In this episode of GTM Innovators, host Kyle James sits down with Sara Rosso, Senior Director of Product Marketing at Smartcat, for a wide-ranging conversation on the evolving role of AI in global go-to-market strategies. From her roots in engineering to her marketing leadership across companies like HubSpot, Automattic, and now Smartcat, Sara shares how AI is transforming content localization, translation workflows, and the very nature of cross-border collaboration. Tune in as we explore why AI might just be the new Babelfish—and what that means for product marketers, enterprise teams, and the future of global communication.
Subscript to GMT Innovators Series on the following platforms:
- SoundCloud: https://soundcloud.com/research-859405782
- YouTube: https://www.youtube.com/playlist?list=PLsoV6fwX4cpGR2Hg98rc1e1k2-cOqCbhb
- Spotify: https://open.spotify.com/show/1gvDzcl0jxpPIfu9WYu6U4
- iHeartReadio: https://iheart.com/podcast/258127960/
- Pandora: https://www.pandora.com/podcast/gtm-innovators/PC:1001097038
- iTunes: https://podcasts.apple.com/us/podcast/gtm-innovators/id1790738579
- Amazon: https://music.amazon.com/podcasts/1b000615-31cc-49dd-a5d8-f80d5098bf2d/gtm-innovators
Transcript:
Kyle James 00:00
Kyle, hello, hello. Hello, everybody. Welcome to another episode of GTM Innovators by 3Sixty Insights. I’m your host, Kyle James, and today we’re diving into, I don’t know. I like to think about this as AI is the BabelFish. Joining me today is Sara Rosso. Sara, I never seem like anybody’s name right is Rosa. Sarah is the Senior Director of Product Marketing at smartcat. Welcome to the show, Sarah.
Sara Rosso 00:40
Thank you. Yeah, I’m glad to be here. So I always like to
Kyle James 00:44
start these things out, like learning a little bit more about your origin story and then kind of how it brought you to smart, smart cat. Then we’ll talk about all the really interesting stuff y’all are doing, kind of translating and solving this whole Babel Fish layer. We’ll just use that for the people that are fans of Hitchhiker’s Guide to the Galaxy, right? I know we’ve got some nerds out there, but you have such a fascinating background, you know, having spent time at HP and Ogilvy and, you know, automatic and HubSpot leading up to where you are now. So I would, you know, I was just like, this is much for me as anybody else, kind of leading into this. Like, talk to us a little bit about that journey. I’m sure there must be some interesting parts along the way that kind of led you this and kind of what you’re doing now, kind of in Product Marketing,
Sara Rosso 01:25
yeah, I think, and probably similar to other product marketers who very few of us have had a very linear journey, I started off pretty technical, actually. So that was a systems engineer doing network, network engineering and HP. And, you know, moved to Italy, lived there for 13 years, and through a technical position and an ad agency, which was Ogilvy, doing everything but coding, so managing databases and all their email infrastructure, you know, became more of a let’s like, kind of like a techno whisper, and trying to translating the technical pieces to non technical people. And so just started moving more into, like digital strategy and consulting. And so I think that the two halves of me, of the like, the want to know how something works, and getting deeper into something, and then, like, being able to tell somebody who maybe doesn’t want all those details is actually what a perfect match for product marketing. So, you know, I think
Kyle James 02:18
everybody my bad joke, it’s like, I’m the guy, if you’ve ever seen the movie Office Space, I’m that guy, like, I talked to the engineers, dang it. I’ve talked to the customers, dang it. Like that translation area, you just do it more, and probably writing, right? Yeah.
Sara Rosso 02:31
And I think, and also kind of thinking about, like, what, what the customer gets out of it, and how, how it fits into, you know, their bigger goals, right? So, whatever, whatever, most of the things we have are not complete solutions for everything anyone’s wearing, you know, working on. So I think that’s the part you had to kind of get in their heads. And I like that. So, so, yeah, so, you know, and I think I don’t know if everyone on here is a housebot alum, probably not. But like, you know, I had a Senate HubSpot as well, where I was speaking more to developers and the technical people who are customizing the platform. And, you know, spent 10 years at automatic at as a very early employee, and that’s wordpress.com, VIP, and did kind of both the scaling piece of it as well as, you know, putting in systems and, you know, growth, growth product marketing. So now I’m a smart cat, and I think you know what drew me to this. First of all, AI is going to change our jobs, whether or not we want it to. And, you know, and I keep saying someone’s like, AI is coming for our jobs, and I think in some cases, yeah, there’ll be some professions that maybe disappear, but I think it’s probably more likely that a peer is coming for your job, who knows how to use AI. And I think that’s something that, you know, like we need to be learning as much as we can. Back to
Kyle James 03:45
that, like, peer, like, there’s definitely a conversation in that little element. But keep going,
Sara Rosso 03:50
Yeah, I mean, so for me, it’s like, you know, I’ve always been kind of on the technical bend. I love software. I love kind of playing around, you know, I had a sabbatical and I started making gpts Because I was also like, on the fence, do I get closer to AI, what do I do? And started making gpts and just kind of like, and then just start getting that, like, thrill, that excitement that I got when I started blogging, when I started making my own be books, when I started, you know, hand coding my websites back in the day. So I think I wanted to follow that a little bit more and and Spark as a company that’s doing really cool things with AI, you know, really deep roots in language, and not just like translation, because I think you know beyond, like a simple translation, but getting smarter every time. You know it’s really built for enterprises. So anytime you want it to learn, and you want it to get better every time, or you’re doing it in in a workflow, we have AI and human oversight. So that part was really interesting, in terms of how the collaboration is, but we’re also getting into content generation and some other things. So that’s why I was excited to join and, you know, plus living 13 years in Europe, like, you realize that, like, English is not the center of the world, and it’s really important and sometimes really expensive for people to think or. Or roll out things in multiple languages. So I think that was kind of a really interesting intersection, and it did spark my, you know, now that I’m playing with the product and stuff, it’s exciting. That’s awesome.
