening this month, for sure.

Shamanth: Maybe this week! Do you get mobile web traffic to your app at all? And is that flow impacted by ATT and if so, how?

Trang: Actually, no. Mobile web, the way I manage it—I don’t love how I manage it; but we do have a mobile web signup flow. But we don’t have a mobile web product. I do run the mobile install banner on mobile web that has tracking attached to it that tries to direct as many people into the app as much as possible. That’s worked for me so far. I run a script on mobile web that captures all the UTM parameters that I possibly can. We’ll see how that works once all this changes.

Shamanth: Right, so the publishers have a link that sends users to your mobile web page, and there’s a banner on your mobile web page that sends people to the App Store. And you think having that banner should not be a problem because it’s your own website? Help me understand that.

Trang: The banner is actually connected to our MMP right now. It’s the AppsFlyer banner; it’s the smart banner. So they’ll continue to work as well as our MMP continues to work.

Shamanth: Right. And you use that for attribution for which publisher sends the user. Is that accurate?

Trang: Yeah, we use that for attribution for anyone who goes to mobile web.

Shamanth: I’m trying to understand, the publisher sends the user that goes to the mobile web page, and the user clicks on the banner, goes to your app. How does your OneLink link capture which publisher sent the user? How do you figure out which publisher sent the user?

Trang: That’s all grabbed through the UTM that is sent through from that publisher. So that’s passed through the smart banner back into the MMP.

Shamanth: I see. So the publisher link has UTM parameters which are passed through the OneLink banner. I see. Interesting, so it looks like it will not be impacted by the ATT at all because the OneLink banner from your website, that doesn’t change. The linking from the publisher to your website doesn’t change. Would that be accurate?

Trang: I would hope so.

Shamanth: I see, good luck.

Trang: Currently, if a user is on a publisher’s mobile website, the experience is to send them into the App Store. I wasn’t sure if the understanding was that everyone who visits us on mobile is getting sent to mobile web, because that’s not the case. It really depends, but for the most part, they go straight to the store.

Shamanth: Fair enough. And to also take a big step back, you described how you guys have tapped into this world of publishers and affiliates, and really leveraged that to the complete exclusion of something like Facebook. So what kind of products is this kind of an acquisition a good fit for? What kind of products is it not a good fit for?

Trang: It obviously works really well for us. What is it about us that’s worked really well? We are in the consumer space and we built something that actually has really, really, really great product market fit—definitely more so than most other places that I’ve been with. So when you have that, it just means that you have to do less work convincing partners to promote your product and it’s really easy to prove that performance once they do start, like I mentioned, once you do start driving revenue to them.

It works well for products where it’s really necessary to build trust and authority in your brand. This is pretty much the case for all consumer products. But I think FinTech in particular is important here, just because if you’re a FinTech company, you’re most likely building a solution that impacts people’s personal financial well being, their assets, their net worth, their wealth, and that sort of thing. So the more credibility that’s out there in the world that other people have established for you, the more effective this type of strategy will work for you.

Shamanth: Definitely Trang. I know we are coming up on time but this has been incredibly instructive, especially since you have an acquisition mix that is very, very rare. And I can see the kind of work that went into putting this together. So certainly, I’ve learned a ton and especially this is going to be more relevant as people seek to diversify away from Facebook, as people seek to adapt to the ATT changes. So thank you. 

This is perhaps a good place for us to wrap, Trang. And before we do that, can you tell folks how they can find out more about you and everything you do? 

Trang: Sure, you can probably look me up on LinkedIn. And if you want to take a look at our company website too, there’s all the details at Atom.finance.

Shamanth: Wonderful. We will link to all of that in our show notes as well as to your socials, Twitter if you have one. Excellent. Trang, thank you so much for being on the Mobile User Acquisition Show. 

Trang: Great, thanks so much.

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 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!