How to Create Content at Scale with Satej Sirur

Hi, I'm Michael Hartman. I'm Naomi Liu. And I'm Mike Rizzo. And this is OpsCast, a podcast for marketing ops pros and RevOps pros, created by the MoPros, the number one community for marketing operations professionals. Tune into each episode as we chat with real professionals to help elevate you and your marketing operations career. Hello and welcome to another episode of OpsCast, brought to you by marketingops.com, powered by all the MoPros out there. I'm your host, Michael Hartman, flying solo today. Joining me today to discuss how to create content at scale is Satish Surer. I know you told me how to do it. I think I still butchered it, so we'll get there. Satish is founder and CEO of Rocketium. Satish is passionate about startups, technology, and the positive impact startups can create with technology. He started Rocketium in 2015 after a decade of experience spanning multiple roles in tech product business across multiple geographies, India, US. I think you're in the US now at startups like Rocketium, Taxi for Sure, and at trillion-dollar companies like Amazon and Microsoft. So Satish, thank you for joining me today.

Glad to be here, Michael. Thank you for having me.

All right. Well, what I'm, what I'm Audiences knows, like, I try to, try to get, make sure I pronounce names correctly. And I think I, I failed even though we, we tried it before this. So do what we can, you know? All right. Well, so you, let's get into this right away. So you founded Rakuten in 2015. And I'm curious, I'm always curious about, you know, the founding of these companies and things like what led to this, the start of that. Was there something, you know, a problem, an opportunity, a challenge that you saw that inspired you to start the company?

Yeah, so I really wish there were a real problem and an insight that I had that somebody somewhere has a problem that I want to solve, and that is why this company needs to exist. But as it happened, this was a pet project that I had in my head for a decade. And when the previous company where I was working got acquired by a competitor, I was flush with cash. I had no ideas about what to do next because for the few years before that, that company was my whole life. So I didn't really have options. So the only thing I knew was this pet project that I had in my head. I said, let's breathe life into it. And no real idea of who wants it, except I want it to exist. That is how Rocket Team started. But luckily, in a couple of years, we realized some people have a problem. We tried to solve that. Then we realized another set of people have a problem. But 5 years ago, we got into the space where we are today. And thankfully, since then, we have not changed direction, no more pivots, but we are very excited about the space in which we are operating. Large teams, marketing, creative, trying to solve problems for them that are not solved by the 3,000 marketing software that exist, not solving it. So that's why Rocketeam exists.

Oh, I think it's more than 3,000, but yeah. No. So it was a bit of a Field of Dreams project, right? If you build it, they will come kind of thing.

That was the mistaken view with which I had started. But yeah, I learned my lesson the hard way.

Well, but still, I mean, I think there's something to be said about how innovation can come about from something like that. All right. So you've called what your company does creative ops at scale, right? So First off, so as you know, most of our audience is in marketing operations. So I'd love for you to sort of define what you mean by creative ops and how it intersects or relates to marketing operations as well.

Yeah. So as the folks who live and breathe marketing ops know, it is very hard to describe what you do to your mother because it's all the little things. It's all the stuff in between. It's the cement, it's the glue, it's the intangibles that you're trying to solve for, right? Ultimate goal is very clear what you're trying to do, but the The way you do it, the tools that exist, it's sort of this in-between worlds kind of stuff. So that is really what CreativeOps is. The core problem that we are trying to solve for our customers is that on a day-to-day basis, they are working on getting the right content out so that they can run campaigns that reach people. Eventually, the goal of any marketing content, whether through creative or anything else, is to persuade somebody to your point of view. And in our case, it is images, videos, GIFs, that sort of content that has to be made. As we work with large agencies and large brands, they have diverse audiences across regions, people with different preferences, many channels in which they need to reach these people. So the scale of content really goes up. The number of teams and people and voices that have to be involved in this process goes up. And all of them are on different tools. The tools that they have for the job, they are best in class but not really built for the ops of it, which is all of these teams have to make content at scale, give it to somebody else to do the reviews, incorporate all the feedback, check for platform guidelines, brands' own best practices, and then analyze and say all of these thousands of decisions that we made, which of those is working, which of those is not working. Now they are doing that in 6 or 7, 8 different tools, passing stuff to each other. And that is what Rocketeam is trying to solve. How do you make sure that marketing and creative teams are able to put out the right kind of content across channels to the audience, do it quickly, do it without too much pain, and do it without too much spend?

