Every marketer wants to tailor their experience to match what customers want, but personalization can often be more about what marketers want to show than what customers want to see. What if you could replace outdated A/B testing and adapt your content based on user behavior using reinforcement learning?
Welcome to One Amazing About Eikona. Today we’re talking with Nir Weingarten, CEO & Founder at Eikona and he will be sharing one amazing thing about their platform with us today.
Watch the Episode
About Nir Weingarten
Resources
Eikona: https://www.eikona.io
We’re proud to be a media partner for #MAICON26 – Oct. 13-15! Learn how AI can power your marketing and business and help you grow smarter. Use code AGILE150 to save! https://aglbrnd.co/r/7fe458ced0f04658
Reach your customers with Reddit. Spend $500 in ad spend, get $500 back in ad credit! Learn more: https://advertalize.com/r/491818c79fb1873f
Chaser is the only Slack-native project management platform that helps teams turn messages into tracked tasks, automate follow-ups, and maintain team-wide visibility, without adopting another tool. Now integrated with Claude and other GenAI tools. Learn more at trychaser.com and use code AGILEBRAND for a 3-month free trial (normal trial is 14 days).
The most influential minds in software, AI, and engineering leadership will be at WeAreDevelopers World Congress North America, September 23-25 in San Jose. Learn more: https://aglbrnd.co/r/60a7299222a7bcf1
Transcript
Greg Kihlström (00:00)
Every marketer wants to tailor their experience to match what customers want, but personalization can often be more about what marketers want to show than what customers want to see. What if you could replace outdated A-B testing and adapt your content based on user behavior using reinforcement learning?
Welcome to One Amazing Thing About Econa. Today we’re talking with Nir Weingarten, CEO and founder at Econa, and he’s going to be sharing one amazing thing about their platform with us today. Nir, welcome to the show.
Nir Weingarten (00:36)
Hey, Greg, it’s a pleasure being here.
Greg Kihlström (00:39)
Yeah, looking forward to seeing this in in action in a minute here. Before we do though, why don’t you give a little background on you and your role at Eikona
Nir Weingarten (00:47)
Sure. So my background is AI research, started in academia, doing my master’s degree in machine learning, then in the industry as a data scientist. Have a lot of passion for AI and automating processes with AI. In this specific domain, creative processes that have some aesthetics into them. It’s something that’s just so much fun to work on. And we founded Icona a year and a half ago.
VC backed company, HQ and Tel Aviv. And we’re helping retention marketing teams adopt a new type of business process in the retention marketing and process, which enables it to scale and to be much more monetized.
Greg Kihlström (01:32)
Right, great. and so we’re gonna take a look in a second here, but before we do, can you talk a little I know you touched on it just then, but can you talk a little bit more what are we gonna look at today and who’s the audience and and why is it important to them?
Nir Weingarten (01:47)
Sure, so we’re gonna see how this process actually looks like in production. the audience is retention marketers, and but it can I think be very interesting also for any marketer, especially performance marketers. And the situation today in retention marketing is that the process mostly looks like this: you have a marketer, they write a brief, they hand it over to a creative person and a copywriter sometimes, iterate back and forth until they get it.
to the final draft, then the marketer needs to take that somehow insert that into their marketing automation platform, add the links, schedule a campaign. That takes a long time and it has two major bottlenecks that we’ve identified. The first one is creating the different creatives, which inhibits us from scaling and testing. The second bottleneck is actually operating the marketing automation platform and
Sending these all out. So obviously you waste a lot of time with your team, but the second thing you do is you end up sending out blasts. And you know, this is not the case, for example, in paid media, right? In paid media, we have an algorithm picking out different creatives, sending them to people based on their intents and their engagement.
But in owned Media, which is such a major revenue driver for B2C companies, the cheapest media channel that we have, and somewhere we it’s a place where we have a lot of first party data on and we still send out blasts, and we think this makes this channel very much under monetized.
Greg Kihlström (03:26)
Right. Yeah. Yeah. I agree. Well yeah, let’s let’s take a look. Well let’s see it in action here.
