iGaming Daily

James Ross (00:44.29)
Everyone loves a good discount, whether that's a renowned Black Friday offering that consumes most of the Western world every November or a free money offering in iGaming. It's fair to say on the outset, marketers use these methods as an effective acquisition tool. However, do firms need to focus on their marketing generosity and look to keep it in check? This will be the topic of the latest episode of our CRM Focus series on iGaming Daily, supported by OptiMove.

As I'm joined by Optimoves, Shai Frank, the SVP product and general manager Americas at Optimoves, who produced a four part blog series on the matter. How are doing Shai? okay?

Shai Frank (01:21.538)
Hey, yes, great to be here and great to participate in this great

James Ross (01:27.796)
Thank you very much. What I've really loved about this CRM Focus series, we started off with Penny for the first few episodes and Penny, Penny loves to talk. I absolutely love having Penny on the podcast. We get through two questions, but it's really informative and it's great. But what I've actually liked now is we've started to kind of incorporate more people within OptiMove and just kind of navigate more through your company and kind of what you do in it, X, Y, and Z. And it's good to have you on Shy. I read your blog post, your blog series. Really fascinating, really in depth.

And it made me kind of think out of kind of the box of marketing, which I never really thought of before. And I even spoke to our marketing guys within SBC. And I was like, have you seen this? Like you need to have a look. It's pretty cool. So I'm.

Shai Frank (02:10.52)
Yeah. Yeah. We are, we're pretty proud of, our product and platform, but, recently we, we, we realized we need to start talking more about the stuff that our clients are doing with our, with our system and the great possibilities they present to brands in general, marketers in particular, and CRM and generosity is, definitely a topic that we really think we can

brands and operators take to the next level. So happy to join and geek out on this stuff as much as possible.

James Ross (02:47.77)
I will, you know, geek out away, do as much geeking out as you want. I will implore you to do it. But what I will say to the listeners, I've left links in the description to the blog series of this podcast. I'd advise you before we start to open these blog series, just so you can follow through as well to read what Shai put on, because we're going to be talking about it, but it's going to be a bit more in depth in the article as well. It will help you when we go through

So links are in the description below. Open them. I'll give you five seconds. One, two, three, four, five. That was a quick five. And we're going to delve into it now. So first off, shy, kind of, this is obviously four parts of this blog series. So there's quite a lot of information, but what were your key takeaways from this series?

Shai Frank (03:42.466)
think our main key takeaways is that we know from the iGaming industry that it's very promotional, it's very bonus heavy and brands are using a lot of promotional money to acquire and retain players, which is understandable. I think our main takeaway from this entire series was that there is definitely an enormous opportunity to optimize

both the spend as well as the player experience that comes along with it. So essentially, hitting two birds with one stone, you can really improve your player experience and provide the right promotions and the right bonuses at the right time, right level to both make your players feel that they are being heard and their brand is feeling their need

anticipating their need and provide them with great experience while at the same time significantly improving profit margins and not overspending on over generosity. So this is the great, the main key takeaway from that. And obviously we can share more about the specific tools and the strategies and the tactics that brands can use when they do that and how OptiMove can help them achieve

James Ross (05:10.232)
Yeah. You mentioned in the first part that it's important for marketers to acknowledge generosity as an important first step. Why is

Shai Frank (05:22.54)
I think, you know, people are creatures of habit, right? yeah, so you work in an industry and you're used to doing the things you're used to doing and they become like a nature to you. So, you know, yeah, when I want to acquire a new player, I'm obviously going to run a campaign that has some big deposit X, get Y in bonus bets. It's obvious, right? It's what we've always done.

James Ross (05:27.265)
for that phrase.

Shai Frank (05:52.47)
So I'll just do it. sometimes half of the of a problem is acknowledging the existence of a problem, or even framing the fact that what you are doing is not the greatest way to accomplish what you are trying to accomplish. Just this acknowledgement is half of the way to creating a better result.