Kyle James 05:10
So let’s kind of like, dive in a little bit more for people, just kind of, you know, so they could get a better understanding of, kind of what smart cat does. Like I joked at the beginning, right? It’s kind of the Babel Fish element. But like, what are some use cases and like, things that people are using it for? You know, what are pain points that it’s solving now for customers?
Sara Rosso 05:28
Well, I think marketing is a huge one, actually. So you know, when you think about doing global, or, let’s just call them even, like, multi multinational campaigns you may have. And actually, this is a perfect example Ogilvy, right? I used to work at a place that we would receive ads from brands that I won’t mention, that had, you know, and they needed them translated into Italian. So the old, the old way, was you had a partner that was local, and that you could, you send it to them, and you waited, and it came back, and you use it. And I think that’s not the way that you should be doing it anymore, right? I think there’s, there’s better ways to do it, and so that that’s one of the pain points, is like getting and getting all those folks in one place. So, you know, translating campaigns, people are doing marketing materials, but we, we have it all the way to legal and compliance. You know, that’s that’s really think about legal notices and things like that. We have teams that are using that to send it out in native languages, where people have to understand, like, really important information. If you give it to them only in English, you really risk so you know, learning development is also another place where we excel and do, like, course, translation and all. And then that gets really deep into, like, company strategy, upskilling, making sure that, you know, global companies can actually reach their own internal employees and keep them longer and train them up. So there’s that’s kind of a taste, but, yeah, it’s all, there’s a lot. So
Kyle James 06:54
what do you think? Like, I’m just curious, right? Because these things are good and they’re getting better all the time. Like, what is the hit rate that they’re correct or they miss things, or that they hallucinate? When you’re talking about, like, the translation content piece, is it? Is it 95% plus, now that it’s it even gets dialect and things correct all the time, or, or, How much do you have to kind of watch that and have a person over, you know, double checking everything?
Sara Rosso 07:18
Well, surprisingly, I would actually say a lot of our customers want a person in the loop. So human in the loop, it’s actually important to them, so that they can have it checked. And I think, I think what we’re thinking about now, and this is, this gets back to my own profession later, is like we think about this 8020 right? Sure, like before, you would have a human person do it 100% and then, you know, you go through and go, I think we’re looking for those 8020 solutions now where we can get to 80 or 90 or 95% and then you you save that skill, skilled human labor, for for looking at the nuances, for looking at those corrections. And then, you know, you have other tools where you can give it glossaries, and you can make edits that it learns from. But I think that’s the point that we’re, we’re we’re trying to tap into one is bringing everything in one place, so even if you have an agency, get them on the platform. But secondly, those shoulder taps, which, you know, I don’t know if you’ve ever been the recipient of a shoulder tap, where someone’s like, Hey, can you look at this for me? That still happens in pretty much every company. So it’s also being more respectful of those folks time, so that, you know, you have AI, we have, we have a marketplace. So a lot of people use AI translation, then a freelancer and save that final slice for someone who really only has to look at a little bit of it. So I think that’s, that’s kind of when I think about AI is like getting you to a really good place, yeah, and then as you use it more, obviously, it gets to know you, and it could get you to a great place. But I think that’s, that’s kind of the time saver, right? It’s and it’s a speed because another thing too is, like, you may think of it as, like, linear, like, I do this and but like we do, you could do, you know, 12 languages in a minute. Like, I mean, you could do it all in a in a group. So I think that bulk processing piece of it, too is kind of hard to think about
Kyle James 08:58
well. And I’m curious too, because I’ll admit, I haven’t played a lot with the translation piece of AI. But one thing I really enjoy about using AI is like I write, like I talk, it’s run on sentences and all that. I imagine these things probably put out pretty grammatically correct content too, right? Oh
Sara Rosso 09:15
yes, yeah. And that’s what then we start getting into things like content generation in different languages. I mean, if you have source content, obviously it should be picking up on how you’ve already decided to write it formally and formally. It understands that context. It understands tone. You can give it additional instructions, but you can also do, you know, like, if you have brand names or if you have specific phrases, colloquialisms don’t translate. I mean, the one example is like, we use a lot of baseball in the United States, baseball, you know, that’s a home run, that kind of stuff, you know, the AI can know to translate that when it goes into another country to something that’s more relevant, like maybe, you know, football, yeah, football slash soccer, right? As they would call it in Europe, which would be football, right? So, I think those are things that you can. Can Teach AI as well, but it can get you there pretty close.
