CXO Roundtable — Agentic Distribution for Growth

746
01:59:38.700 –> 01:59:55.499
Julia Nimchinski: Phenomenal session. Thank you so much, Maureen, Brandi. And now, we want to welcome back Mary Shea, one of the most forward-thinking GTM AI innovators. Mary is a co-founder and chief growth officer at Meerkat, a futurist and former forester and outreach executive.

747
01:59:55.840 –> 01:59:58.990
Julia Nimchinski: Mary, welcome, such a pleasure, how are you doing?

748
01:59:59.240 –> 02:00:15.369
maryshea: I’m doing fantastic. It’s great to be here, and I have to say, my friend Brandy really fired me up. I was able to listen to the last 10 minutes of that session, and, absolutely fantastic. So, I’m excited for our panel here today, and to dig in whenever we’re ready.

749
02:00:16.240 –> 02:00:20.610
Julia Nimchinski: We are! Let’s do a quick round of introductions, everyone. Josh, let’s start with you.

750
02:00:34.520 –> 02:00:36.640
Julia Nimchinski: Can you hear me?

751
02:00:39.410 –> 02:00:40.850
Ashley Stepien: Yeah, we can, you know…

752
02:00:41.420 –> 02:00:43.340
maryshea: I cannot hear Josh, so…

753
02:00:43.340 –> 02:00:44.730
Ashley Stepien: Not here, Josh, yeah.

754
02:00:48.520 –> 02:00:50.489
maryshea: And I’m just kind of glad it’s not me, the one.

755
02:00:50.490 –> 02:00:51.160
Ashley Stepien: Where’s the problem?

756
02:00:51.160 –> 02:00:53.109
maryshea: Because it’s always someone, isn’t it?

757
02:00:56.190 –> 02:00:57.139
maryshea: There we go.

758
02:00:57.720 –> 02:01:03.320
Julia Nimchinski: Live TV? Yeah, let’s do a quick round of introductions. Mary, take it away.

759
02:01:04.150 –> 02:01:24.070
maryshea: Yeah, so my name is Mary Shea, and as Julia mentioned, I’m co-founder and chief growth officer of Meerkat. Meerkat is an agentic AI company that really looks to blend the human connection, empathy, and creativity with the boldness of an AI teammate. And,

760
02:01:24.070 –> 02:01:47.700
maryshea: I’m really excited to, be here as part of this conversation and moderate everything. I’ll give you just a little bit more details on Meerkat. The big difference is that we are focusing on team situations, so it’s built not just for individuals, but it shows up in your team meetings, in Zoom, Slack, email. It answers questions, recalls contacts, conducts research on the fly.

761
02:01:47.700 –> 02:02:11.190
maryshea: captures action items, and we also have a messaging app so that the conversation never ends. You can continue the collaboration, coordination, and productivity after your meeting. So that’s a quick overview on Meerkat, and Julia summarized my previous activities pretty well. I’m a former Forester analyst. I do like to take a look at the future, and I’m really excited to moderate this panel today.

762
02:02:11.190 –> 02:02:19.140
maryshea: So, that’s me. Mario, do you want to, give a quick background? And we’ll go around the, around the horn here.

763
02:02:19.140 –> 02:02:39.759
Mario Moscatiello: Yeah, sure. Thank you so much for having me. Mario, head of growth here at Airbyte. For those of you who don’t know AirByte, we are a data movement platform, so, you know, we help companies move data from all of their SaaS tools and databases and so on and so forth into data warehouse or data lakes, to do stuff like analytics and power AI models and so on and so forth. Yeah, very excited to be here.

764
02:02:40.320 –> 02:02:43.940
maryshea: Fantastic. Great to meet you, Mario. Ashley.

765
02:02:45.970 –> 02:03:07.679
Ashley Stepien: Yeah, happy to be here. Ashley Stepien. So, I’ve been in the marketing space for a long time now, with companies such as Webflow, RAMP, and now Hex, leading marketing. But yeah, I love, love these conversations, being on the cutting edge of AI, but also just, like, practical application of it. That’s what I’m really interested in talking about.

766
02:03:08.170 –> 02:03:12.660
maryshea: Totally. Thank you for that, Ashley, it’s great to meet you. And Sandy, you’re next.

767
02:03:13.230 –> 02:03:20.360
Sandy Diao: Hey everyone, I’m Sandy. Super fun to be part of this panel, a topic that’s top of mind for myself and a lot of the companies I work with.

768
02:03:20.360 –> 02:03:39.570
Sandy Diao: These days, I lead growth at Drift International, which is my own growth advisory, and I work with AI startups and corporates and enterprises who are looking to build AI products and use AI to accelerate their workflows, and come from a background as a former growth operator, where I led growth at companies like Meta, where I worked on Facebook, Instagram, and WhatsApp growth.

769
02:03:39.570 –> 02:03:52.900
Sandy Diao: companies like Pinterest, Indiegogo, Descript, and now get a chance to work with dozens of other companies as well. These days, I also teach the topic of growth, marketing, and distribution at UC Berkeley and Stanford. So, excited to meet you all today.