So, I mean, up until you talked about the performance stuff, it sounded— it seems like there's some overlap with what I've I guess, was it digital asset management? Is there a component of it that's like digital asset management, or is this something that works hand in glove with a digital asset management platform?

It's more the latter because the job of asset management is to be the system of record for all of your content. Once stuff has been approved and ready to use, it has to be there so that different people can take it. But how does it even get there? Somebody has to decide that, hey, here is my marketing brief. for this audience, this kind of campaign, this kind of objective. These are my assets of photo shoots and models and products and all of that. This is the copy, these are the translations. I need to put all of that together into these 5 or 10 or 20 different base designs and multiply and get hundreds of pieces of content.

Okay.

Once I make those, they are sitting in an asset management software, but most times it is directly deployed onto paid ad channels or your CRM and things like that. But asset management is probably best suited for your core assets of your products and models and fonts and those kinds of things. And there is a component of that within Rocketium, but we integrate with the top DAMs as well.

Okay. So, it's more like the creative ops is more about the production of the assets and then some— and then it's also with the deployment of them as well.

Not really the core management. Yeah.

And the review of them. Got it. Okay. Yeah, that makes sense. Okay. Thanks for that clarity. So, you know, having been in marketing ops roles and things related to that for a while, I, you know, I know I've worked with like creative directors and creative teams, but I always struggle with like, we actually, I can think of a place where I worked where creative, there was a creative team and there was a content team. Right. And I always struggled with like, why didn't they, what, you know, why were they not more integrated? And it's not like they didn't work together. It's just. 2 separate teams, different objectives, right? Maybe not aligned. But so I hear creative ops, so I think creative type people, but I have a hard time separating that from content. So who would be the primary users of, you know, or the beneficiaries of creative ops? Is it, you know, anyway, go from there. I'm just curious, like what, what parts of a marketing organization or an organization in general would be beneficiaries?

Yeah, so when we look at the broader ops space, there are many ways to slice and dice it. Today we have limited ourselves to focusing on 2 teams. One is the creative team and one is the performance team, and so much so that our products are named Rocketium Creative and Rocketium Performance. So when it comes to creative teams, these are the ones who are predominantly spending their time on design tools like Photoshop, Illustrator, Figma. So what we are trying to do for them is to say, Here is a familiar experience like those design software which Rocketium has built from scratch. You don't have to use that if you don't want to. You can import your Photoshop, Figma, Illustrator files and templatize it. The goal of that being when you take that design template and the brief that comes from your marketing team, you can combine those 2 and multiply the content across different sizes, across the different content formats so that you don't have to do that copy-paste work in your design software. Now people have the capability to do complex animations and clip masking and whatever other crazy design things that they would like to do in our software. But the value of that comes in the multiplication of the design. And it's not only multiplying it blindly. There are certain things that you would like to check. For example, certain words are not allowed. Alignments have to be checked. Contrast between the background and foreground has to be looked at. Is it readable? There are guidelines such as Americans with Disabilities Act about how accessible this content is on the internet. So the creative team has to do a lot of these things manually, which is why they would like to do more creative and content coming together. But there's a lot of such operational work that they have to do on a daily basis. So what Rocketeam does is it checks for many of these things on its own, flags some, autocorrects for others. It also has a review process right here so that you don't have to then download these files, prepare it for somebody else to review, and then get the feedback. So the feedback happens right here. Everything that Rocketium Creative does is looking at the different steps of repetitive, manual, automatable things that designers do so that they can focus on the purely creative things and then come in, in bits and pieces and just check whether everything is okay. When it comes to Rocketium Performance, it is doing something similar for performance marketers, specifically solving the problems around insights and analytics of the creative choices they've made. So this part of the product connects to the paid ad platforms, or you could upload your own custom data if it's not a standard platform. We extract the platform metrics. We also extract what's in the creative so we can figure out what sort of models and products and layouts, colors, copy. We can figure out more detailed information about it, saying what sort of interaction is between the model and the product. For example, is the model next next to the product, holding it, using it, and that sort of information we extract so that when a performance marketer is planning for their next campaign or they want to figure out where am I really leaking my budget, what do I want to improve, they can look at that and say that, okay, great, I never had this insight before that every time a product is outside the packaging versus inside, one works a lot better than the other. So that's what we do for So, that's really in the 2 ways we solve this problem.

