ABOUT MICHAEL: Linkedin | Twitter | Marketing Memtics

ABOUT ROCKETSHIP HQ: Website | LinkedIn  | Twitter | YouTube


KEY HIGHLIGHTS

💡 What is meme-mapping?

💎 How is meme-mapping advantageous for advertisers.

📚 Borrowing memes from different verticals.

🧮 How many tests to run and how often? 

⚖️ Pre-ATT and post-ATT – what changed?

✅ Managing approvals at scale.

KEY QUOTES

The secret behind best selling business books.

He actually did some research to find the top 150 business books that sold over a million dollars in sales and he started to look for patterns. He also noticed some recurring patterns in the title. He said quite often, it was the topic of the book as the second word and the first word that you pair with it would be something that’s not usually associated with the topic.

So in “The subtle art of not giving a f”, not giving an f is not normally a subtle thing. In Atomic Habits – habits are not normally seen as atomic.

A lot more innovation is happening in the gaming industry.

Gaming companies are usually at the forefront of creative testing.

You’re not selling a pair of shoes that’s going to arrive. You’re selling a distraction like a way to entertain yourself. It’s the best business in the world in terms of margins if you get a successful game, but there’s so much competition because of that.

So it’s really a marketing game the gaming people are playing. Evolutionary pressure is going to create a lot of innovation. A lot more innovation is happening in gaming creative testing than there is in slower-moving industries where they don’t really feel that pressure as much.

What is memetic arbitrage?

You can almost think about this as memetic arbitrage. You’re finding a generally popular meme and then building your association of using it as a honey pot really to get your less popular but more commercially valuable memes into that. 

I think in general, innovation happens when you cross-pollinate between industries. Apparently Steve Jobs got a lot of the idea for the sleek Mac design from studying blenders and food processors.

Overwhelm the bottleneck.

There’s a really good book that I read.

I think it’s called The Goal. It’s this one that’s specifically about manufacturing but the same principle applies. You have to overwhelm the bottleneck. So the way we used to do it was with a creative lake. You should be filling up your reservoir of already approved creatives. I think the key is it needs to be modular. 

You can’t say “I want to show this one ad.” You have to say “Can you approve these six stock photos and these seven illustrations?” As long as it’s modular like that then you can always roll out a new variation. You add another copy line that’s one copy line but you could try it with 20 different images. So that’s 20 new variations to test.

Accepting and adapting to loss and grief.

I remember that the day he died and I had to go to the funeral, I didn’t take the day off. I had a meeting with my team. I was smiling and when the meeting was over, I changed clothes and rushed to the cemetery. It was such a surreal experience because one minute I was talking about scaling, creatives to help improve performance and the next I was standing and looking at my friend for the last time.

FULL TRANSCRIPT

Shamanth:

I’m very excited to welcome Mike Taylor to the Mobile UA show. 

Mike:

Thanks for having me again.

Shamanth:

I’m excited to have you back because you are among the few people that have been writing with a lot of insight and experience around creative. I’ve certainly appreciated a lot of what you have written and I’m excited to dive into these in our interview today.

You’ve written about meme mapping – the word meme has a very specific connotation in what you write about and what you speak about, and it’s not just funny cat videos.

 So can you help explain what it means. 

Mike:

So most people don’t realize that the word meme actually predates the internet. It was coined by the evolutionary biologist Richard Dawkins. If you’ve read the Selfish Gene, that’s where he coined the term meme in 1976. 

It’s an analogy to the word gene and what he’s trying to say is, just like your genes can evolve over time and you develop stronger features or functions as a species to help you fit the environment that you’re in, the same thing happens with information. 

A lot of the cultural artifacts that we have, anything literally from ways of making bows and arrows to different words or phrases that we use that have meaning that we might use in a specific industry, all of those things replicate over time and they evolve and you can actually apply the principles of evolution and some of the tools from biology to the evolution of information as well as genes.

Shamanth:

Yes and you’ve also talked and written about how memes are borrowed. 

So applying the idea of memes to advertising, how might an advertiser come up with a meme map or a territory of ideas or concepts, for whatever they’re advertising? 

Mike:

Meme mapping is a way to figure out the application of mimetics to marketing – and I was doing it for a long time. So functionally how it works, a really good example I found recently was, James Clear, the author of Atomic Habits said when he wrote his book, he really really wanted to make sure that it was positioned to be a best-seller. So he did an analysis.

He actually did some research to find the top 150 business books that sold over a million dollars in sales and he started to look for patterns. He also noticed some recurring patterns in the title. He said quite often, it was the topic of the book as the second word and the first word that you pair with it would be something that’s not usually associated with the topic.

So in “The subtle art of not giving a f”, not giving an f is not normally a subtle thing. In Atomic Habits – habits are not normally seen as atomic. 

So he basically did this analysis, noticed that pattern and because he had seen 150 of these books and started to see what was emerging, he could then use that pattern for his book title as well. This is essentially what designers and copywriters do. 

Meme mapping is a way to accelerate that. It’s coming from the perspective of someone who’s data-driven but through running Facebook ads, Google Ads, etc. 

I had to learn how to make creatives. I wanted a framework. 