770
02:03:53.540 –> 02:04:02.409
maryshea: Fantastic. I’ve got to add you to my list. I’m sure we will have lots to talk about during and after this conversation. Love your background, Sandy, thank you.

771
02:04:02.620 –> 02:04:05.920
maryshea: Vanessa, give us a little background on yourself, please.

772
02:04:05.920 –> 02:04:24.649
Vanessa Schneider: Hi everyone, nice to be here with all of you. I’m Vanessa, I look after marketing here at Descript, which makes it incredibly easy to create, edit, and share video with a very user-friendly interface, but also a lot of AI and agentic editing features built right in. It’s a tool for anyone who wants to make video but doesn’t know how.

773
02:04:25.780 –> 02:04:35.160
maryshea: Fantastic, thank you for that. And it looks like we have Josh’s notetaker here, but Josh, if you are here, please reveal yourself. If not, we will, jump right into this.

774
02:04:36.590 –> 02:04:53.040
maryshea: All right, I hope that he will catch us at some point in the discussion, but, you know, let’s start with you, Ashley, if you don’t mind, and I always love to go from kind of, like, general to more specific, but, you know, from a marketing perspective.

775
02:04:53.300 –> 02:05:00.160
maryshea: What does distribution look like, generally, in the Gentic era? Tell us a little bit about how you’re thinking about it.

776
02:05:00.490 –> 02:05:15.049
Ashley Stepien: Yeah, I’m looking at it from two angles. So, the first is sort of, again, like, practical application. And that’s in three categories for me. It’s in listening, it’s in creating, and it’s in distribution, in the more literal sense.

777
02:05:15.060 –> 02:05:31.629
Ashley Stepien: When I look at listening, it’s really trying to gather as much information from either the market, from our sales experiences, from social, from customer interactions, from product interactions, etc, and trying to really kind of form opinions and make sure that anything that we’re distributing

778
02:05:31.630 –> 02:05:53.940
Ashley Stepien: echoes what we’re hearing in the market, and it’s incredibly relevant and personalized and right on. So a lot of those listening tools have been just, making huge leaps and strides in terms of, like, helping inform our strategy a little bit more. That next bucket of creating, so we use Descript, we’re big fans of that, but there are so many great tools out there for just creating content.

779
02:05:53.940 –> 02:06:03.569
Ashley Stepien: Creating workflows, creating more personalized engines, etc. So we’re doing a lot of that bucket as well. And then, the third bucket, which I think is

780
02:06:03.570 –> 02:06:22.500
Ashley Stepien: maybe the most interesting is around distribution. So, how can we automate across channels? Our growth channels, our paid channels, our website, our lifecycle experience? How are we using tools to make those decisions for us, and always kind of an always-on mindset?

781
02:06:22.500 –> 02:06:22.940
maryshea: Right.

782
02:06:23.170 –> 02:06:33.679
Ashley Stepien: the area of, like, tool application that I get my team to think about in those three buckets. I also think there’s another element here of, reshaping the marketing org.

783
02:06:33.970 –> 02:06:42.520
Ashley Stepien: And taking a good, hard look about how we’ve been built historically. And can we build the org to reflect listening

784
02:06:42.620 –> 02:06:55.670
Ashley Stepien: creation and distribution in a more interesting and effective way. So I’ve been looking a lot about flattening my organization, taking out those people who are maybe more dot connectors in favor of more builders.

785
02:06:55.670 –> 02:07:04.259
Ashley Stepien: And trying to make sure that AI is sort of prevalent throughout the build lifecycle of everything that’s leaving the building.

786
02:07:05.610 –> 02:07:26.459
maryshea: Yeah, I really love that, and thank you for framing it up in such an articulate, easy-to-understand way. I love talking to marketers and analysts, so you’ve made a very complicated topic pretty simple to understand. And Vanessa, I want to talk to you more about Descript offline, because I think there may be some synergies with Meerkat and what I’m looking to do.

787
02:07:26.460 –> 02:07:50.279
maryshea: But, before we leave, Ashley, I’d love to talk a little bit about ROI and metrics. And I was listening to the last piece of Brandy’s session before I joined, and they were talking about the types of metrics you want to bring to the board. I think we’re at a place now where no one really cares about vanity metrics. How do you sort of translate what you described into metrics that could really tell

788
02:07:50.300 –> 02:08:01.010
maryshea: not only the board, but the executive team, if you’re making positive progress in the right direction. And I’ve always found that to be a little elusive, particularly when you’re using edge technologies.

789
02:08:01.310 –> 02:08:13.200
Ashley Stepien: Yeah, yeah. Well, I think that the good news is the metrics haven’t changed significantly from what we’ve always been worried about, but the way I think about it is

790
02:08:13.260 –> 02:08:15.940
Ashley Stepien: Early signal and late signal.

791
02:08:15.940 –> 02:08:39.169
Ashley Stepien: So, if you’re a growth marketer or any sort of growth engineer, we’ve been trained in this for a long time now. We’ve been running experiments looking for those, early signals of success before we roll it out on a scalable level. So, more literally, like, what are the metrics I’m looking at to see if things are working and, like, are we being effective? At the top of funnel, it’s traffic.