Nir Weingarten (03:32)
All right, let’s go. So we talked about the first bottleneck being creative production, right?
What we’re gonna see now is basically the first step of this new process, you start off in Slack. And so you have a Slack channel and you can talk to our bot there. The bot is called Ike. Now, Ike has connected to your existing marketing automation platform.
Has learned your brand, has learned your different SKU, and is now ready to help you automate the process of creating content. So, what you’re gonna do is we’re gonna call Ike, and then we’re gonna give him a command called Fullboard, basically meaning create an email from scratch. And for the purpose of this demonstration, we’ve created this fictitious brand for handbags.
So let’s say we want to create an email for a 50% discount for one of the products in this fictitious brand, the City Toad. And we’re gonna add it in a brief. So we’ve prepared something here beforehand. this is a brief like you would give a creative team saying, you know, what you would like to focus on, etc. And we’re just gonna send that to Ike.
Ike responded, now he’s starting to work. He’s gonna get back to us in a minute. And I would just say until he gets back that this is basically what’s happening now behind the scenes is that Ike is working inside of Figma and immolates the process of a creative marketer and breaking out different intents for your product.
And then composing this email from scratch also in layers. Here we have it back. Thank you. let’s see what Ike cooked.
So this is our fictitious brand, Marelle from Paris. you have a complete email together with your hero shot and your email body. This is the product to its details. It’s not a real product, but it is conserved between the different presentations, and this is practically good to go, but you know, let’s say we want to.
edit something here, right? So I can ask Ike to change something, like let’s ask Ike to change the hero image.
So, one of the key constraints we we’ve heard from people using AI to automate creative processes is that it’s really hard to get the brand right and the product right. And then it’s really hard to insert edits. So one of the things we’ve did here is tuned models over our partner’s specific brand to get their products and brand right, but also made our agents work inside of Figma.
Like you, you’re seeing only my Slack. So I’m gonna share my Figma later. So this is a link to your Figma and you can actually edit their the email in layers. Yeah.
Nir Weingarten (07:13)
So here go, and we’ve edited the hero image. Now there’s some people in the background. Everything else remained the same. and this is how we ask for edits and create emails. basically from scratch, I would like to ask you, Greg, how much what do you think that you know marketers and creative teams could do with this time now that they don’t have to you know deal with this manual work?
Greg Kihlström (07:43)
Well, I mean definitely and I mean it definitely saves a lot of time. I mean, I think you get to think more strategically and you get to think about more potential options, right? Instead of trying to get one email right, you can you can think in terms of more segments or more personalization and then things like that too.
Nir Weingarten (08:01)
So yeah, that’s exactly what we hear from our partners. So this is one of the things they love the most is you know that feeling that that they got their time back and also that they now have time to do the more strategic things like planning, looking at the wider market, thinking like three, four months ahead. It’s exactly that.
Nir Weingarten (08:23)
So the second thing I’d like to to show is that you know, we covered now how we create emails from scratch.
And this is by the way, this is also created as a template inside of your marketing automation platform, together with all the links and everything needed. But the second bottleneck that we’ve discussed is actually setting this up as a campaign for different segments and maybe even trying to test and understand what worked. so what we’re gonna see now is how we’re gonna automate that process as well and this is going to be also through Ike. So we’re gonna call Ike and we’re gonna say schedule.
Alright, let him think for a second. So, what Ike does now is it sets up a campaign on your existing marketing automation platform, and but it’s gonna set up what we call a dynamic campaign. Meaning it’s gonna take that one piece of email that we’ve created and span variations out of it that try to explore different purchase intents and different messages.
So Ike’s gonna look at past performance and stuff that he knows that worked in the past, but it’s also gonna look on what your competitors did and what did people write on your Shopify comments and different trends that it saw. It’s gonna compose a bunch of variations for you to approve such that we can test them at scale when we send out the content. So it’s gonna take a couple seconds.
Yeah. There you go.
