Scaling user acquisition demands you think big and act bigger—even if you have a small team to manage and optimize ongoing campaigns. Our host Peggy Anne Salz from MobileGroove catches up with Zach van Driel and Adam Turowski, the Co-Founders behind MiriGrowth, an app marketing agency specializing in UA on Facebook and Snapchat. They get past the hype to deliver solid advice (in an edgy way) around how you can achieve scalable app growth through rapid creative testing and bid optimizations—no matter your size or budget.
Transcript
Hello and welcome to Mobile Presence. I’m your host, Peggy Anne Salz, with Mobile Groove, where I plan, produce and promote content that allows my clients to reach performance goals and scale growth. And today we’re going to take that literally because we are talking shop – I’m looking forward to that actually – talking shop with MiriGrowth – it’s an app marketing agency focused on user acquisition on Facebook and Snapchat, and also looking at scalable app growth through rapid creative testing. That’s going to be the topic of today’s show, and my guests today are Adam Turowski and Zach van Driel, two UA specialists and also the Co-Founders of the company. Hey guys, great to have you on the show today.
Yes, thanks a lot for having us, Peggy.
I’m excited. First of all it’s good to have two people because then we can have a conversation and creative testing is hot, it’s something we’re hearing about a lot. I’m hearing about it even as like the core of competitive advantage for app companies and marketers. Is this what kind of level creative testing has taken on?
Yes, we think so. I think this is kind of the key assumption behind our company and what we feel is when we talk about Facebook, for instance, Facebook is really good at targeting, like look-a-like audiences are pretty much the best you can do in terms of targeting and it’s really hard to find anything else. In terms of bidding, they are also really good, at the beginning you could bid on CPIs and they were doing it really well for you. Then on post install advance now, there’s also value optimisation and based on that what we think is that really where, as a company, as an advertiser, you can make a difference is really with the creative.
This is where all of a sudden you can have one creative that can have a click through rate that is four times better than the previous one and that can really change the trajectory for the promotion of a product. And the rest is also important, we just feel that it’s not – it cannot have the same level of, the same magnitude of differences. Would you agree with that?
Yes, exactly.
I was going to ask you, Zach, that makes sense because also apps need to be effective but we’re hearing a lot more about sort of like emotive marketing. You’re sort of clued in on the power of the creative, the importance of testing the creative. What are you hearing out there in the market? Is it something that app marketers are getting their head around?
Yes, I think so. I think more and more maybe a year ago when we started talking about it, it was more as something a bit different that we’re doing or kind of the pace at which we’re testing creative, definitely everyone we speak to now is in complete agreement about having to test enough creatives, being able to actually deep dive into the results and understand what you’re testing, what performs well, what’s performing just below your kind of expectations and what you should iterate on and what you should maybe kind of save to revisit early today. So, it’s definitely something that is definitely like in agreement kind of in the space.
I mean, that’s good, we’re in agreement on this. I am fully, I saw it coming a little earlier, I thought, you know, it’s data driven marketing, when we all sort of have the same data, when we all have our attribution partners and everything in place and we understand the data that’s driving the marketing, then we’re going to have to differentiate elsewhere, you know. When we’ve all done our platforms and we’ve all seen the results, where is the differentiation? That’s very much in the creative, now it’s going to be the question about how to do it and how to do it at scale. I mean, that’s a big question, I won’t pose that one to you but maybe we can look at what you’re telling your clients when they’re just getting started because AB testing as a rule is the way you do it but there are some other considerations and it’s not that simple.
Yes, so I think the way we kind of go about creative testing, as you said we definitely AB test everything. How we split the creative decision or the ideation of what creatives we do, we like to do some exploratory tests, in each test, so that’s exploring completely new ways that we want to advertise the app, so that’s a new selling point or a new visual attached to that selling point. And then within that, we probably have about 70% of the creatives that we’re doing as iterations, so that’s iterating to try and get that small incremental improvement on the creative performance.