792
02:08:39.520 –> 02:08:45.469
maryshea: And is it relevant traffic? I think we are going to be in the fight of our lives for attention right now.

793
02:08:45.470 –> 02:08:47.150
Ashley Stepien: and websites.

794
02:08:47.150 –> 02:09:10.969
Ashley Stepien: are completely changing in the way that people are interacting with them. We’ve always known they’ve been coming later, visitors have been coming later and later in their buying journeys. Now it’s later than ever with all of the tools available to them to do their research before they get there. So, I am really concerned with driving not just traffic, but relevant traffic, and that is one of my big metrics of success right now.

795
02:09:10.970 –> 02:09:21.329
Ashley Stepien: Is are we getting in front of the right audiences, and we’re pulling them in? The more middle of funnel metric that I’m looking at is, accelerating the sales cycle.

796
02:09:21.650 –> 02:09:38.599
Ashley Stepien: because we’re capturing them later in their buying processes, are we able to really calculate that, you know, we’ve taken a 90-day sales cycle and gotten it down to 67 days, because we’ve caught them later and we’re giving them more relevant content? And then the last, more bottom of the funnel, or latent.

797
02:09:38.600 –> 02:09:44.680
Ashley Stepien: metric that, you know, we all care about at the end of the day is efficiency. Are we getting better?

798
02:09:44.710 –> 02:09:59.919
Ashley Stepien: smarter, more scalable in the channels that we’re distributing across. Ultimately, that comes down to CAC and payback period. If my CAC and payback period is starting to go down, I know that I’m getting more scalable, and that’s ultimately the end goal.

799
02:10:00.490 –> 02:10:17.279
maryshea: Fantastic. Super, super helpful, and I see some head nodding here, so it looks like the rest of the group is aligned with you. Thank you for such thoughtful responses there. I’m sure we’ll come back to you. But I’d love to switch to Sandy for a moment, and I love your background, so thank you for sharing that.

1447
04:06:53.530 –> 04:07:15.369
Mary Shea:

800
02:10:17.280 –> 02:10:24.160
Mary Shea: You know, if you put on the lens of an investor, operator, and now a strategic consultant, You know.

801
02:10:24.160 –> 02:10:46.469
Mary Shea: I want to talk a little bit about distribution strategies, and the ones that you feel are most effective and defensible from an Agentic AI standpoint. I was reading the other day in an article, I think it was GoToMarket Fund, where they actually said that your brand and your distribution strategy is going to start to become your defensible moat in some ways, because so many different people are able to spin up

802
02:10:46.470 –> 02:10:58.919
maryshea: agentic AI companies with the possibilities with vibe coding and so on and so forth. But I’d love to… I’d love to hear you, sort of opine a little bit on, on, on this topic. It’s really important.

803
02:11:00.500 –> 02:11:25.490
Sandy Diao: Yeah, thanks for the question, Mary. I think that the first thing I want to say about what is defensible versus not, I think is unique to the circumstance of every company, right? Because when you look at the distribution, the go-to-market, the growth operations of any company, they have a unique setup that is tailored to their industry, to their market, to their customers, to their channels. And I don’t think there’s necessarily a one-size-fits-all. If you do this, then your strategy is considered defensible, it’s a

804
02:11:25.490 –> 02:11:50.460
Sandy Diao: distribution moat that you can preserve forever. And instead, I think more of the strategy is how do you move with moving waters and change with the changing tides, and is your company able to adapt to some of these changes that are happening? And so, with that said, maybe I think it would be interesting to kind of share with you all some of the most common, I think, distribution strategies and use cases that I’ve seen

805
02:11:50.460 –> 02:11:56.879
Sandy Diao: AI be able to accelerate, and that are already battleground tested and driving, you know, growth results today?

806
02:11:57.110 –> 02:12:13.930
Sandy Diao: And the first one that I want to start with is a bit of an obvious one that all of us have probably touched upon, which is that when we’re thinking about distribution, there’s always some component of interacting with a customer, or what is the messaging and what is the creative, if you will, that is being shown to the customer.

807
02:12:13.930 –> 02:12:38.860
Sandy Diao: And the game has really changed for creative development at scale. So, some of my favorite tools include the likes of, you know, Descript, of course, and then more specifically, Pinpointed to performance marketing and paid. Tools like Creatify AI, for example, actually allow brands to create these really cool and compelling UGC, or user-generated content style ads, talking head on a background with a product that’s being displayed without ever having to hire a single on-screen talent

808
02:12:38.860 –> 02:12:48.050
Sandy Diao: talent or hire a video editor. And the goal here is not to hire less people, but the goal is to figure out how do we test more nimbly, and how do we create these different variants, because

809
02:12:48.050 –> 02:13:13.050
Sandy Diao: audiences are so vast and so broad, and we need to be personalized to every corner of the universe in terms of the formats, in terms of the platforms that we’re targeting. And then there are yet other tools around creative development. Like, for example, you can use a general-purpose image generation tool like Gemini’s Nano Banana, that makes it much, much faster to iterate on static creative assets in addition to video, of course, and whether you’re developing

810
02:13:13.050 –> 02:13:23.570
Sandy Diao: a header for email, a banner for social media, assets for a landing page. You know, these tools basically allow you to have sort of this creative iteration at your fingertips that we didn’t have before.