Kyle James 10:05
Well, that’s, that’s cool, too. Like, and going back to your, I didn’t know we get here this fast, but like, going back to your comment at the beginning of like, it’s not taking your jobs, but thinking about the people that could leverage that. Like, I imagine Traditionally, these translators also had to be, like, what I would just say earlier, like, great. Grammatically clean up, right? Clean. And they don’t need to do that anymore. They need to know the local dialects more. So the people that have those skills kind of that’s where it’s probably changing a little bit, right, right?
Sara Rosso 10:30
And you can give edits, and it can learn and like, in our specific case, but I go and I think that’s where you start to get into like, these enterprise solutions versus like something you and I can just pop into a web browser and use like, you’re not going to get the going to get the continuous learning piece. You’re not going to get the country specific. And so for some people who you know, they they operate in seven official languages of the company, they’re going to need something a little more sophisticated than that. So, yeah, that’s
Kyle James 10:54
cool. Like, and I’m thinking too, like, I recently had a conversation with a company beekeeper who who has a frontline mobile app for frontline employees, right? And a lot of frontline employees, English isn’t their native language. Whether they’re in the warehouse or, you know, a job site or, you know, restaurant or whatnot, they’re, you know, so having something that you could translate and keep it in in in their native language, like it’s super, super useful to them. But it’s got me thinking too, like, there’s probably a lot of these cases where that 80% is good enough just for, like, trying to communicate things to them, where I don’t necessarily need it cleaned up, perfect for, you know, marketing purposes, but if they understand what I’m doing, and I’m just able to, you know, do it very quickly, like you said, in in minutes, and it, you know, but on A website or in a mobile app that translated, that’s probably incredibly powerful, I’d imagine,
Sara Rosso 11:45
yeah, and I think, I think the difference is the 80% now can happen in maybe a minute or two minutes, whereas before to get there again, you’re sending it to somebody. They’re elaborating it, they’re sending it back. You’re not so I think, like you still have the opportunity, if you’re let’s just say it’s between 10 to 20% that’s where you’re spending your time, yeah, and you’re not spending on it, waiting and that kind of thing. So I think, I think that’s, I mean, like, in getting back to kind of, like, the skilled person, you know, well, the example you were giving, that’s another example of how folks are using it for, you know, like, safety, safety and compliance, like, so like, think about how important is like on on job compliance, or on job safety, for something, if you don’t understand something in your native language, you might injure yourself or others, or customers or, you know, so that’s like, something that’s we’ve also seen quite a few people that just love being able to make sure it lands, yeah. So like getting into legal, getting the compliance, also getting into safety for sure. So,
Kyle James 12:42
so let’s, let’s go there. You said something really interesting that’s got me thinking, like, what industries are you seeing get the most success out of this? Because we’re talking about stuff all over the board, but I imagine there’s certain ones that this is exponentially more valuable and useful for than others, right?
Sara Rosso 12:56
Well, I think it’s that’s a really good there’s so many layers here, right? One is like adoption of AI and what industries? And I would think that’s something that in some industries, they are definitely embracing it faster, sure, right? And so when we can come in with a use case like I just told you about, in an industry that maybe isn’t adopting in other places, like manufacturing, that that proves some business value to them, right? Pretty immediately. So that’s, that’s one example, but I would say, I’d say we’re seeing AI adoption, like, pretty strong across a lot of places. Maybe some of the like, you know, we’re helping them do, like, manuals and medical things, you know. So I think, like, it’s, it’s pretty widespread. It’s more about who, who’s putting the effort into either communicating with customers in different languages, communicating internally different languages, or they’re trying to, like, generate revenue and and that gets back into the marketing. So I don’t know if I would like again, I think this is, this is like a winners losers kind of conversation on AI, right? So I think it’s just like, there’s going to be some people in every industry that are going to be acting faster, and they’re probably the ones they’re going to be winning compared to their competitors, because they’re going to start to see that exponential speed, the efficiency, that kind of stuff. Well,
Kyle James 14:11
it’s interesting, because, like, the more I keep thinking about, like, the translation layer falls into these kind of, like, no brainer things for AI, right? Like, you’ve got things like, we hear the number one case I hear a lot of business talk about is, like the call transcripts, right? Give me summaries, be able to recap calls, and then all the intelligence other people in the company can pull from that. That’s one, you know. Another one you hear this is, this is clearly one of those, right? Like, it’s a no brainer why you shouldn’t be embracing this to do this kind of stuff and translate faster, but, even for a go to market motion, right? If you’re trying to go to New, you know,