So, it's creative teams and performance marketing, or I mean, the performance marketing I think is a relatively new term. Call it demand gen or demand creation type teams. Is that the primary focus?

Yeah, it's the demand gen is the more B2B world, which is what you and I live in. Within the more retail brands or agencies or people who are living and breathing Facebook and Meta and Google ad platforms, platforms, it's performance marketing, paid marketing, but yeah, broadly it is demand gen.

Okay, got it. Okay, some more, okay. Oh, that's interesting 'cause I hadn't tied performance marketing as a thing that's with B2C type organization. Okay, so I have a really tactical question about that. So one of the struggles I think, I know I've seen creative teams run into is how to handle, how to produce creative that renders well in both dark mode and light mode. So are you able to help with that as well?

So marketers have even more challenges because if it goes onto Facebook or Google somewhere, you have a little more controlled walled garden of what the content appears next to. But as soon as it goes onto, let's say, a Google Display Network or Facebook Audience Network, I mean, these are technical words not everybody needs to know, but the idea is that when it goes on to a third-party website, you have no idea what it's next to. It could be a jarring, you know, bright-colored website, and yours is a more subtle design. So some of those challenges exist. You cannot always solve for it, but really what you want to stay true to is A, your brand, but B, more importantly, who your audience is. If you know what your audience cares about, sometimes they can be forgiving. Sometimes they're on light mode, sometimes on dark mode. You cannot really control for that, but Knowing what your audience cares about, what sort of message resonates with them, that is probably what's more important because you cannot always control the thousands of websites on which your content goes, what it is in line with, stuff around it. Maybe there's an incendiary story, you can't always control it. But really, you want the message to connect with people. That's really what you want to focus on.

Yeah, I get all that. I guess I'm thinking more like, are they really like I worked with marketing. I had worked with one marketing team in particular that was very, like, the quality and the match to brand standards of colors and things like that was really, really a priority for them. And one of the things we'd run into is, you know, when you say send an email to somebody and they look at on their, on their phone, if their phone— if a person is using dark mode, right, it— like, the colors actually get out of whack and it's strange like that. So that's really what I'm talking about specifically. I get, I get, like, you can't control your, like, what's around your content when you use an ad network or something like that. But in terms of, like, how images— can you help with making images that will work well both in light mode and dark mode? If someone— if you can't get— because you can't really control for that.

Yeah, it's, it's not very easy to do, but the, the way we try to solve for some of these things is to say, what are your brand's guidelines? What are the platform's best practices? Can we automatically check for that? Can we suggest changes that you should be making? We don't have anything specifically for light mode versus dark mode, but different brands have their quirks and their needs. So somebody will say certain words are just not allowed, or our logo has to be larger than every other logo. Sometimes people are manually looking for that and eyeballing, you could make a mistake. So having software do those checks and flag those for you is invaluable for companies that operate at a meaningful scale.

Okay. Yeah. And that makes sense, right? And I, like, I remember, 'cause I was actually in that particular one, there were 5 brands I was supporting with one business and they all had their own you know, brand guidelines for color, you know, color palette and things like that. And some were fine when you go to dark mode and some weren't, right?

Yeah.

Depending on the mix. And that was, it was really hard to handle that. Okay. So you, I think we've touched on this a little bit, but maybe you go a little deeper. Is it, well, 2 parts to this, right? So first it sounds like, you know, when you talk about doing things at scale, it sounds like there's a lot of automation, and things like that where you can do some, we'll call them standard kinds of checks, even though it's probably oversimplifying it. But it feels like this is most, probably most useful for larger enterprise type brands. At the same time, like I can imagine growing smaller organizations might be interested in something like this. Like what's, like what are the signals that in terms of their business and maybe have benchmarks, I don't know, that would say, hey, like, this is something we need to be thinking about, you know, because we're spending a certain amount of time manually reviewing stuff and all that. What would that be like? So first off, am I right in my assumption that the types of organizations that most benefit from it? And then second, like, what are— for those that are not, because they might be smaller, like, as they grow, what's the signal?