811
02:13:23.570 –> 02:13:40.299
Sandy Diao: And is that a defensible mode, per se? Not necessarily, but if you’re able to use these tools to basically use the data that you have and make these changes, I think that’s pretty much as good as it’ll get in terms of you being able to inch forward and make kind of these steps towards

812
02:13:40.300 –> 02:13:43.310
Sandy Diao: Getting the outcome, the growth outcomes that you desire.

813
02:13:43.870 –> 02:14:07.000
Sandy Diao: The second use case and distribution mechanism that I’ve seen really kind of hit the next level is SEO and programmatic search broadly. We won’t even necessarily get to cover GEO or generative engine optimization before, you know, more of the traditional, how do I get my website and my web pages ranked in search results, whether they’re generative engine or traditional search engines?

814
02:14:07.000 –> 02:14:27.709
Sandy Diao: We now can use a combination of AI tools plus agents to do everything from researching personas, writing copy for positioning, and you can even use these agents to essentially autonomously execute landing page content. So you add to a list of topics in a spreadsheet, and then that spreadsheet will basically trigger off a number of research workflows, write copy and…

815
02:14:27.720 –> 02:14:52.690
Sandy Diao: This is really in contrast to the sort of pre-agentic world, where we would, you know, in my experience, leading SEO initiatives in companies, we would oftentimes spend, you know, weeks or months prioritizing lists of verticals or use case landing pages, and say, this month we can only take on 5, or, you know, next month we’ll scale up to 10, and then 20, and then 50. And now, you know, I’ve truly seen teams who start building SEO pages day zero for

816
02:14:52.690 –> 02:14:53.839
Sandy Diao: the first time.

817
02:14:53.840 –> 02:14:56.390
Sandy Diao: 0 pages to 50 pages within a week.

818
02:14:56.390 –> 02:15:04.389
Sandy Diao: Right? And these pages start getting impressions, they start getting ranked, and so distribution is truly accelerated on the programmatic SEO front.

819
02:15:04.700 –> 02:15:12.159
Sandy Diao: And maybe the last one that I wanted to touch upon is more of a broad use case, but in a lot of the go-to-market work that we do, there’s

820
02:15:12.160 –> 02:15:33.500
Sandy Diao: actually a ton of research work that happens in the background. This is, you know, my experience, but whenever we’re kind of in those go-to-market planning cycles, like, hey, let’s launch a feature, let’s launch a product, there’s just so much work that we do, scanning, competitor creative, keeping track of competitor pricing, you know, you know, myself and the team, we’re stuck behind computers doing a lot of these sort of

821
02:15:33.500 –> 02:15:39.729
Sandy Diao: Manual, repetitive research and planning tasks. And, you know, we’re kind of in this really cool

822
02:15:39.730 –> 02:16:04.660
Sandy Diao: time now where both managers and executors alike, we can use agents to essentially augment repetitive tasks, like research using agents. So, some of my favorite tools include using, Manus AI, for example. They have scheduling built in, so you can essentially run a research task every single day, every single week at a certain hour. So imagine, like, the simple task of tracking competitor pricing. You can imagine have Manis read through 50 landing pages, or pricing

823
02:16:04.660 –> 02:16:28.450
Sandy Diao: pages on a daily or even weekly basis, have it outline the insights and the changes directly to you, and then you focus your work on getting to distribution much, much faster, and execution, you know, moreover. So that’s kind of my philosophy and my perspective on thinking about how do we use, you know, these agents to, you know, kind of think about building this more defensible, or this moat, if you will, for broader distribution strategies.

824
02:16:29.310 –> 02:16:32.100
maryshea: Yeah, that… there’s so much,

825
02:16:32.340 –> 02:16:47.110
maryshea: content you’ve just shared there that really resonates with me, Sandy, and so I think to maybe summarize and come back to the original question, the distribution strategy is not necessarily a defensible moat, but it can allow you to out-execute your competition, right?

826
02:16:47.110 –> 02:17:11.420
maryshea: I think that’s sort of the big takeaway, and I love that. I’ll share just a little bit of background on our company now, because we’re in the process of unveiling it to the market. We already have done that, and we’ll be doing that commercially in October. But, I’m a former CEO, a former analyst, a former C-suite exec. We have, five people in the company. We all have our own Meerkats, and…

827
02:17:11.420 –> 02:17:26.229
maryshea: We have our team meerkats, and we are… the… the ability for us to, churn out really high-quality work is like nothing I’ve ever seen before, and so while I function as a co-founder and a C-suite exec, I’m also…

828
02:17:26.379 –> 02:17:44.869
maryshea: a social media marketing manager, too, and I could never have done that in the past without, Agentic AI and without all of the incredible tools that are out there, and so I’m very eager to learn more about what’s out there, because I do see it as a huge differentiator on the execution side.

829
02:17:47.020 –> 02:18:03.730
maryshea: So, I think that might be a nice, sort of segue to turn to, Vanessa and Descript, and if you can, talk to us a little bit about how Descript works and blends this human creativity aspect with AI.