What this means in practice is we want to AB test, as you were saying, we want to test 5 creatives each week for each product and with that we’ll look into kind of the volume of data that we’re getting to make sure we’re going to make statistically significant decisions on the creatives and make the decision from there as to how we’re going to iterate and what we’re going to, I guess, roll out. If we find a new winning creative that works well in this testing environment that we have set up, we can then roll that out and scale it to all these performance campaigns, targeting the right audiences, so that’s matching the right creative with the right audience and able to kind of benefit off that, improve performance on the creative at scale, as you were kind of asking. Does that answer your question?
Absolutely. I’m also hearing there are 5 creatives testing – what was that per week, did you say, Zach?
Yes, per week, per app. So we find video ads to be the best performing creatives across the board, we also like to explore other ad formats just to kind of make sure that we’re always testing the right thing and within that, we also test different lengths of videos so we’re really – we don’t want to make any assumptions as to what we think will work well, we just want to make sure we’re exploring broadly enough and also iterating like with enough focus to try and like learn it as we go and also find new pockets of performance because what’s really going to make the big difference for your campaigns is if you can find that one new creative which is completely different to any of the other creatives you’ve been running that performs, as Adam said, sometimes 4 times better than any other creative you’ve run in the past.
And just to kind of add to that, the iterating and exploring new concepts will help you maintain performance so obviously if you’ve found one ad that performs well, it’s going to fatigue over time so people are going to be less responsive to it, click through rates will decline and performance will start to drop off, so if you want to maintain the performance and the level of scale that you’re at, you’ve got to iterate and make sure you’re maybe adding a kind of fresh look or a fresh skin to a creative concept to maintain the good performance.
So, it’s definitely 5 testing – I’ve got loads of questions, by the way, because this is such a hot topic and people are even asking me – I’m going off to a conference next week in Berlin and the question – “So how do you – what’s the scientific approach to testing creatives because we don’t have one really yet?” AB testing sounds good but at the end of the day, I’m also hearing that even if you reject something, don’t throw it away, don’t put it in the bin as you would say there in England, where you, Zach, and Adam are based, right, but keep it because you never know – it may actually be a hit later. It was just maybe not the right time or you know preferences were different. I mean, do you ever find yourself torn when you have an AB test and then it’s like “Is it really, really the one?” because it’s not that conclusive, is it?
I think, yes, I guess maybe a couple of things you touched on there. In terms of whether creative may not work at one point in time and then it could work a couple of months later – that’s definitely true.
It’s a headache right there for marketers because it’s like AB testing, yes, but don’t toss it is what we’re saying, right?
Yes, exactly. I think the way we approach it is maybe try and at least refresh it in some way at a later time when you test it or at least explore like you’re revisiting a concept, you’ve had some time away from it so maybe have some new ideas of a different way to visualise it or some other copy that could go well with it, for example. And also maybe not exactly the opposite but kind of something a little bit different to what you said is also true. If you have a creative that maybe worked a couple of months ago or even a year ago and this has fatigued over time and it’s no longer relevant for your targeting, if you refresh those creatives, we found that that can sometimes also be like a very successful approach to like revisiting a previous concept that you’ve tested and maybe refreshing it with some other learnings that you’ve had over the last period of time –we’ve seen that can like help you find a well-performing creative as well.
And just to add for this, the thing is when you test a creative or when you revisit it six months later, if for instance someone else is working on the same concept, they will also deliver a slightly different ad and maybe just because their approach is slightly different, the performance is going to be completely different. The thing is that when we have an ad that is generating more engagement, we make assumptions on why it’s generating more, a higher level of engagement but we don’t really know why because an ad is like, you know, it’s so many parameters at the same time, so anything could influence this performance.
So, yes, again, revisiting the same concept by someone else at a later point in time can really lead to different results.
Well, actually great segue because time is the question right now, I’m running out of it. We do have to go to a break but listeners, don’t go away because as you can hear, we’re having a great conversation about creative testing which is going to be top of your agenda in 2019 and beyond, so don’t go away, we’ll be right back.