Yeah, so you're right, the company This product was built for large companies, people who are operating at a massive scale. In fact, it started with one of the largest companies in the world, but that is not the expectation that, hey, unless you're Fortune 500, you cannot use this software. It definitely works for companies that are a little bit smaller in scale as well. But really, if you have 2 or 3 people in your team, you are making maybe 50 to 100 pieces of content a month, you do not need something like this. Because you could spend half a day and do the copy-paste work in a design software, do the reviews manually. If your ad spends, for example, are less than maybe $3, $5, $10 million, then maybe the stakes are not— of course, for each company, the stakes are equally high, but the amount of the dollars that you would lose versus bringing in a software like this, we have a customer success team that we deploy with you, we do quarterly business reviews, there's a lot of investment from our side as well. This is probably overkill for you. So typically teams that have more than 10 team members involved in this process between designers, creative directors, project managers, performance marketers, 10 or more team members, anywhere from 2,000+ pieces of content a month and $5 to $10 million of paid ad spend at a minimum, that is where they really start seeing value in something like this.

I see. Okay, so that's a good thing. You know, I can imagine probably not like startup-type companies, but call them teenage companies, right? Yeah, mid-sized that are growing could easily get to that.

And the pitch, if you hear it, what you're trying to do, trying to bring in technology, trying to make versions of content for each audience, that all sounds like motherhood and apple pie. Who is going to dispute that? Of course I want it. Even if I'm a one-member team, I want it, right? But The outcomes are great. The ways to get to that may not be a software like this. You should anyway do it. You should try and automate as much as you can. There are great AI tools today, and you don't even need to look for specialized tools. The LLMs are doing a lot of this today. But as you get a lot more scale, there are more egos in the room, there are more voices that need to be heard. That's when you need more specialized software.

Yeah, yeah, that makes sense. Yeah, I think what you hint at is right. There's trade-offs along, all along the way, right? So, and it's a good point, right? This is something that probably everyone, if they're not thinking about it, who's in a creative world or, yeah, creative world is, is, should be thinking about how to, how to handle that. So that actually brings me to another question. So I mean, we talked about enterprise companies, large brands, high volume. Like, can you maybe take that and give us an example? Like, you know, if you can share a real example of a, Maybe one of your clients, like, what does this look like on a day-to-day basis? Yeah, what's a kind of like example project or problem they're doing that, and then what, how would it, how would that happen with the tool?

Yeah, so let me give you a couple of examples of 2 different worlds. One is the agency world and one is a brand world. In the brand world, what had happened was this company, which I'll not name, preparing to go IPO, trying to cut down costs. They had let go recently 40% of their team, which is unfortunate. And now that they had let go of that team, they said, we still want to do the same amount of work, and that's why we want to bring in automation. The tools that they were using were not enough, and that's how they discovered Rocketium and we came in. What they were able to do with that is the designs that they were making on the other tools that they had, including Photoshop and others, They were able to bring that in, create a repository of standardized designs. And then the performance marketing team, which was on a very tight treadmill of 2,000 pieces of content every 2 weeks, they used to give them a very structured Excel with all the copy versions for every country and all the different products for which it had to be done and so on.

Mm-hmm.