Sara Rosso 14:46
I think people are skipping it, though, that’s the thing, is that. So when you I think, here’s how I think about I think that people see the value, right? And I think that people see the value of communicating. I think that before, though, they think it as a delay, right? Or something that happens at the end of a motion where the groups we’re working with are bringing it, bring it up to the front, right? They’re putting it into their plans. They’re putting it into and so then they’re able to do and like we have examples where they’re like, they did one language. Now they’re doing eight languages every rollout is eight languages every so I think that’s the shift is, is that companies who aren’t getting it and figuring out how it can change engagement or adoption or acceleration, are tacking it on at the end, and I just get it out, and when we show them the speed, then they’re like, Okay, actually, this needs to be a part of, like, the core project and the strategy of us doing this out and like, and, you know, I think we’re starting to get into this, like, do more with less, kind of era, and I think that’s going to be around for a while. So it’s also like, how do we find new sources of growth? How do we find and that’s when you can start to think of doing things like experimenting in new markets and trying things where, you know, we just came out with a this is, I’m actually really excited about this. We just came out of the image translator. So it’s image to image translation. So preserve your background, yeah. So you can give it a finished image, and it’ll give you back a finished image with the translation. But you can do some tweaking and move you can change fonts. You can move things around. You can you can hide
Kyle James 16:13
it’s like text embedded in the image that you can then swap out like so think
Sara Rosso 16:17
of an ad. Think of an ad with text in it. We can, we can do an ad like you. So here’s a great example. I’m a marketer. I got ads for my designer, who is very talented, but also very busy, and maybe, you know, like, you know, he gave me this finished ad, and, you know, I ran it, and it’s, you know, out of the six, the six ones that we ran, this one’s actually performing. Well, why don’t we try it in other markets, instead of having to go back to him and have him re elaborate everything, which he’s maybe moved on to another project or something, I can pop it in the image translator and get a finished image and do some quick little like, if I need to reposition things, and then I have a finished JPEG or PNG or whatever. So it’s like, it’s image to image, and then you also get the benefit of all the learning and the glossaries and things like that. And I think this is again, back to the whole thing where I think about his 8020 time, I might just be like, Hey, I did this in 12 languages. Can you just go in and do your polishing on it where he’s not getting a text document and then placing things, I mean, like, think of even about that, like, how you might do that in the past. Here’s the translation file of all the translations. Can you go in and swap those, those kinds of things. So I think also respecting his, yeah, respecting his time. If you wanted to have him in that flow, you could just pull him into the platform and do that. So I think that’s a it’s about like what these new workflows look like, what the AI and human oversight and human in the loop workflows look like. And I think that’s going to be really interesting to see how we change. That’s a use
Kyle James 17:45
case. I didn’t even think it was a now, you’ve got my mind spinning all these other use cases. Like, do you? Do you have companies that use this to, like, do the rewrite translation inside their product, right? Like they have language Absolutely.
Sara Rosso 17:57
Yes, yeah. I mean, I didn’t, I didn’t get the product engineering when we were listing off of this. We have companies that are using it for rapid prototyping. So if they are doing interface changes, they can do it so you can do it directly with like our integrations. We have companies who are translating videos on the fly so that they’re primarily English speaking legal teams
Kyle James 18:19
instead of just the text based. That’s crazy. Oh yeah, we do
Sara Rosso 18:23
audio, video, image is and then doc, like we do, like about 80 content types, so you can do content to content translation, so PowerPoint to PowerPoint, or, you know, you know, that kind of thing. That’s not like primary, that’s the thing. It’s way beyond text, though. That’s cool, but I think, yeah, the the product and engineering teams are using it more for software localization. We’re seeing people translate, you know, product screens, diagrams, those kinds of things.
Kyle James 18:54
And I’m wondering too, like I wrote a question I want to come back to in a minute, because I think it’s kind of the philosophical where this goes. But I imagine too, like, if I’m sending somebody an email, I can have it, then translate it, then they can open it up and instantly read it in their native language. I mean, there’s some stuff that does that now, but if it just it’s a plugin that does it natively. It enables teams to communicate with each other faster. Guys this. There’s so many use cases that just seem like a little hanging fruit, just to run with.