And we are back. Welcome back to Mobile Presence. I’m your host, Peggy Anne Salz with Mobile Groove and we have today, Adam Turowski and Zach van Driel, two UA specialists, experts and Co-Founders of MiriGrowth and right before the break, we were talking about, you know, how to refresh and when to refresh creatives but of course then there’s also the question when do you just throw it all away, try something new, push the envelope? Let’s start with the first, refreshing – what qualifies as a refresh?
It’s a good question, I think even internally we maybe have slightly different opinions on what a refresh is but in general, we like to make sure we’re not doing like refreshes that are too subtle so we’re not thinking about maybe changing like one tiny thing, like changing the colour of the copy. If we’re talking about refreshing a creative based on colours, we’ll try and change a couple of things at once – so, like the background colour, for example, any colours of assets within the ads, so to try and change the look and feel of the ad without really changing the concept, if that kind of makes sense.
The other idea of like a refresh could be changing a small part of the ad, so changing one segment of the ad and seeing if that’s something that could change like, have an impact on performance and by a segment I mean maybe one portion of the ad is showing slightly different concept or maybe split up the ad into three parts, so you’re advertising three different parts of the product or three different selling points. Do you have any other thoughts on that?
To be clear, this would be about ad formats, banner but also video – you would do a similar refresh in video?
Yes, we’re talking specifically about video, this is where we see the best performance.
It’s also the hottest ad format, you know, I mean, I’m seeing some articles that tell me the banner is back but we’ll see about that. What about pushing the envelope, being totally creative but being totally different, Adam – do you have some views on sort of like when is the time to throw it all to the wind and say “Yes, I’m going to charge out in a different direction” – is it something you do with your clients at a certain point or is it when you say “Let’s try something completely different”?
So, what we’re trying to do is to do this on a weekly basis. So, every week we want to test new concepts and the reason for that is you may find one ad that is outperforming all the others but this will probably not last forever so you need to look for another concept that will work and replace the current best performing ad once it drops in performance a little bit. And finding this new concept, it may take a bit of time. All our creative tests are not positive, we don’t always find something that works better. So, this is why every week we need to come up with new concepts and we need to test them.
Now, every now and then, we’ll find something that works, maybe we’ll be right in time for when performance is starting to drop and we need something new to keep performance at current levels or to scale up the spend for a client, or if we’re lucky, we can just try to create a backlog of concepts and we’ll have this backlog and we’ll just release a new ad every now and then when we see that we need it, when we see that performance is dropping.
How long does it – is this an ongoing, because you’re saying you’re trying to do this every week, I’m hearing 5 ads a week per app, I’m hearing innovate, try something different, you know, push the envelope as well every week – what does this look like for an app marketer right now? To put the manpower and the resources, maybe not in the monetary sense but just this sounds like a big – it’s a big ongoing task – what does it mean really? I mean, do I need to have at least 5 people on my team to work with you to make this happen? It’s something that requires some effort. How much effort really?
So, I mean, we’re not talking about you know like your 30 second TV ad and doing that five times a week. These are really short formats, 10, 15 seconds in most cases, sometimes a bit longer but that’s the exception, and we, I mean, on our team, we have two designers at the moment and they are responsible for a few apps and they will do that, yes, they will simply develop – we will define the concepts with them and then they will work on them.
I mean, what we’re seeing more and more and that relates a little bit to a question you asked earlier, more and more people or the teams we’re working with, they also have someone in the marketing team that is responsible for the user acquisition side of things, so they will have someone whose job will be to do just the creatives for marketing. And basically, yes, I think this is the way that the market, the direction in which the market is evolving which means there will be like one or two people on the team that will also do the creative side of things for the app.
How does that work best to engage with you at MiriGrowth? Should I be a certain type of app, a certain vertical, a certain category, a certain size because it probably works best if I can dedicate someone to really interlocking with you and working through these creatives?
Yes, I think really we have loads of different cases where it works quite well, so we have cases where we’re really integrated with a team and we’re not just coming up with ideas ourselves, we’re discussing ideas on calls, we’re talking about feedback, about things we could be doing differently, different ideas we can test but also we’ve had cases in the past where given that like I guess this is our job, this is what we do all the time, it’s quite easy to at least come up with different concepts to start with and once you understand an app and you really begin to kind of test enough selling points, you can kind of snowball from there.