That spreadsheet could be fed into Rocketium, and then you start with a base design, it automatically makes all the different sizes for each platform. And then the spreadsheet comes into it almost like doing mail merge. You would get the 2,000 pieces of content for every 2 weeks that they would have. The first set of review happens right here, and before you can even assign it to the performance marketer to review, you have already discovered many things that have to be fixed. You go and fix those things, they review it, then one click, all of the files get renamed according to your taxonomy, because this is a very common thing, especially in the B2C world. Your files have to be named according to a very complex format because that is how you are naming your ads in the ad platform. So the automatic file naming happens. Every ad platform has strict guidelines for how large the creative file should be. So Rocketium automatically resizes all of these files. So each of these steps that would have taken them half a day here, quarter day there, 2 days somewhere else, is completely crashed. And so now a team that is 40% smaller is doing the same amount of work in a fraction of the time. And now they can do more value-adding work like coming up with new creative ideas, looking at analytics, which also Rocketium has for this team to say what sort of creative choices are working. For the agency that I talked about, there, there was a clear need which was to say that my clients are asking for more cost efficiency. I have a 40-member team that is working on this client and they are saying, how can you cut this down? If I have to bring in Rocketium, I would like to free up at least 10 people. And now they were able to figure out how to do that in a fraction of that. And the process, the rest of the process was the same. So whether you have an in-house team that is doing it or an outsourced team doing it, everywhere the problems are the same, that there's a lot of copy-paste work, a lot of same thing that I keep doing, and again, that can be automated. So this was on the creative side. Similar thing, like I said, exists on the analytics side, which is these both of them don't use it. Another team does it where they go to every platform, download the data, clean it up, put it in a single Excel, and then they go and manually look at every creative, tag it, then do analysis. Rocketium automates that process—a single dashboard where you see cross-platform data. It extracts the creative information so you can see in charts saying. product inside the packaging, outside the packaging, 2 models, 0 models, all of those kinds of charts. And it gives you directionally what sort of creative choices you should be making, how should you be updating your campaigns. You get all of that information. So the before and after is very stark in all of these cases.

Yeah, I want to— so I want to come back to the metrics, but I have a— so we— what you described reminds me, we had, um, not too long ago a couple of guests on who are in video production. And we were talking about like how that has evolved quite a bit over the last several years. And they mentioned a particular example of a big change for them has been there are now tools that say, and I was thinking of like a retail type or consumer product good where I'll just, I think the example they used was a cooler, right? And the cooler may come in, 6 different colors, right? You do the shoot, you get the one color, and then they have tools that automatically generate the 5 other colors that can then be used as necessary through different channels. I mean, is that maybe a small example of what you're talking about, but is that the kind of thing you can do in an automated way to get things at scale?

That too. So you have starting assets, you want to make some changes to it using AI, Typically, people would go back and forth between Photoshop and a product like Rocketium or do the same thing in Photoshop. And now, thanks to AI and all the advanced tools that we have available, you can do the same within our product as well. Same thing when it comes to translations or rephrasing copy and those kinds of things. All of those AI tools are built into Rocketium.

Interesting. Okay. Yeah, I was playing— anyway, I don't need to go off on that tangent, so I'll hold on that. Okay, so you were hitting on analytics. So even though I'm not sure that every marketer knows what it means to be data-driven, I think everyone's claiming that they either are or want to be. So can you go a little further into how you go from— I think in particular, I get the idea of consolidating the results from the different platforms, ad platforms or ad networks, That sounds pretty straightforward. Again, oversimplifying. I think, how do you then— how do you get the attributes of the content that are important? I think in particular, I would expect video to be a challenge more so than images, because I've seen— I have seen digital asset management systems that do a really good job of auto-tagging content when it's loaded in, but How do you do— how does that work?