Sara Rosso 19:20
Yeah, support teams are also another great, you know, like, maybe either understanding something or a screenshot. They can do a quick translation. I mean, you know, that can just help them get, get to again, the 8020 right? Instead of, like, sending or shoulder tapping somebody to, and I’m thinking, that’s thinking, you know, we’ve been talking, I guess, primarily like English into but I’m thinking more about any language back into English. So if you have a small team, then also you’re able to help them get that much closer to being effective, you know. So there’s kind of, it’s kind of like outward and also inward, in terms of global,
Kyle James 19:58
well, and you. Talked about, kind of, like, some of the nuances of, like, different dialects and like, like, what sort of safeguards are in place? Is that where the kind of the human person kind of comes in just to double check that at the end. Or how do you kind of ensure some of that, and some of the cultural stuff and things that happened in kind of an a that’s a really
Sara Rosso 20:18
good point. Yeah, I mean, I think depending on the if it’s a regional dialect, I think we have something like 140 languages that we support, or, like, up to 280 language pairs, or something like that. But I think, yeah, I mean, it depends, it depends on the content, you know, and I think you can give it instructions as well. If you wanted to do, like, chain of thought prompting, or, you know, like, customize the actual engine, kind of orchestration. You can do all of that. That’s that’s getting way, like, for me, I start to get interested in that, right, like to go, like, way under the hood. So the thing is, if you have those kinds of presets or instructions, immediately, you may know, like, let’s just give an example that you go, okay, in in Italian. And it should know this, I think Ann Isabel, it’s like, we always use the formal, like, you need to use the formal all the time and this, and like you may even if this is written in a like, like a informal English, it needs to be formal in this market. That would be something that you could tell it ahead of time and give it that kind of you shouldn’t have to, or you could probably correct it. But I go, that may be something that companies spend a little time setting up, and then, you know, they’re set so
Kyle James 21:26
it’s definitely not something that people just run. It needs some sort some people do, though,
Sara Rosso 21:31
again, it depends on, it depends on the visibility of the impact and where they’re going and and I will say, some people say, we we do this, and then we wait to hear back. And I just spoke to a customer a couple days ago, and I won’t name who they are, but on the legal side of things, they were doing only in English. Now they’re doing 15 languages, and they previously, they were sending it to an agency, and it was like 3x the cost, and they were still getting complaints about the translation. And she said that now we’re sending it out, and we’ve gotten no complaints, and people are responding to us in their own language, which means they’re, like, connecting with the content in their own language and then replying, you know, natively. So she said that that’s gone down. So I think in her they’re also kind of monitoring to see, like, what is feedback. So you know, everyone’s going to kind of roll it out differently and look for those signals. But, yeah, it’s different.
Kyle James 22:20
That’s cool. So, so how does like, smart bug support, like the continuous learning feedback, right? Because that sounds like an example of when you’re getting that back from customers today, it’s not doing how do you support that? Like, is there prompts inside the app or, or, you know? Yeah, so, so,
Sara Rosso 22:36
so in smart cat, we have an editor, and that’s where you can review segments and you can give it, you know, feedback, and that’s something that could like, then becomes part of like your language graph. So as you make those edits that you shouldn’t have to do, like, if you’re correcting something it should. But then some folks actually seed it with with things like glossaries or terminology pairs and things like that. But, but, yeah, there’s a mechanism for you that’s a layer that becomes a part of your like, I said, the language graph that you can be correcting it, and over time, you shouldn’t have to do as many of those corrections, or very few of them,
Kyle James 23:11
right? If it gets familiar with kind of, like, what you’re trying to
Sara Rosso 23:16
accomplish there, yeah. And by project, I mean, we’re kind of talking generality, but I go, like, if you had a project that had specific instructions, you could, you could do some prompting to, you know, I mean, let’s just think about it. You’re completely different audience. It’s completely different, like, experiences. Or you could do it if you needed to, and just kind of like, prep it for that.
Kyle James 23:38
So now I’m curious, like, what have you seen, like, what are some really innovative, like, productivity gains that, like companies have seen, like, in kind of this space or, or have you seen anything that’s like, you’d almost even call maybe a breakthrough that’s kind of happened from, from what what companies are, how they’re thinking about using AI translations.
Sara Rosso 23:58
I mean, I think, like I said, one of the biggest pieces is, like, bringing it closer to their outcomes and more less of an afterthought, doing their strategic planning when they start to do this, I think, you know, the image translation is huge, because there’s very few, or definitely no enterprise that I can think of that are doing this right now. So that’s going to change. Like, probably you haven’t even been thinking about localizing those, and they’re like, that’s just, that’s too much work. That’s too like, we can do text, we can do all these things, but like, like, this finished images, it’s not going to work, either if it’s in learning content or if it’s in, I don’t know, a product diagram or something you you kind of have been not considering it. So I do think that is, we do consider that a game changer. And I do think that’s going to open up. I mean, it’s in beta right now, so I go, that’s going to probably open up some really awesome things. We’ve already seen e commerce, that people are using it to, you know, show products which, you know, there are some stats about people are more willing to buy from, you know, sellers who are showing products in their native languages. So we know that, we know that they all, they’ll connect to that. So I think that’s probably the, one of the first. Places that it’s going to take off, as in E commerce and in marketing, but we’ll probably see the effects of like, using that much later.