830
02:18:03.730 –> 02:18:09.020
maryshea: And yeah, tell us a little bit about that, and then I’ll ask you a follow-on question.

831
02:18:09.590 –> 02:18:33.250
Vanessa Schneider: Absolutely. So, it’s so exciting to hear so many people excited on this call about using, Descript. So, Descript is a tool that allows anyone to either bring in recorded video or generate video featuring, models like Nano Banana or generated avatars, to create content and then edit it using really intuitive tools like text-based editing and drag-and-drop design capabilities.

832
02:18:33.250 –> 02:18:58.019
Vanessa Schneider: But then we also have nifty AI tools that will correct your eye contact, give you an AI green screen, crispen up your sound, stuff like that. We also have an in-app agent that can become your agentic co-editor, and you can slough off a bunch of the work you don’t want to do, or don’t know how to do, and have the agent do it for you. And so when I hear, Mary, you talking about being an executive who also has to produce a bunch of content for your social channels to be

833
02:18:58.020 –> 02:19:17.390
Vanessa Schneider: an authentic spokesperson on behalf of your brand, Descript is perfectly suited for that kind of workflow, because it’s not going to be what you focus the majority of your day or week on, but it means that you can always be creating really distinctive content that you’re happy to post without it becoming a huge time suck, and without having to rely on creative professionals.

834
02:19:18.059 –> 02:19:33.970
maryshea: Yeah, absolutely, and I can’t wait to learn a little bit more. So, tell us a little bit about how you think distribution changes when, you know, you’ve got the collaboration between creators and agents, and this is where…

835
02:19:33.969 –> 02:19:45.620
maryshea: you know, the bulk of my work is, both, you know, as a co-founder in scaling up Meerkat, but also, you know, as a thought leader in that piece of my profession. I’ll never let go, and I always just…

836
02:19:45.620 –> 02:19:51.129
maryshea: see this human-machine collaboration as something that’s very exciting. So tell us how that… how that works.

837
02:19:51.440 –> 02:20:13.429
Vanessa Schneider: Yeah, absolutely, and as we all know, collaborating with agents is truly the only way teams can keep up with the demand for content right now. So there are two converging trends at play. First, because every brand and executive and media conglomerate is vying for our attention. The volume of content being served up to us is truly nuts.

838
02:20:13.430 –> 02:20:32.239
Vanessa Schneider: And then second, the expectations we have for the content we consume at work are increasingly shaped by the content we consume at home. So watching YouTube, looking at social media, joining a conversation on Reddit, it doesn’t matter that we’re at our desks, we are expecting content that is relevant and that provides unique access.

839
02:20:32.300 –> 02:20:55.369
Vanessa Schneider: And so the net of this is that enterprises need to make loads more content, and it needs to actually be good. So your prospects want to learn about your offering with engaging videos, your customers want authentic and personalized tutorials, your colleagues want to be entertained when forced to watch their mandatory compliance training. So to have a fighting chance of making all this stuff and having it land.

840
02:20:55.410 –> 02:21:14.110
Vanessa Schneider: Teams need to bring agents into their creative workflows. And in my experience, in our own team, and what I see when teams use Descript, that workflow is a continuous back and forth. So, teams know their message, their offering, their audience better than anyone. They’re not going to be outsourcing that to AI.

841
02:21:14.110 –> 02:21:14.980
maryshea: Right, right.

842
02:21:14.980 –> 02:21:32.849
Vanessa Schneider: AI can help them ideate on concepts, it can help them come up with, like, a million throwaway versions of copy or visuals, and it can be great for first-pass editing, so whether it’s doing, like, a first edit or a rough cut, and AI is great for doing scut work, like removing filler words from a video that no creator.

843
02:21:32.850 –> 02:21:33.570
maryshea: 15.

844
02:21:33.570 –> 02:21:43.439
Vanessa Schneider: wants to do. So teams probably won’t hand over their final edits to AI, but they can use it to be more independent and more ambitious in their work.

845
02:21:43.440 –> 02:21:53.919
Vanessa Schneider: And it’s a great way to plug skill gaps. So, like, if someone on the product marketing team, or the sales team, or the support team wants to create something really visually distinctive, but they don’t have a design background.

846
02:21:53.920 –> 02:22:01.289
Vanessa Schneider: that’s no longer a problem. You can really make, like, a one-pager and a demo deck, and a launch video all just by prompting.

847
02:22:01.670 –> 02:22:26.660
maryshea: Yeah, it’s incredible, and I’ll just thank you for that, and I’ll just share a little story about what I did yesterday, which was I created… we have a Final Frame Friday, where we create sort of light, animated videos, and we have a digital guru who is a part of our exec team as well, but he’s quite busy, and so I, went to our agent and created the visuals, and I’m not gonna spill the beans, but it’s exciting, it has a

848
02:22:26.660 –> 02:22:30.639
maryshea: pirate theme, for tomorrow, so look for that out there on LinkedIn.