Yeah, so when it comes to imagery, like you said, there are enough tools that exist. Video is a lot harder. We have, to begin with, we have focused on static imagery, and we have capabilities for video, but a lot more can and should be done in video. But the way our value is coming in 2 ways. One is that at scale, being able to pull all of this information from different platforms and running this data ingestion pipeline, which is to say for every piece of creative that comes, go and extract the standard 80 attributes, which we have worked with customers. Because we work with so many people, we have figured out the standard attributes to extract. We call them lenses. So there is a product lens, a model lens, a messaging, branding, layout. All of these are different lenses. Within that, there are subcategories of different attributes that we would have, right? So like I was giving examples of What is the interaction between the product and the model? There are 5 standard ways that we have figured out that we would have within that, right? So being able to extract all of those, each one of those requires a bespoke way of figuring this out. AI today is good, it's getting better, so maybe in the future it's all done with one prompt saying that extract these 80 things for me. But today we have to figure out color in a single, in some way, extract the text in a certain way. Take that text and interpret that in another way. So there are multiple things that we do. So our system requires multiple things working in tandem. When it comes to video, it would work in a similar way. What we do is we extract the keyframes out of your video because if you had— if it's 24 or 30 frames per second and it's a 15-second video, you are looking at 3,000 frames and doing this level of attribute extraction for each one of those is crazy. You won't really be able to make head or tail out of that. So instead what we say is, can we extract the 5, 10, 20 keyframes within the video and then tag those instead so that you can analyze your video in a much better way? So that way it could be what is the starting frame, what is the first highlight which is in your video, how do you end? There is a peak-end rule that you know about, so how is it ending? So there are those kinds of things that we would do. But yeah, video is a lot harder. Working with that is a lot more complex. We are doing a lot more work in that. But the idea is that our value comes in standardizing these attributes because like you said, a lot of people are data-driven, they have strong views of how data has to be used, but almost no one has done creative-level analytics before because there weren't tools for this. So our value is in standardizing some of these things and then working with customers to discover more because let's say somebody is in the media and entertainment space, they want to know if the creative is talking about a venue or about a player or about statistics of a sport and so on. So that cannot be extracted automatically with the current system. So you would add additional AI tools to extract that. That is what we do.

Yeah, that makes sense. And now I'm thinking we had, oh gosh, what is his name? John, it's a guy who works for one of the major, studio studios, and he described that they're using AI for doing some of that kind of stuff too. So, and getting data, which is really fascinating because I think that's, that's probably a huge area. So how, so, okay, so now you got, I'll call them standard dimensions of attributes for the con— the creative and maybe a little bit of content too, it sounds like. especially if copy, if you keep copy with it. So, and then you've got the performance. How is that used? How do you then think about linking the performance to that? I mean, is there any like A/B testing that's going on? How's that handled? So you can see if you have slight variations in creative, you can kind of evaluate that, or are you looking across more of a broader portfolio? It's like, use the term model, I assume that's the people in the creative, you know, the attributes of the, you know, is it better with, you know, a woman or a man or dark hair or blonde hair? Like, I don't know what level you're getting to, but I can imagine like, are you finding, are you looking for things like that? Like, hey, when we target people for this product, you know, we should use, we find better performance With this kind of model?

That's exactly the sort of insight you're looking for, because typically what people would do is they would start with a hypothesis and say that because we are trying to target young moms or young dads, we think that this sort of creative would work. Now, most marketers and creative people are not wrong in the choice that they've made, but a lot of times this comes from tribal knowledge or gut feel or Just kind of looking at the past couple of campaigns and doing this, but saying that every time in the past when we have done this X decision, whether it is showing a model which is just cut off till here or the whole thing, nobody thinks at that level. They would just kind of look at it and say, this looks a little off, it's kind of not balanced, can you make it bigger or smaller?

Right.

But it's not backed by data, and you don't have to go crazy all the time, and every single thing has to be defined by data saying how many pixels below the right edge is your logo. You don't want to go crazy, but somewhere if you had a little more data to tell you that, hey, in your awareness ads that you're running, your logo needs to be big because every time your logo is small, people don't know who you are. You're running an awareness campaign. And so you may not know the reasons. Data cannot tell you the reasons why somebody reacted a certain way to something, but you can see patterns like this that We noticed that in your awareness ads, your logos are small, and whenever your logos are small, outcomes are poor. Now you could make a case and kind of retroactively put a story behind it saying that because it's an awareness ad, they don't know about our product, and maybe that's why the logo needs to be big. In an awareness ad, if you put a button saying shop now, that doesn't make any sense. You are trying to make people aware about your brand.

Right.

Maybe telling them that visit stores to see the product in action or some such thing, right? So, these kinds of things you would discover. The goal is to surface information that you might not have planned a priori, but you've kind of subconsciously made design choices.