In terms of the different apps, like what’s an ideal like moment for an app to be working with us, it also varies quite a lot, I think, in terms of verticals, as you were saying, we do work with apps from a lot of different verticals. Recently we’ve been working with a lot more gaming apps where we’ve been seeing some really good results with our approach, so this has been kind of over the last six months we’ve been working more and more with gaming clients. We’ve also worked in the past with health and fitness apps, education apps and we’ve seen some really exciting results there. And we’ve also worked with apps from soft launch and also apps that have been around for over two years and are really well-established.
So I guess kind of in answer, there’s a lot of different cases that we’ve had going and we’ve seen good results in a broad mix of them.
Yes, the only thing I would add is the best clients to work with is the clients that – it sounds a bit cheesy but that challenge you a bit, but it really is good to have someone that tells you “Why did you not test this, why did you not test that?” or “If we tested this, maybe the next thing we should test is that thing” because first of all it adds an extra person in the creative process and it’s quite hard to keep on having new ideas constantly, but also it just forces us to have more innovative concepts.
Yes, and I guess in some way challenge our assumptions basically, if you have someone who’s challenging you, you’re like challenging your assumptions or your best practices.
Because as much as we say that we don’t want to have assumptions, we obviously have some form of bias in whatever we do so it’s always good to hear it from someone else that maybe – maybe the fact that video, for example, we do a new video, maybe we should be doing more aesthetic ads or something like that.
I mean, you talk about bias and things like that and I’m looking into that a lot, that’s one reason why people are saying there has to be an element of machine learning here, I won’t go there because that’s way out, it’s also not perfect, I’ve seen some great articles about what the machine learning has chosen for creatives and they were really, really bad. How about you as a person, what are some surprises you can share because we’ve all probably been to those conferences where they say, “Okay, raise your hand, was it creative A or B that one?” and we’re usually wrong. I am at least, and the audience is too. So, it’s a hard one to call. What are some surprises or what can you share along the way? Maybe a lesson or two that tells us how to keep an open mind, put it that way.
Yes, it’s a good question. I think in short on my side, generally the ads which I think are going to work best, don’t, and generally the ads where maybe I’m a little bit less sure of whether it’s going to work actually end up being better performers. So, it’s really hard to kind of make assumptions on what you think will work and what doesn’t although there’s always maybe something underlying that you’ll in hindsight be like “Oh, actually this is related to a previous ad concept I’ve seen before which I didn’t just look at it in the way that I saw at the time”. I don’t know if Adam has a different opinion to me on that one?
No, no, I totally agree. I mean we could often say that – that’s why we don’t really want to have assumptions because we never know – we can never guess which creative is going to be the best performing one and yes, we’re always surprised by the results. I mean, I think, again, there are so many – you could break down all these ads into so many different parameters and we can’t really do that, it’s too complicated but I think that’s why trying to predict which ad is going to perform at the end of the day is something that is very hard to do without actually testing it.
Yes, that’s exactly what I’m hearing. I’m just hearing really test, test, keep testing, keep the ones you reject and sometimes, as one of my friends and colleagues, you just have to get a hammer out and break everything – I thought that was the best advice I’d ever heard because what will emerge from those pieces might fit into a whole different concept. It’s still similar and related to your app but just a little bit different. We have to go to a break, I can’t wait to get back because I’d like to ask you some questions about what I think is going to be important in 2019 and I want to hear what you think is going to be important going ahead as well. So, listeners, don’t go away, we’ll be right back.
And we are back. Welcome back to Mobile Presence. I’m your host, Peggy Anne Salz with Mobile Groove and we have today, Zach van Driel and Adam Turowski, UA specialists and Co-Founders of MiriGrowth. And guys right before the break, we were talking about what we’re trying to do, probably can’t be done, which is approach creative testing as a science. It’s an art, the creative part, science, tougher to handle, probably something we’re not going to crack anytime soon but of course there is best practice – you can offer what you’ve seen by working with your clients, working with app marketers. Let’s just walk through those, maybe Zach, you kick off with one. One really important piece of best practice or a key learning that you can share.