849
02:22:30.640 –> 02:22:42.639
maryshea: And I was actually able to go into, Google AI Studio and share my screen and have the AI help me, refine my

850
02:22:42.640 –> 02:22:58.600
maryshea: prompts better so that I could get the design effects that I wanted, because when I was doing it without that, it added an extra 90 minutes to 2 hours to it. So, seeing every day I’m learning something new, and, that was just really outstanding. So I could hand a very,

851
02:22:59.280 –> 02:23:12.769
maryshea: clear video script, along with all the visuals I had created, and then our designer will take it across the line, and, you know, if you told me 2 months ago I’d be doing something like that, I would have said, you’re crazy. So, it’s… it’s a really exciting time, isn’t it?

852
02:23:12.940 –> 02:23:24.110
Vanessa Schneider: It’s such a great feeling, and it invites a sense of play that I think is really refreshing, like, day to day, when you’re sitting at your desk. Like, the fact that you can make stuff that surprises you is such a unique pleasure.

853
02:23:24.320 –> 02:23:27.999
maryshea: Yeah, it really is. It was… I mean, it’s so great, because as…

854
02:23:28.190 –> 02:23:42.760
maryshea: as a C-suite exec, I’d always have to, like, have… tell my ideas to some person, then I’d have to go 5 levels down, and then I’d have to go find that person, and now it’s, like, happens almost instantaneously, and I can do it myself, and it is such a…

855
02:23:42.760 –> 02:23:49.460
maryshea: a liberating feeling. So yeah, I’m pretty excited. But before I leave you, Vanessa,

856
02:23:49.460 –> 02:24:08.109
maryshea: Tell us a little bit about reach, and, you know, it’s getting… I heard you mention Reddit, and it can be time-consuming if you don’t kind of have a good system to hit everywhere you want to hit, and I’ve been experimenting a little bit, trial and error, so, you know, LinkedIn obviously is my first stop, and then LinkedIn company page, and Facebook, which is…

857
02:24:08.110 –> 02:24:14.680
maryshea: pretty low impact, and then I’ve been hitting blue sky for my own, you know, reasons. I don’t want to use Twitter, and

858
02:24:15.010 –> 02:24:23.260
maryshea: So, it’s a little trial and error, but talk to us a little bit about amplifying Reach and how you think about that, how to spread the content efficiently.

859
02:24:23.580 –> 02:24:32.359
Vanessa Schneider: Yeah, absolutely, and as Sandy mentioned, I think right now we’re all drinking from the AEO fire hose. LLMs definitely change how content spreads.

860
02:24:32.360 –> 02:24:43.890
Vanessa Schneider: We have this new surface that puts a slate of citation sources into the spotlight that haven’t been there before. My personal shorthand for this is long form, didactic.

861
02:24:43.890 –> 02:24:44.820
Vanessa Schneider: content.

862
02:24:44.820 –> 02:24:58.659
Vanessa Schneider: So forums, review sites, FAQs, explainer videos… LLMs are going to these sources to pick up credible and complete answers to the questions people are typing into their prompt box.

863
02:24:58.770 –> 02:25:07.190
Vanessa Schneider: But for me, what’s even more exciting than how agents are changing the way content spreads is how agents can actually help all of us amplify our reach.

864
02:25:07.190 –> 02:25:23.249
Vanessa Schneider: Ai is really great at helping you scale your content operation. It can help you reach global audiences with translation, or translated voiceover, or lip-syncing video to authentically deliver your message to a global community without forcing them to read subtitles.

865
02:25:23.250 –> 02:25:37.539
Vanessa Schneider: We’ve all seen examples of how AI is great at creating media. We talked about Nano Banana, it’s great for images, avatars, custom B-roll, and this means that teams don’t have to rely on stock libraries to tell a story that’s really ownable.

866
02:25:37.540 –> 02:25:48.910
Vanessa Schneider: And AI is really smart at helping you repurpose your content. You can get more out of everything you make. So one of my favorite ways to use the Agentic co-editor inside Descript is to give it a long video.

867
02:25:48.910 –> 02:26:09.719
Vanessa Schneider: and have it find the best clips for social, exactly like what you’re talking about, Mary. And it’s so much easier to scale and repurpose your content with AI, so you get more juice out of every asset that you invest in creating. So, I think that reach, with new audiences, with global audiences, comes from allowing AI to get into the process to help you scale and repurpose.

868
02:26:10.350 –> 02:26:16.130
maryshea: Awesome. Thank you so much for that, Vanessa. And, Mark, I’d love to turn to you.

869
02:26:16.520 –> 02:26:19.240
maryshea: Let’s talk a little bit about,

870
02:26:19.570 –> 02:26:32.669
maryshea: Use cases. So, what are the use cases that you’re most excited about as you think about agentic distribution, and, you know, what lessons have you learned along the way that would be beneficial to folks listening in today to the session?