If you frame a problem in the product world where I'm coming from, framing of a situation as a problem or as an opportunity is a very important thing to do because then it lets you think, OK, so maybe I can do the same thing, but different players can get different levels of generosity. And all of a sudden, I'm more profitable, which is what we all care about. So I think it's important to, first of all, acknowledge that you are being either

generous in general or over generous in particular or in specific situations. And then once you acknowledge that you can say, okay, is it important enough for me to actually do something about it? And if yes, so let's

James Ross (07:04.302)
You also mentioned predictive models. How can predictive models assist marketers in kind of aligning generosity with shopping behavior?

Shai Frank (07:14.584)
Yeah, that's a big topic. I think when you have a lot of data about your customers and you can gauge what they're doing and what customers like them have been doing in the past, you can deduce a lot of conclusions from that. You can do it manually through reporting and analysis, but you can use machine learning if your data is structured in a good way and you have a platform that allows you to do

like OptiMove, then predictive models can help you anticipate what might happen if you do something. So you can look at the segment of customers and anticipate their future lifetime value based on their characteristics and based on their engagement so far with you, based on their gaming preferences, their betting behavior, their patterns, the times in which they're more active, the types

types of games that they're more active on. So it really helps you anticipate their future value and then also deploy the right CRM strategy for each of those segments. So if you have high value but also high risk of churn group of customers, you might want to focus on that more than on low risk of churn, low value players, right? So, and by value, I mean not just the value that they've

compounded so far, but also their predicted value going into the future. So you might see this is a player or a group of players that have a high likelihood of becoming top spenders with me. They haven't yet. They're still in the beginning of their journey with me, but predictive models show me that they have high potential. So I want to focus on them and retain them and maybe shift some of my promotional budgets on that group of customers versus someone else. Or if I have a predictive model about

chance of becoming a bonus abuser or an at -risk player from a responsible gaming perspective, then you can tame your strategy with them and anticipate the problem before it hits you in your face. So, Bravely Models play a big role there in the overall personalization and optimization strategy.

Shai Frank (09:39.188)
and this is why we put a lot of emphasis on

James Ross (09:42.778)
In your answer, you mentioned machine learning and there's two letters which has been circulating in this industry for the last 12 months and that's AI. What role does AI play in kind of managing and optimizing market generosity?

Shai Frank (10:00.14)
Yeah. So there's a lot of hype around AI, which is great because people in the industry or people on the vendor side of the industry have been talking about AI for ages. since the OptiMove has been providing AI solutions for more than a decade now, AI is becoming commoditized with the prolification of, know, touch BT and tools like that. So now everyone.

every kid can use AI, which wasn't the case up until, I don't know, 18 months ago. But AI has a lot of facets. So machine learning is obviously an aspect of AI and deep learning and large language models that are now powering those ChachiPT experiences are another flavor of AI. We chose to use AI mostly for predictive models as well as

decisioning and optimization. So you can use AI to generate content for a marketing message. And we obviously do that too. But for the context of generosity and optimization, when you use AI for decisioning, can move from a strategy of test and learn to a strategy of continuous optimization, which is a big thing because we all know

running tests and experiments is not an easy thing to do. You have to set up your test, have to set up your experiment, you have to analyze the results, and then you have to draw conclusions from the results and deploy a strategy based on that. So imagine that you wanna test different levels of generosity to a certain segment of players. This is already an advanced mindset. And let's say you're running your

test or experiment in the middle of NFL season, just as an example, So you're saying, I'm going to offer three levels of bonus generosity to this segment of players on a certain campaign or some journey. And you have a diverse group of customers that you are testing this strategy with. Now, it could be that based on your test results, you realize that there is a

Shai Frank (12:24.78)
sub -segment of players that don't need much incentive to be engaged. So you don't have to spend too much money on them. And all of a sudden, so you run this test, you run this, you get those results and you adjust your strategy based on that. So here's a group of players that get more and a group of players that get less based on these test results. Now time goes by, comes February, NFL season ends and all of a sudden this group of players that you didn't give a lot of generosity to,

because they were already kind of spontaneously engaged, their engagement level dropped. And why is that? What happened? So maybe these group of players were, you know, big NFL fans. So as long as NFL season was happening, they were engaging, right? They were playing, they were very highly active, highly engaged, didn't need a lot of incentive. And as soon as NFL season ends, all of a sudden,