Yes, good question. I think one really important kind of, yes, I guess, lesson that we’ve learned is to try and focus on one message, so when you’re testing a lot and you’re testing 5 ads a week, you can maybe get sidetracked and kind of like either maybe have an ad concept that’s maybe a bit vague or also have an ad concept that focuses on too much or tries to sell the app too much within a video ad. On Facebook and on Snapchat, users really don’t want to pay attention, you need to be able to grab their attention very fast and be able to like communicate your message in a very short amount of time, so having like one key message is a key part of that.
So I guess that the way that would look is maybe having one piece of copy that’s short and sweet and also maybe having one piece of like visual to accompany that, whether if that’s for a game, it could be game play, or if that is like another app, a health and fitness app, for example, it could be showing off one key selling point of the app. I guess that’s probably one from me.
That’s a good one, it’s about focus. Adam, why don’t you give me your second?
Yes, sure. The other one that is really important is we know that first impression is key, we know that people look at an ad for an average of 1.5 or 2 seconds which means that the beginning of the video needs to be really impactful which means that, for instance, you really cannot afford to have an introduction to the ad. So, like the logo of the company just like appearing on the video really slowly for 2 seconds, that cannot happen because that will – I mean, people will just go somewhere else.
And it also means that the main messaging of the ad, this one idea, one single feature that is promoted as Zach was saying, it also needs to be on the screen from the beginning of the ad and it can’t appear after 2 seconds.
So really grab attention but also be consistent, it’s not just like wowing me but really seeing that you’re repeating that message in the video in a different way but being on message.
Yes, I mean, as we said, quite often we have quite short ads so it’s also very difficult to have more than one, to talk about more than one feature within 10 seconds but also yes, it’s true that there’s no – we just focus on one message, we don’t really go from one message to another.
I think these two assumptions, focusing on one message, I think, is maybe something we try to challenge every now and then as one of our key assumptions, we have found moments in the past where if you actually work on a slightly longer ad, so we’re talking maybe a 30 second ad, this can work well in communicating a more detailed maybe app or a game that’s a lot more – there’s a lot more going on that you need to communicate. So this isn’t something we’ve actually found that works for all cases which is why we wouldn’t with that approach for a lot of apps we start with but it is something that we do challenge every now and then and we have found cases in the past that does perform well.
Well, it’s been great having you both on the show. I know you do a lot more, you do, you know, automation of budget allocation, bid optimisations, Facebook, Snapchat, loads of stuff – I hope to have you both again back soon. In the meantime, how can listeners stay up to date with what you’re doing? Maybe you have an excellent blog, you’re doing some stuff on Medium – how do they reach out to you, maybe they just want to ask you and say “Hey, give me some advice” or even “I have an app I’d like you to optimise for me” – how would they do that?
Good question. I think we’re a small company, we’re already trying to work on kind of sharing the knowledge that we also came as doing what we do – we are in the mid- to long-term working on a blog where we will be sharing our information – that hasn’t started yet. In the meantime, if people want to reach out to us, we’re on LinkedIn and also on Twitter as well, and on our website they’ll be able to kind of reach out to us and speak to us directly through there as well.
And we’ll have those also in the show notes – your website again?
Is mirigrowth.com.
There you go, pretty simple. And listeners, if you want to stay in touch with me or find out more about how you can be a guest or sponsor on Mobile Presence, then you can email me, peggy@mobilegroove.com, Mobile Groove is where you can find my portfolio of content marketing and app marketing services.
And that, my friends, is a wrap of yet another episode of the show. You can check out this and all earlier episodes of our show by going to webmasterradio.fm or you can find our shows on iTunes, Stitcher, Spreaker, Spotify and iheartRadio simply by searching Mobile Presence. So friends, until next time, remember, every minute is mobile, so make every minute count. We’ll see you soon.