Right. That's interesting because I can imagine— actually, my thought when you described that is, if I were in their shoes, I really don't care. Like, why? the one performed better than the other. I just wanna know that it did. And if I'm smart, even if it means that, 'cause I've seen this, the best example I'll use is when I've worked with marketers on emails, right? There's sort of this school of thought, especially for brands that are really, really focused on their brand creative and visual aesthetic kind of stuff. when you send an email, they really wanna make sure that that's really incorporated into the email, which leads to all kinds of challenges with just building the email and everything else. And my experience is like virtually every time if I do an email that's, I'll call it, you know, high design versus one that's basically text, the text one almost always performs better.

Yeah.

And so I think if people are open to the idea of listening to the data, it doesn't mean you have to make it like do that. It's also an every time thing. It's, but I think it's important to go like, you know, if I really care about the results, I need to check my ego a little bit. Like if it's not the most thing, it's the same thing with, I do the same thing, by the way, with copywriters because, you know, they get caught up in, should I use an Oxford comma or not? Should I have a hanging, you know, an orphan word on a paragraph like that? I'm like, it doesn't matter, nobody else cares. You could have a misspelling, right? Um, and I've actually—

And that is the, that is the tension between creative and performance. That's the 2 products that you have. So the idea is you want to strike that balance because you don't want to go too data-driven, too much about performance, and let's only care about this, because then you don't really have a brand identity. You are no different from a lot of, let's say, AI slop is the winner. You just generate some crap and put it out and it performs the best. Is that what your brand is going to stand for? So at some point, maybe you are still okay to have subpar results because you want to have a certain impression in the market, you want to be known for certain things. So some of this, even though we are really pushing for performance, because at least on the creative side this is completely absent from the discourse, that is what we are trying to bring in. We, we absolutely see the value in making subpar decisions at times because in the long run, maybe that is for the best.

And, and I think there's something to be said for, I've got 2 like examples of scenarios that I know about where like the repetition of something, even if it's not performing well, can generate results as long as you continue to be persistent on it. The one is—

As a parent, I can tell you that is true.

So, so, so the one I have as a parent, like putting, putting foods in front of your kids, it's, I think I've heard research, something like to get 'em to try something to really be comfortable eating something new, you have to do it, like try it 10 times, which is hard, right? When you've got a hungry child, right? But I, I don't know that we committed to 10 times every time, but we have kids who have pretty broad tastes, right? And we try to always put new stuff in front of 'em and set up guidelines. Like we don't, you don't have to finish everything, but you have to at least try everything that's, you know, that's presented. But the other one goes back to a work one, which is I worked with an agency that was helping with some database marketing stuff, but they were, one of their clients I think was one like Visa. And one of the interesting things that they shared about that, that they learned, and this was back in direct marketing days, right? Sending mailers out, which still happens today. But they found that the best performing piece of direct mail was the third one. So what they did is that like what that changed for them is didn't really change what they did on the first and second one per se. It changed the timing. So they moved through the first and second ones as quickly as possible to get to the third one. Because it was this like, whether it was conscious or subconscious, right, that people were receiving it, it started to sink in like, oh, this is a brand. Maybe it's that awareness piece. Again, like why it was happening, the psychology behind it, I don't think it really mattered. It was that I like the idea of like you've got these sort of the tension between the creative and the performance marketers. My concern would be if I was in kind of a leadership role there, it was like, what if I have performance marketers that tend to lean more towards the way you typically think of creatives, right? And they are less likely to try new and different things because they only— or they're more likely to try new and interesting things and not focus on the results. Seems like they're in the wrong job. But yeah.

And sometimes that could happen. Sometimes the loudest voice in the room might not be the most rational. But yeah, well, that's where I was talking about our customer success team and how we work with customers, because it's not just building software and throwing it over the wall. It's about what's the goal that you have signed up for with this company? Because the goal of that company is not to use Rocketium software and be proficient at it. It's for whatever goal that they had. And that's what we, whether through reviews or constant engagement with our customers, that is what we focus on.

Yeah, that makes sense. Okay, one more question and we can wrap up here. But so a lot of talk about AI and tools. You've brought it up, even some of the more, I'll call them more generic tools, LLMs, things like that. What advice would you give to the marketing teams and agency leaders that are looking to try scaling their creative production with those tools or with Rocketium, right? What would you be like, what's your best piece of advice for them who are just starting on this or thinking about starting it?