You know, they need more incentive to stay engaged. You don't want to wait until next September to acquire them again, to reacquire your existing players, which is a disease we see with a lot of brands and operators. They keep acquiring their customers again and again and again. And the cost of acquisition is obviously enormous. So what happened is that you ran a test, which is a good thing.

and you deploy the CRM strategy based on the test results, which is a good thing. But if you think about it, running your future marketing campaigns based on test results from now is like skating to where the puck is instead of skating to where the puck is going to be. In other words, your test results are becoming obsolete.

as soon as the test is over. Now, when you are using AI to optimize your campaign strategy, it's essentially as if you're doing tests, test results, new tests, new results, a new deployment of strategy every day, all the time, on an ongoing basis. So it could be that you're running an experiment for two or three levels of generosity.

Shai Frank (14:51.032)
And instead of doing it once and selecting the winning strategy, you continuously do so. So with the same group of players, during NFL season, AI will choose a less generous offer to them. But as soon as NFL season ends and their engagement levels start to drop, AI will automatically adjust and divert those customers to the more generous journeys compared to before.

So using AI can help you overcome seasonality like that, changes in player preferences, changes in your catalog assortments. So some games, you introduce new games, you introduce new experiences, your sunsets, some games. This is relevant also outside of gaming as well. So think about it as a retailer.

Why do you change your product or software? So how do I know which products a customer will choose? So this is the same thing. Then it's applicable for gaming as well. So AI can help you just continuously optimize and cater for changes that obviously happen in your player database without you even knowing.

James Ross (16:06.254)
Perfect. And you mentioned journeys and I think that's a nice little segue to the next question. And just for the listeners, you might want to skip to, I believe it's either blog three or four for this. We're going to talk about Optimus self optimizing journeys. Can you just explain to us what that is before we delve a bit more into

Shai Frank (16:20.579)
Yes.

Shai Frank (16:26.786)
Yes. So, like a good product person, will start with describing the problem. imagine as a marketer, you have all kinds of campaigns and experiences that you have in mind. You have all kinds of life cycle campaigns. So, how am I communicating with new players? How am I communicating with active players? How am I communicating with high risk of churn?

reactivation, things along these lines. And you also have a bunch of ad hoc offers, right? So specific events. You mentioned Black Friday in the introduction, but same thing goes with the Euros or Copa America that just ended or Super Bowl or March Madness or any one. So you start lining up campaigns and journeys.

for those events. You have birthday, you have responsible gaming content, you have all of these different mix of offers and campaigns and journeys that each one on its own make a lot of sense, hopefully. But then players are humans that behave in very unpredictable way. And all of a sudden, James, on a certain day, you become eligible to four different

campaigns because you're in the middle of your welcome journey. But it also happens to be your birthday. And it also happens to be on the verge of March madness. And there's something lined up for you because we identified that you are a basketball fan. And for whatever reason, also put you in a high risk of churn bucket. So you're also eligible to some, you know, please stay with us kind of kind campaign. So.

James Ross (18:23.054)
Yeah.

Shai Frank (18:25.026)
What do you do as a marketer? Or what do you expect your platform to do for you in such a situation? One option is to just send you everything that you're eligible to receive, which could be a good strategy in certain situations, but many times it's not because not only it's spamming too much messaging, these are also conflicting.

messages and conflicting offers, very inconsistent experience. So as marketers, you don't want that to happen. So what we see or what we saw happening in the market is either an attempt of marketers to tame the beast and create these automations from hell, that you are creating all kinds of splitting conditions and exit criteria in any of the journeys that you create.

to try to avoid situations where people are eligible to one thing to like eject them from eligibility of another thing. But this is unmanageable as you scale, right? We actually have this Slack channel where we post these automations from health screenshots of journey that people are creating. Look at this crazy thing, what they've done here. And then, know, the CRM manager that built this monster leaves and the

can figure out what's going on there and there's no way to troubleshoot. So that's one thing we see people have tried to do. On the other hand, when they couldn't manage that, they say, okay, let's do frequency capping. So we won't send you four messages on that day, we'll send you only one because you're capped. But guess what? Frequency capping solves for spamming. It doesn't solve for relevancy.

it sends their first message that was scheduled, but not the best message that was scheduled. It could be that your most generic newsletter was scheduled for 8 a .m. in the morning, and so you receive that, and now you're capped. You're not getting your deposit abandonment campaign. You're not getting your high risk of churn. You're not getting your March Madness offer just because they happen to be scheduled for later in the afternoon. That's not a good strategy.