Kyle James 25:08
What do you see this? What do you see this going like, what sort of innovations do you see? Or even, is Marquette kind of cooking up because, like, it seems like all the base stuff is handled now, or, or it’s good enough, but is it just continuing to get mentally better and better to, like, 99.9%
Sara Rosso 25:24
right? Or, yeah, I think, I think there’s levers of quality that I always think that you could be getting better and there. So I think that’s going to be, I would also say that expectations are getting higher, right? So, I mean, people are kind of expecting AI to work, right? And sometimes it’s not, I mean, like in general, like, we know it’s, it’s a machine, and there’s and there’s different layers on top of that as well that gets it to great output. But I think now, the way we talk about AI, like, we are setting bars high, and we’re going to have to, everyone’s going to have to have product experiences that match at that. So I think we will continue to invest in quality and making sure that we can, you know, delight people with what, with what it can do and how quickly. So I think, you know, quality, speed, you know, like I mentioned, being able to process things in parallel. And I mean, like, like, the old way of doing it, like you couldn’t send and say, translate this in 17 languages for me next week or in 30 minutes. And, I mean, and that’s, that’s the kind of gains that some people are getting. So I think that that’s going to be interesting. But I think beyond that, I think getting into repurposing, remixing, reusing, you know, then we start to blend the content generation. But I go not just generation, but like from source materials that they’re using. We already do that with learning content, so you can upload courses and things that that you’ve used, maybe they need updating, maybe they need and, you know, you can actually do that with our AI, but I think we’ll probably be getting into other ways to do that. But I can see that being great, because you already have the things that you want to communicate there, and you can remix them and experiment and and then also reach those global audiences. Yeah,
Kyle James 27:04
and we’ve been a really good point that I don’t even think is fully set in with me yet. But it wasn’t what six nine months ago that we were all joking because AI was telling you to put glue on your pizza
Sara Rosso 27:17
because there are strawberries, and it was disagree, and how many R’s were in strawberry, I think
Kyle James 27:22
stupid, random stuff, because it dug it up on Reddit. And like, we’re not seeing any of that anymore.
Sara Rosso 27:26
I think we’ll still, I mean, so you mentioned hallucinations earlier, and I don’t know how much this audience is thinking about that. So you know, the big one, of the biggest bewares, right, is hallucinations, which is making stuff up, if we had to making stuff up. And you can definitely see that if I start to go into something marketing, and it’s making up data points, and I have to remind it on something, I actually think that we’re gonna, we’re gonna come to a really interesting crossroads of the people who built AI. And there’s a lot of ways we could dig into that deeper, but let’s just say you’re building a performance product. Let’s just start with that as like a base layer. And that’s the tech techie in me. You have a you want a product that performs well, so you put in something and you get something out in a reasonable amount of time. That’s one thing. I’m actually running more into omissions, omissions that I don’t get as full or as deep or as a expansive enough of an answer. And I think that is the intersection of building something performant and trying to get an answer back to, you know, the person who’s working with it, and actually going deep. And that’s why we have things like chat GPT deep research, and that’s why Gemini has deep research, because some of these queries are going to take more processing power. And so I think that’s going to be the interesting piece when you have a non expert, right? So in some of the cases, when I saw that there are omissions, I knew that there was supposed to be in there. But a non expert, someone who wasn’t familiar with the material, would have been like, Okay, this is enough. And so that’s where I think we’re going to have some interesting learnings as we’re marketing professionals, as we’re GDM professionals, like that’s still the great intersection of like expert plus AI is going to be better work and great work, not like anyone’s gonna be able to do this work. I think that’s the piece that we have to keep, kind of keep in mind, I haven’t seen as much about hallucinations. The omissions has bit me a few times. I fight with AI sometimes and argue with them, like, Hey, are you sure you didn’t leave that out? Oh, yes. I mean, so that’s the kind of thing that I go we’re going to have to train each other and, like, also keep the software accountable for that. But I think that’s going to be some of the some of the risk, right? Because the software engineers who built the back end don’t want a query to run for 17 minutes, right? Right? To give you an answer. So that’s some of the stuff that I think is really interesting to see. Yeah,
Kyle James 29:49
it’s all good point. And I’m just trying to think, like, I don’t see it arguing with me and coming up with as much crazy answers as it used to. But to your point, like. Maybe that’s because I’ve trained it, or it’s trained me, or we’ve worked enough, but, but also, I think the other thing that I’ve worried about the whole time is like people like us that have been in the workplace for decades, like we’ve built up this level of like competency, right? Like we’ve got the scars on our back of having to learn what is right and wrong and and now these things are good enough where junior people just won’t be able to recognize it. And that’s, that’s the biggest fear I have about all of this right now,
Sara Rosso 30:26
yeah, and I think that’s, that’s going to be something in a leadership role we’re going to have to figure out is like, and actually, we’re about to run an internal an internal workshop on, like, generating content with AI and teaching our sales and customer facing teams to do that because, you know, being a small company, we can’t actually ghost right for everybody right now, we want to teach them how to turn customer conversations and things into insights that they can then, you know, and then that’s the thing we need to remember. AI, can write a lot of stuff, but you have those customer insights, you have those conversations, those examples that are true and real, that you can bring and AI, that’s the 20% that you know, and AI can help you with the 80% that that’s what I kind of see, is that, like I could have sat down and written that blog post, and I use AI off and on for things, you know, but I use it more for editing, I would say, to critique my work, Probably personally, but I go like, but, but maybe two hours later getting to that 80% and I have them do a couple iterations, and, you know, maybe in 15 minutes I’ve gotten to an 80% that I can modify. You know, that 20% I’m still going to kind of, that’s my expertise, and that’s my viewpoint. So that’s, I think that’s going to be the piece we’re going to continue to kind of negotiate with AI is like, what do we bring? What kind of insight, what kind of, what kind of scars, what kind of, you know, mistakes that we’ve made that help us and then work with it to, like, get us to a better, like, faster, to things, I think is going to be really helpful.