The way I do it is, I go out like James Clear and build a swipe file of basically all the examples of success in a specific domain that I want to conquer. Then I tag those recurring patterns and think. 

For example you’re looking in the Facebook ad library and you might look at D2C e-commerce businesses. You can spot some recurring themes in the ads that they use.

Once you have that map, the list of tags that either is associated with the successful examples or with the least successful examples. Then it’s really up to you to choose where you go on that map.

It’s not going to tell you to make this decision or don’t make this decision. But it’s really important to know and then you can decide what to copy, where to innovate and where to do something different from the rest of the category.

Shamanth:

Yeah, it’s very important to have some sort of taxonomy, or some systems of tagging into a classification. 

I know you used the word swipe file, that’s certainly been used by marketers and direct response copywriters for over a century now, where they’ve collected ads even before there were digital print ads, and having a system of saying here’s why we think this might be winning.

You also just mentioned and reached out about our creative newsletter which I’m sending out.

That’s definitely built on the same premise which is: can we tag different ads that may be seemingly very different?

For gaming, it can be somewhat easier because there are player motivations, and players seeking progression and relaxation. For non-gaming, you can have different kinds of motivations which could be, does this show the ease of use? Does this show a cute character? 

So I think it’s important to figure out what that taxonomy can be.

Mike:

Yeah, it can be for gaming in general. I am almost envious of you being so deeply embedded in the gaming industry because gaming companies are usually at the forefront of creative testing.

You’re not selling a pair of shoes that’s going to arrive. You’re selling a distraction like a way to entertain yourself. It’s the best business in the world in terms of margins if you get a successful game, but there’s so much competition because of that.

So it’s really a marketing game the gaming people are playing. Evolutionary pressure is going to create a lot of innovation. A lot more innovation is happening in gaming creative testing than there is in slower-moving industries where they don’t really feel that pressure as much.

Shamanth:

One of the trends that came out over the last couple of years in gaming was fake ads. That’s very common in gaming. There are fake ads but there are also ads that are run for non-gaming apps. 

Certainly, a number of wellness apps have what we call soothing satisfying ads. There was a time when every medication app was showing rain falling or a tree blowing in the wind.

These are based on TikTok trends.

A lot of these ads and advertisers are borrowing from different memes that may not be native to their vertical. 

So in those cases, if an advertiser wants to borrow from an unrelated or an adjacent meme, how do you recommend thinking about which ones to borrow from and which ones to avoid? 

Mike:

You can almost think about this as memetic arbitrage. You’re finding a generally popular meme and then building your association of using it as a honey pot really to get your less popular but more commercially valuable memes into that. 

I think in general, innovation happens when you cross-pollinate between industries. Apparently Steve Jobs got a lot of the idea for the sleek Mac design from studying blenders and food processors.

I think in general you shouldn’t just copy memes. Just because something is popular doesn’t mean you should jump on the bandwagon because you can face rejection. 

When you see a brand posting about whatever latest thing is trending and it’s not a natural home for them it comes across as quite fake and lame. I think people are pretty good at spotting authenticity. 

I think in general what works really well is if you borrow memes from another category that your target customer likes already. A really good example is, we did a lot in the FinTech space with Monzo Bank. They were a challenger bank in the UK growing past 3 million accounts. Best rated bank in the UK actually. But they targeted mostly students.

It’s a meme that attracts more attention and memes that get remembered or stand out, persist longer over time, and get talked about.

Shamanth:

Yeah, certainly. I think that makes a lot of sense. 

So once you have a universe of ideas you’re ready to test, how do you figure out how many tests to run and how often? 

Mike:

This is getting harder and harder. Back in the good old days, you used to be able to do thousands of microtargeting audiences and you could split-test everything. 

I started in Google ads and everything was split-testing creative and we had to do the analysis in Excel and stuff.

But today I think it’s much harder post-iOS 14. Especially if you’re a mobile app you’d only get so many slots in terms of tracking through SKAN. Then also there’s a bit of a delay in the data that you get back. So I would say you should actually test your most existentially important things. 

We used to call this performance branding at Ladder, and we would either pre-launch before a company was about to launch a product or even after launch if you wanted to test a new creative direction we would do that.

We would actually make the creatives and we would run proper split tests but it would be at the big conceptual level. I’ll go back to the FinTech example. 

There are people who care about cheap foreign currency because they travel a lot.

Then there are people who care about splitting the bill with friends. So if you advertise the splitting of the bill with friends that might work really really well with students.

It might work terribly with people in my demographic and if you advertise for an exchange that might work actually well with both groups but it wouldn’t work well with a different group and those are existentially important questions because if you’re deciding whether to focus your product on one of those use cases or not, you can actually test that beforehand with advertising and see what the response is.

They actually have big implications in terms of profitability. Forex is very expensive to supply. Whereas a bill-splitting feature, once you build it, is zero cost. People should really get into the habit of using marketing and product together and testing those really existentially important things and being very wise with the number of slots they use in their testing regime.

Shamanth:

I think the pre-ATT world spoiled a lot of us when we thought “yeah let’s just change this one font or this one color and see what happens.” Sometimes we did find statistical significance but it’s just not practical post-ATT to get such granular reads.