Shai Frank (20:48.78)
So at OptiMove, we chose a different path. We say we want customers to get the best campaign for them at every point in time based on whatever it is that they're eligible to receive. That's what we're trying to do. And the solution is called self -optimizing journeys. So essentially, as an example I mentioned before, James, you're eligible to multiple campaigns at a given day.

What the system will do using AI is evaluate all the different options that may result in sending you each and every one of these campaigns. So if we send campaign A, what will happen? What will happen with James? We know James. We know customers like James, how they previously responded to this campaign A. So we can anticipate or predict. If you get campaign A, this puts you on a path to a journey that looks something like

You will become eligible to something else tomorrow and the next day and so forth and so on and so forth. And we do the same thing. Okay, so what if instead we send you campaign B and what if instead we send you campaign C? And now AI does a much better job than any marketer in calculating all of these different permutations. And eventually AI will choose the campaign today that will put you on a path to maximize your lifetime

and then comes tomorrow and the picture changes completely. You might have called customer support and yelled at them about a promotion that did not activate the way it should have been. So assuming we are integrated to your support system as a brand, let's say it's a Zendesk or something else, we get the signal, hey, there's now a ticket, a customer care ticket open for James. And that changes the entire eligibility picture for you. Now you're all of sudden, we might want to suppress

campaigns that are too much happy, happy, joy campaigns and prioritize something, sorry, you're experiencing something, some bad experience. Here is something to compensate for that. This wasn't in the cards yesterday when we made the calculation, but it is now. So AI will just recalculate the entire journey permutations. And again, choose the campaign today that puts you on a path to maximize your retention rates.

Shai Frank (23:10.976)
and lifetime value. So this is self -optimizing journeys. The end result is that essentially each customer plots their own path instead of a marketer just drawing a journey on a canvas and expecting all the customers to follow those prescribed journeys, which we know doesn't really happen in

James Ross (23:35.002)
Annoyingly shy, we are actually quite short on time. So, and we've got, I have quite a lot more, I had quite a lot more questions, let me say, but we just don't have time for them. I'll try and chase up some of the questions and I'll put them in the comment section of our LinkedIn. But just to round off this podcast, this full series blog post, what is the one thing you would like those readers to take away from

Shai Frank (23:36.14)
Yeah. Okay.

Shai Frank (24:07.832)
think the main point I'd like our audience to take is that there is a huge opportunity for brands, for operators, for marketers to use modern technology to optimize levels of generosity in a way that can produce huge profits to brands while also improving their customer experience.

There is a huge opportunity there. There is a lot of money on the table, virtually or literally. Just use the right tools and with technology today, it's very easy to do that with a few clicks and a few mindsets to make those clicks. You can save millions, and depends on your size, could be even billions of dollars.

on over generosity that you don't have to spend and focus those savings on crafting greater customer experiences and innovative solutions to your customers in a way that will help you differentiate from your competition on your product, on your content, on your offers rather than overspending on acquiring players again and again or overspending

on retaining them where they might stay with you for a lesser offer. I think that's the main takeaway and acknowledging the fact that you might be over generous is a good start. Understanding predictive models and segmentation is probably the next step. And then utilizing AI to move from testing to continuous optimization is the sucker punch that drives this home.

James Ross (26:03.96)
So, shai akadatati for longer than we've got and hopefully we can have you back on to continue this conversation because like I said, there was a lot more we could have got through.

Shai Frank (26:15.042)
I'm happy to join again and thank you for hosting

James Ross (26:19.232)
It's been my absolute pleasure, Shy. Thank you very much. Again, to the listeners out there, I will leave the links to the blog post in the description below if you've not opened them already to browse at your leisure. But apart from that, I've been James Ross. I've been John by Shy Frank. And this has been iGaming Daily. Thank you very