Kyle James 31:55
Yeah, yeah. It’s wild. Let me, let me ask you this. I was holding up on this earlier, but now I think it’s probably the right time to ask. It’s like, with kind of AI able to do this translation layer, like, what does that do long term for language, right? Like, do we? Do we because, because we’ve seen over the last, and now I’m getting, I’m really going philosophical here, but like, this is where the fun part comes, right? We’ve seen over the last, let’s say, 100 years. How many languages have died off? Does this accelerate this, or does this preserve that? Because people can, Oh,
Sara Rosso 32:28
interesting. I think actually preserve it. I think it would preserve it
Kyle James 32:33
because there’s no pressure. Like, I think
Sara Rosso 32:35
it was a tick tock. I mean, I don’t know if you weren’t tick tock enough to add this little thing that goes in your ear and it’s doing instant translations for you. I think, like, the point is we want, we want to globally communicate better, yes, right? And so, so we want to make it easier for companies to do that, faster, for them to do that, and like, for it to make sense, right? To be a part of, like, their core of what they do. And I think that’s what all of this is doing is saying, hey, something that was an afterthought. Let’s, let’s make that part of our core, and then helps them connect with customers, helps us connect with each other. So I actually don’t think that it like I think it can be, I think it’s going to be easier. I mean, hopefully for these companies, maybe they have more official languages. Maybe they bring on a country that they couldn’t operate in previously because they didn’t have I mean, I think there’s it could open up a lot more, yeah, but it’s a good question well, because
Kyle James 33:27
me selfishly saying, like, you know I know one language. I know English, and I want everybody to talk English so I could communicate with them. But, and you see large parts of the rest of the world that you know, learn English as a secondary language so that they can get into the American market, or whatever it is. Or whatever it is, but that won’t necessarily become necessary anymore. So Well, I
Sara Rosso 33:47
think that is your question, though, are we going to be in a world where English is the business language of the world? I don’t know. And I think that’s, that’s probably the one that you know, and we shouldn’t be afraid of that. And we should, you know, we should be figuring I think that is probably the shift that I would see. Is if we can all communicate with each other with ease, more ease, then the actual language that I speak probably comes less important. So whether or not English loses its we’re getting very philosophical here, whether English loses its like foothold as the primary business language. And, I mean, they’re probably countries like China and, you know, India would would disagree with that already, right? I mean, I go, I think that’s, that’s probably something that we’ll probably see accelerate or, you know, I think maybe it just maybe we’ll become more multilingual as, you know, as a I mean, how can we become more multilingual? I think we’ll, you know, maybe be doing business in more and more languages. All
Kyle James 34:42
right. All right. One more follow up to that, because to the multilingual point, right? There’s a lot of data and search research out there that shows that people that are multilingual the way their brain works, because they’re able to cross things in more unique ways, which is super useful in creative aspects. And, like, it’s one of the big things I wish I was better at. You know, I was just enough Spanish to pass in high school and college. But like, do people not continue to flex that muscle and learn those additional languages to their detriment because they don’t need to? No,
Sara Rosso 35:17
well, again, I think, like, I don’t know, but I think that if you think about again, it’s like many people who don’t speak English learn English. I think that’s the piece that you’re I think that those people probably continue to do that potentially. You know, my kids are in immersion Spanish school learning because I want to, I speak Italian fluently. I speak a little bit of a Spanish as well. But I think, I mean, it is a good question. I go like, will they have as much incentive to do that? I do. I think, I think it can only grow until we get to the point where, where we all have something in our in our area, that’s doing instant translation. I don’t think we’re there yet. I think we’re gonna, we’re gonna reach it on some other, some other areas first, right? Whether or not you have a support team that only speaks one language or maybe speaks whatever language you hired them to do, but could handle any incoming language because you have the tools. I think that’s probably what we’ll see. We’ll see these multi multinational, multi global workforces probably not worrying as much about what language they’re working in. Yeah, that’s a little far out.
Kyle James 36:21
No, it’s like, Thanks for like, indulging me, that this is where my mind goes. Like, there’s these really interesting other things that are coming that you know, I hadn’t really thought about, because I haven’t thought about the translation piece, and definitely in a way that you have, Sarah, what kind of wrapping up time here? Because I want to be mindful of your time. What have I not asked you that I should have? What have we not talked about? Talked about any, anything else that you could think about. Well,
Sara Rosso 36:44
I mean, next time we talk, we can talk about product marketing specifically, and how the profession is changing a little bit. But I think, um, I think I just want to continue to talk about that with others. And like, how do we, how do we all, like, evolve in the age of AI? And like, how is that going to make us better? And what things do we start to say no to? So that’s the thing that I’m kind of interested in hearing from other people on the podcast. So maybe that’s like a push forward for you to ask other folks, even if they’re not getting deep into AI. I
Kyle James 37:14
could do that. Let me, let me ask you this, because one thing I’ve heard recently to get your perspective on it is because of the way AI is working and because some of the SEO stuff is changing, right? People don’t need to worry about long tail content anymore, right? It doesn’t, it doesn’t hurt. And so therefore, does Product Marketing become more the central aspect of marketing? Because you’re talking about the pain, the use cases, the way they use a product, and we don’t need to solve for like, Hey, here’s this article that answers 10 things about this is tangentially related to what we do, because AI is answering that you’re not going to Google and searching for it anymore.