871
02:26:33.240 –> 02:26:54.539
Mario Moscatiello: Yeah, totally. I think there is definitely a lot there, and I can empathize, I think, with a lot of what has been said so far. I think Ashley and I were at a dinner together, this was, like, 2 or 3 weeks ago, and we literally sat next to each other, and we’re like, oh my god, you can actually do so many things now, you know, with the Gentin distribution and AI. One of the things that I’m very, very excited about is the SDR role, in that sense, like, I’m…

872
02:26:54.540 –> 02:27:02.169
Mario Moscatiello: hot take for me is, like, I don’t believe in AI SDRs. I actually think, the best, sort of, like, way of handling it is by

873
02:27:02.170 –> 02:27:03.380
Mario Moscatiello: turning every…

874
02:27:03.380 –> 02:27:27.830
Mario Moscatiello: every reviewer’s SDR into superhumans, essentially, and that’s, like, sort of, like, what I looked at. We were talking about metrics, is stuff like, can we have them book double the meetings? Like, can they spend 90% less time doing, like, repetitive tasks, and so on and so forth. And I think, like, the SDR, it’s… I started my career as an SDR, like, it’s the hardest job in tech, like, and it’s harder than anything. You’re being told no a thousand times a month, you know, you’re doing the same thing over and over.

875
02:27:27.830 –> 02:27:52.779
Mario Moscatiello: And so what I did was, like, taking a bit of, you know, sort of, like, the Jeff Bezos principle of, like, he was delivering books the first year of Amazon, and so what I did is I spent an entire week… I have a go-to-market engineer, and so, like, we spent, like, you know, so much time sitting with the BDRs, understanding the workflows, and understanding, like, what are they actually doing? And so… and then you start looking at what are the most successful ones doing, like, on a daily basis, and what are, like, the least productive ones doing.

876
02:27:52.780 –> 02:28:06.480
Mario Moscatiello: And you really realize that a lot of… a lot of that is, like, sort of, like, a lot of yes and no decisions, or, like, sort of, like, qualifying and so on and so forth, and it turns out AI is really good at making yes and no decisions. And so, it’s about, like, taking, like.

877
02:28:06.480 –> 02:28:20.240
Mario Moscatiello: sort of, like, what if the SDR role didn’t exist, and I would be reinventing from scratch? And so it’s really about that. It’s really about, like, using AI, leveraging all the tools, and so on and so forth, you know, to make their experience better, help them be more successful, and ultimately.

878
02:28:20.240 –> 02:28:37.050
Mario Moscatiello: increased ROI, for the company. I think something, you know, someone said before is, like, we’re all fighting for attention, and of course, like, GEO and all of it, I think brand is becoming, like, such an important thing. And so, when it comes to outbound, like, it’s the same thing. With AI, outbound exploded, everybody’s doing this, you know, huge,

879
02:28:37.050 –> 02:28:49.989
Mario Moscatiello: campaigns, and I think, like, helping SDRs, like, be more prepared, solve all of the account research, and helping them automate 90% of their work is something that is really exciting to me. And, you know, it’s very quick to CRI, I think.

880
02:28:50.030 –> 02:28:52.199
Mario Moscatiello: The biggest lesson for me was, like.

881
02:28:52.880 –> 02:28:56.330
Mario Moscatiello: don’t fully build it before you’ve tested enough, you know.

882
02:28:56.330 –> 02:28:56.940
maryshea: We use…

883
02:28:56.940 –> 02:29:21.870
Mario Moscatiello: As world people, we always like to diagram and use the whimsicals and draw everything down and really build all of these complex and over-engineered systems. And he’s like, instead of doing that, we should really focus on understanding what’s working, what’s not working, and then we go and scale it. And I think that that was the biggest learning. I, you know, instead of investing in 15 tools and automating everything and wanting everything to be precise, I’m like, okay, maybe we can just upload a CSV, you know?

884
02:29:21.870 –> 02:29:34.800
Mario Moscatiello: And what we care about is, like, is the messaging resonating? Are we building the right context engine for them to sort of, like, not have to think about the company too much? Like, are we feeding them the right case studies, and so on and so forth? And all of that can be automated.

885
02:29:34.910 –> 02:29:42.820
Mario Moscatiello: For them, and so, yeah, I think it’s, I think that’s definitely one of the biggest lessons. You know, as you scale, you kind of want to keep.

886
02:29:42.820 –> 02:30:07.489
Mario Moscatiello: a flexible canvas for executing, and I think that makes a total difference. At the end of the day, what we’re trying to do is really trying to understand why our customers are buying from us, and what resonates and what doesn’t. And so, like, I’d rather spend the time optimizing the messaging and optimizing the prompting of the AI to get that right, versus figuring out how to, you know, how the leads flow from Salesforce to Clay and so on and so forth. We have to do that as well.

887
02:30:07.490 –> 02:30:18.460
Mario Moscatiello: But, you know, ops, like, good ops, like, is… I think people over-index on it, like, it’s something that I’m really proud about, like, we want to have all of the workflows set up and everything, but ultimately, it’s like.

888
02:30:18.750 –> 02:30:22.009
Mario Moscatiello: Can we look… how fast are we learning, you know, in that sense, so…

889
02:30:22.480 –> 02:30:33.039
maryshea: Yeah, thank you for that, and I will add that I started my business career as an SDR as well, so, that population is really, really close to my heart, and I’d love to just

890
02:30:33.230 –> 02:30:57.669
maryshea: dig in a little deeper here. I think you may have answered the question, but I want to make sure I hear directly from you. You know, what… what… so, number one, you’re saying the SDR is not dead, that role will be around. I assume there will be many less of them, right? And they will be technology optimized, but what… how do you see, you know, the disposition of that role changing? Because it’s not going to be the same role as

891
02:30:57.790 –> 02:31:08.399
maryshea: when you or I would have been doing it. And so, what are the key criteria that a great SDR is going to have in this time of agentic distribution?