Yeah, so what we typically tell our potential customers as we talk to them, introduce what Rocketium does is It doesn't matter what Rocketium is, how good it is, or the shiny new toy that is there. Let's start with your business objectives. What are you challenged from doing today? What, what is unlocked for you if you solve this? Let that be big enough to warrant a change in your business, whether it is paying money for software, changing your process, doing something different. But you need to start with the why. Again, not to quote any jargon, but really to say that you need to have your business goal in mind. Once you have that, you can't just say that let me bring in AI and solve all of it in one go. You need to break it down. You need to understand what your process is. You need to figure out which part of it is the bottleneck because bringing in software will not solve the problem. Let's say the analytics one that we spoke about. If a certain company's culture— this is something we had heard— was that one of their co-founders was on the creative side. So no matter how much data the team went to them with, they used to say that, hey, this is the creative choice, we are not changing this. So what's the point of bringing in software if the decision is going to be vetoed by one of the founders, right? So somewhere you need to know, yes, the end goal is something you have in mind. Yes, the software can solve it, but you also need to know what your process is. How do you work today? What part of it are you trying to solve? So a little bit of the, the why, a little bit of how you work today. and then trying to bring in AI to solve the specific problems that you have.

So, it sounds like there's a cultural component to this to be ready to do something like this because—

Absolutely.

Yeah, okay. And that makes a ton of sense, right? Culture.

And that's a very CEO answer to start with the culture, but I really believe that's the only way this thing can be solved.

No, I go back to this all the time is that, well, if I think about kinds of cultures that I like, it's ones where Yeah, there's a lot of vigorous debate about, say, an important decision with the right set of people. And once the decision's made, I mean, the decision has to be made, but there needs to be a culture that is open to that debate and discussion. Decision needs to be made, but there's also the recognition you have to trust people to do the right thing, right? And don't hold, like, don't point fingers when, like, you're, if you had, if the decision was counter to what your you thought was the best one and it doesn't go well, right? You don't get to say, I told you so, right? Your job is to make it better and try to help solve it. And I think if this is a similar thing, right? Typically the way this shows up, I think, is these controls over review and approval and all that kind of stuff. And if it's what you described, maybe not the CEO or founder, but like a senior person who is the one, like, you know, go all through this work, you think everything's approved, and then there's this one person who could veto every fucking thing you did. Right? To be blunt. And it's like nobody wants to, then nobody wants to get into the effort of doing it. And it's really, really disheartening when that happens. So if you have that and you don't rec— like if you recognize that that's the case, then bringing any of this other stuff is not really gonna solve that problem. So I wouldn't discount the culture answer is my point.

No, perfect.

So anyway, hey, this has been great. Any, any last bits of Wisdom you want to share, Satish, before we wrap up?

No, I would say, uh, the only thing I would tell people is that this is a glorious age that we are living in. We have magical technology from the future that has been somehow pulled back in time and handed over to us. So do not despair. Do not think that everything is going to change, you're going to lose jobs. Make the most of this. When computers came, people had similar fears. When the internet came, people had fears. AI is probably going to be much worse if you let it be, but I think it's magical technology. Let's use it to make our daily lives better, and hopefully we make it to the other side safe and sound.

Yeah, sounds good. Well, I appreciate it, Satya. If folks want to follow up with you and learn more about CreativeOps and scaling it and all that, what's the best way for them to do that?

So, I'm fairly active on LinkedIn, so feel free to reach out to me. Otherwise, our website does a great job of laying out what we believe in, what problems we solve for our customers. So please visit rocketium.com.

Sounds good. Perfect. Satish, thank you so much. It's been a fun conversation and I think we covered a lot of ground. I think it'll be helpful for our audience. So thank you. Appreciate it. Thanks for having me, Michael. As always, thank you to our audience. Thank you for supporting us and providing your feedback. If, as always, we always are open to ideas for topics and guests. If you want to be a guest or have a suggestion, please drop a message to Naomi Liu, Mike Rizzo, or me, and we'd be happy to engage you with that and figure out how we can make that happen. Until next time, bye everybody.