Sara Rosso 37:52
Well, and I think I’m just talking to my my friend who runs SEO, and, yeah, definitely, there’s some consideration. I think, I think there’s a different question here, if I can turn it around, which is, where are people going to be putting their attention in the future? And that is the piece, and whether or not AI emerges as this, like, is that the first place I go? And for some things, it is right. And, you know, I do think the real time we’re still, we’re still not quite like, not all the models have real time, and that’s, that’s the that’s when we get it, nonserious hallucinations. But I go, if you have a person who’s not savvy and is using a model that only has information from November 24 you know, so I go, I go, versus, where do I go? Where’s my first software information? And that probably will hit the tech folks first, and, you know, we may lead the way. But I go, the average person, where are they looking for information where? Where’s their attention? So, like, as a marketer, I can reach them, I think that is going to go through some major shifts in the next couple years. And I think that’s the part where probably need to be well, B to B is going to be different than B to C. You know for sure, I think, I think that B to B buyers are going to continue to look in certain places for information. But yeah, like you said, our relationship is going to change with the with how we search,
Kyle James 39:05
yeah, yeah. And I think it’s teeing it up to where, you know, product marketers could be more of of the primary purpose of marketing, personally. And let me ask you this too, like there’s always a debate you see now, like, where do you put the product marketer Right? Like, do you have a strong take on that? Because, I mean, it just, kind of, my answer has always been, kind of, it depends as a living product or marketing, but what are you seeing? And kind of, yeah,
Sara Rosso 39:32
that’s, that’s, I mean, I think also, what do they own? And that starts to get into company maturity, that starts to get into, you know, by the like, you know, is it the first product marketing hire? I think, like for me, and that’s where I spend I am. I get excited by products. I’m always spending a lot of time with the product and and so whether or not I’m there with them or very closely working with them, and I think, doesn’t matter as much to me. I think. More, I become a translator for the rest of the marketing team too, right, in terms of understanding limitations, opportunities, that kind of things. But I think, I mean, the wonderful and sometimes terrible part of Product Marketing is that it can live anywhere, and it can almost do anything. So I think, you know, I guess for people listening, it’s more like being thoughtful about, you know, when you’re building out your organizations, what, what are they going to be responsible for? How you going to track it? Because that’s another thing, right, where, how do you have the person split their time between product adoption and, you know, like GTM, funnel generation is really interesting as well. So so much that’s feels like another hour we could fill with
Kyle James 40:39
that probably is what we need to come back and dive into that, because my background is a lot in products, so it’s one of those things that I run into a lot. And, you know, having spent a lot of time building marketing products, like, there’s some aspect of product marketer in me too, and then content producer all this, right? But it’s it. There’s very much that’s a chicken or an egg, and they’re very strong opinions about both. And the only way I end up it’s like, well, it depends. Like, and you have to think too, is product LED. Growth has become bigger and bigger like, well, that is all product marketing, right? Like, you’re not doing all of this top of the funnel, brand awareness, you know, marketing and sales driven effort. It is all product driven.
Sara Rosso 41:22
Yeah, yeah. I mean, there’s the it depends. Is the answer, I think, really so it’s the answer in the intersection. That’s, I keep saying, we’re at the intersection of so many things. It just depends where you are on the
Kyle James 41:34
map. Totally. Well, awesome, Sarah, this has been awesome. I really appreciate you taking the time to join me today, and incredibly insightful, and you’ve got my mind spinning in all the different ways that I’ve never even thought about how we can use AI as a translation area, but what, how could people help you? How could people find you, and what are ways that the audience can kind of support you and connect with you?
Sara Rosso 41:55
Yeah, definitely. I think LinkedIn is still the place I had a sub stack, but I haven’t really moved it over from my old newsletter so much. So LinkedIn is a great place to connect with me. And I think that’s there’s some interesting stuff happening there. I think, you know, like, in terms of what people are sharing, I have, like, open office hours. If people want to chat, I love I’m a networker. I love to meet people. So, you know, working remotely 15 years, always interested in having a chat with with people. So, yeah, would love it if people reached out. And then, you know, of course, we’ll learn more about smart cat. Happy to chat about that as well. Awesome,
Kyle James 42:30
awesome. Love it. Well. Thank you so much again. And for everybody out there, if you enjoyed this episode, five star review, nothing less. Please Subscribe, Like, Share review. And if you have an idea for a guest or something you want me to talk about, please feel free to reach out at research at 3Sixty Insight.com, and with that, everybody, keep growing and we’ll talk to you again next week for another GTM Innovators, thanks everybody. Thank you.