So this is a link basically to the Icona app where we actually control the dynamic campaign. So I’m gonna click here and I’m gonna reshare my screen.
So here’s an example.
of our campaign that we’ve just scheduled, right? So this is what we call a dynamic campaign. We we can see it’s scheduled, and this is the control version that we’ve approved, right? Together with a subject line and preview that we can edit. but below it we have variations that are pending our approval as marketers or brand creatives.
Each one of these variations basically is based on the original email and brief, but tries to poke at different purchase intents and messages like we talked. And so you can see that the product is displayed differently. And here we have a piece of social proof that it probably
Nir Weingarten (11:21)
took from the website. And we can talk try to poke on FOMO, we can try to poke on some product attribute. And if you want to understand what Ike did.
And we can click on this brain symbol and Ike would explain, right?
yeah, that’s that’s cool and you can also edit. So you can leave a comment to edit for Ike like we signed a second before, and you can also open this in Figma.
Cooler. Right. So yeah, this just comes in layers, right? So this is not just one image, it comes in layers. You can just edit the text manually if you’d want. And once you’re satisfied, you can just select the variations that you’d want. And now you’re ready for your dynamic campaign. That once sent is actually going to look like this. So here’s an example.
Of a of a sent campaign. So this is a different campaign that’s already been sent. And what you see here is the outcome, the performance outcome of this dynamic campaign, when you have the different variations that you’ve approved, right? Competing and through the algorithm for different populations, using your first party data. And here you see this like
Plot of the algorithm running through time allocating traffic dynamically. And at the end of the day, we can compute very measurably the amount of incremental uplift and clicks or in revenue that this smart allocation of traffic actually produced. And we can tell our partners hey, this dynamic campaign produced a 30% increase in RPR or in CTR and or in the any other metric.
from our experience, this is anywhere between twenty to forty percent uplifts
Greg Kihlström (13:18)
wow.
Nir Weingarten (13:18)
in revenue that we see consistently and across time compared to a control.
Greg Kihlström (13:25)
Wow. Wow. That’s great. Yeah. Yeah. I mean, this is the definitely definitely qualifies as amazing in my book. So I mean a few things that stood out to me. I mean, you know, first of all, just the the way that you can interact with it using an interface, you know, s a lot of people are used to using Slack or, you know, things like that. So, you know, a very natural way of interacting with that. But yeah, I mean the the Figma integration as well is really interesting because I think a lot of the outputs
That we tend to get to kind of to what you were saying are just it’s a flat file that if either you like it or you don’t. And then it’s kind of this regeneration process of you kind of cross your fingers and hope for the best when it regenerates that it that it did what you needed, but actually being able to to interact with those layers if if you want as well. And then of course the you know, I thought it was really interesting to see the transparency on, you know, creating variations, but also, you know, you flip the the tile back and actually see why it’s doing what it’s doing. so yeah, I I lot lots of lots of interesting stuff there. And I I would imagine it’s you know, it’s helping with the collaborative process as well, because people can really see can start to see the the thought process behind it, right? And and start to again back to the your earlier question, it enables people to think more strategically about what they’re doing, right?
Nir Weingarten (14:45)
Yes, and can also span to other channels and the organizations. So if you find a winner
Greg Kihlström (14:48)
Hm. Yeah.
Nir Weingarten (14:50)
on email, you can take it either as a retargeting campaign or reach campaign or even an acquisition campaign because maybe Ike will tell you, Hey, this worked really well for your prospect. So just take it as is to paid media.
Greg Kihlström (15:04)
Yeah. Yeah. Love it. Well, yeah, this is great, great to see this in action. would love to would love to get my hands on it for a couple of clients I work with myself. But for those that want to learn more, what where should they go?
Nir Weingarten (15:17)
Well, I recommend visiting our website. It’s ikona dot io. That’s e-i-k-o-n-a dot io.
Greg Kihlström (15:25)
Great, great. Well again, I’d like to thank Nier Weingarten, CEO and founder at Icona for joining the show. You can learn more about Nier and Icona by following the links in the show notes.