Mike:

Honestly, it was always hard to get to statistical significance. We were kidding ourselves most of the time. I ran a growth agency so I was guilty of this as well. We ran 8000 experiments at Ladder and only a third of them actually were statistically significant.

We kept track but most people don’t know. I think now I’m a radicalist on testing – it should always be opinion driven unless it’s existentially important. In this case, you need to put a good budget behind it.

There needs to be a very big change because statistical significance isn’t just a function of how much budget you spend or how much traffic you get or the conversion rate but it’s also like how big the difference is. Like if you’re just testing white colors and borders, Google can do that.

They have billions of searches a day. But you can’t do that if you’re a mobile game. So leave split-testing everything to the big companies, if you’re just trying to figure out what’s the next feature to build or what type of game should I be making.

Shamanth:

Certainly. So what are some of the ways that you manage approvals at scale? 

Mike:

We got blindsided by this at Ladder quite a lot. We started off with small clients, even idea-stage startups, and then eventually we started to pick up bigger clients and we were working eventually with Fortune 500 companies in some cases. 

I ran some Facebook ads for Facebook once which was pretty fun. I was signed off on by Sheryl Sandberg way back in the email chain. I never got to talk to her but, when you have a bottleneck like that, when legal has to approve the ads, when ad approval might take a long time or even just when you have a lot of internal stakeholders that want to give a helpful opinion.

They want to make sure that it’s on the brand, that it’s good, that their ideas have been heard, then I think that you have to optimize everything else in the process around that one bottleneck. 

There’s a really good book that I read.

I think it’s called The Goal. It’s this one that’s specifically about manufacturing but the same principle applies. You have to overwhelm the bottleneck. So the way we used to do it was with a creative lake. You should be filling up your reservoir of already approved creatives. I think the key is it needs to be modular. 

You can’t say “I want to show this one ad.” You have to say “Can you approve these six stock photos and these seven illustrations?” As long as it’s modular like that then you can always roll out a new variation. You add another copy line that’s one copy line but you could try it with 20 different images. So that’s 20 new variations to test.

I think modularity is key. Building up that lake of approved creatives is key. Then just always make sure that you have something ready to demo in the legal review. 

I did some Google ads for Capital One. 

We had one creative approval session every three months and if we missed our session or if  the copywriter didn’t finish in time then we’d have to wait another three months to get things approved. So we made sure that copy was done weeks and weeks ahead of that date,

and when we went in we had more things to review than they had the time to review in the session. 

So we always built up that work in progress ahead of that deadline. So that however much time they would give us we would always like to have something for them to check.

Shamanth:

Yeah, and just the way you mentioned The Goal I think that’s very appropriate because I have also been thinking about applying the theory of constraints to knowledge work. What he talks about is making everything subservient to the bottleneck. Build up a buffer before the bottleneck which are the principles you just outlined and described because the approval bandwidth is so scarce.

If you have approvals once in three months then you just can’t take it for granted.

Mike:

You can actually do a lot of things that would normally be seen as wasteful, but actually in light of the bottleneck not really wasteful.

For example: if I had someone on my team, there’s like a 10 or 20% chance they’ll produce something good but like there’s like at least a 50% chance that we’ll need revisions.

I would not just give that task to one copywriter or one designer. I would actually go on Upwork and find three more. I would actually pay four times as much, just so that I’m sure within those four people I have something usable. 

I knew one of those bets would pay off. Then in the grand scheme of things if you have a $20000 a month client in terms of fees, paying $100- $500 for creative is nothing.

If you show up with bad creatives or you don’t show up with creatives you could lose the 20000.

So while it might seem inefficient to have four different people working on the same task but in light of the bottleneck, it’s actually efficient.

Shamanth:

We do something very similar with UGCs where we have to hire actors outside of our team. The entire project is delayed if somebody’s bailing at the last minute. So we hire more actors than we need, and that’s helped us.

Mike:

One of my clients right now is Recast, a marketing mix modeling tool.

We send out one blog post every week in the newsletter but I actually have 12 weeks of blog posts built up now because I’ve just been building them up and I’m still keeping the deadline of one per week.

Shamanth:

This has been very fascinating and certainly, I think we could do another episode on, “the theory of constraints.”

This would perhaps be a good place for us to wrap up this interview. 

Before we do that can you tell folks how they can find out more about you, everything you do and also your upcoming book. 

Mike:

The two of the concepts I talked about – meme mapping and performance branding, they’re both based on my experience at Ladder and the book is about 80% through editing. It’s called Marketing Memetics.

It’s on the topic of memetics. It includes those examples and it’s the book that I wish I had when we were this data-driven team of creative testing media buyers and we had no idea what to do in branding conversations. We had no idea how to come up with good creative ideas when we couldn’t test as many times as we liked. 

So it’s really the handbook for getting over that. It’s marketingmemetics.com if you want to see that. On Twitter, I’m @hammer_mt. 

Shamanth:

Excellent. We’ll link to your book, your Twitter and LinkedIn as well.

It’s been a pleasure having you Mike. Thank you. 

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!