892
02:31:08.770 –> 02:31:17.379
Mario Moscatiello: Yeah, I think, like, some of the interesting things is, and look, my entire SDR team is great. I think since we started doing outbound, like, nobody has missed a quarter, everybody is really.

893
02:31:17.380 –> 02:31:17.750
maryshea: Wow.

894
02:31:17.750 –> 02:31:41.719
Mario Moscatiello: But what you’re seeing is that the gap in between the… and I’m saying it in the best way, the gap in between the experienced SDRs and the SDRs that are coming out of college, it’s getting smaller and smaller. And so what you optimize for is actually building a team of athletes, versus, like, building a team of experts. And this is something that I think really resonated with me. I’m stealing from somebody, like, this is not, you know, my invention by any means, but I think it’s, like, what I’m

895
02:31:41.720 –> 02:32:05.430
Mario Moscatiello: optimizing is for endurance, performance, and so on and so forth. It’s like, what is the grid? Like, why are they willing to put in so much time? I think when it comes to outbound pipeline, it’s the time you put in, is the effort, is the pipeline you get out, at the end of the day. Of course, with technology and everything, hopefully we uplift a lot of the metrics, but I’m optimizing less for experience and more for, like, attitude and, like, who these people actually are. And so, like, that’s why I get excited, because

896
02:32:05.430 –> 02:32:18.729
Mario Moscatiello: it means that, you’re basically, like, the… you’re leveling up, like, you’re leveling the playing field, and people are not failing because they are not learning how to use, you know, some… these crazy tools, you know? And I think that that’s the other thing is,

897
02:32:18.730 –> 02:32:42.230
Mario Moscatiello: I think, like, a lot of the tools and a lot of, like, what’s being built is not actually built for SDRs or reps in general, like, for AEs. This is built for group people that are super huge nerds and can figure out, like, all of the scoring and the algorithms and everything. Like, these people just, like, want to wake up in the morning and know that those are the 100 accounts they need to focus on, and that’s it. And so, like, the more we can help them do that and think less about everything else, I think the more successful they will be.

898
02:32:43.170 –> 02:32:57.600
maryshea: Yeah, I agree with you, and, I think, sort of, you know, sort of pulling out some of the key themes, you’re saying, you know, have some patience, don’t go all you can eat immediately, as you think about scaling, so,

899
02:32:57.670 –> 02:33:08.140
maryshea: So, less is more in some ways, and then once you’ve got things nailed down, you continue to add in more tools and increase your capabilities. Is that accurate?

900
02:33:08.140 –> 02:33:20.549
Mario Moscatiello: Yeah, there is also… that’s, I think, the first principle. I think the second principle is also aligning the entire go-to-market organization. Like, do we know… like, it’s not even do we know what our ICP is? Do we have a shared understanding of what

901
02:33:20.660 –> 02:33:44.059
Mario Moscatiello: what ICP even means. Like, when we talk about use cases, what are use cases? You know, like, especially selling horizontal products. When you sell vertical products, I think it’s… you have a smaller… you tend to have a smaller TAM, but the use cases and the personas are more defined. When you’re selling a horizontal product, like, you just want to make sure that the entire go-to-market organization has a sheer understanding, because otherwise it’s… it’s going to create friction in the process. And then once you go and scale.

902
02:33:44.060 –> 02:33:46.939
Mario Moscatiello: It’s harder to, like, rip and replace some of those, like.

903
02:33:46.990 –> 02:33:51.669
Mario Moscatiello: notions and context, like, within the intelligence layer you’re creating, essentially, so…

904
02:33:52.110 –> 02:33:58.610
maryshea: Yeah, awesome. Thank you for that. And, I want to get to Josh, but I also,

905
02:33:58.880 –> 02:34:07.770
maryshea: being slightly selfish here at the moment, I want to put a pin in Let’s Talk About PLG after we, talk to Josh for a few minutes, because

906
02:34:07.770 –> 02:34:25.720
maryshea: So many times when we think about scaling and growing, it’s always, like, in the B2B, in the enterprise, and we’re starting out with PLG, so I want to make sure folks who are growing their businesses, at least initially, with that motion, that we can get some of your expertise as well. So think about that as a topic, and I’ll come back and we’ll do a round robin.

907
02:34:25.720 –> 02:34:32.979
maryshea: But Josh, welcome, I’m glad you made it, and let’s talk a little bit about the role of customer success, and

908
02:34:32.980 –> 02:34:46.849
maryshea: You know, it’s interesting, I have a lot of friends who lead customer success organizations, folks that I’ve hired over a period of years, and I have so much admiration for the role, and then at the same time, you see,

909
02:34:46.890 –> 02:34:50.019
maryshea: you know, what happened over at Salesforce, and

910
02:34:50.530 –> 02:34:56.919
maryshea: different folks in that, position losing their jobs, and so I’d love to…

911
02:34:57.020 –> 02:35:10.140
maryshea: Hear from you on how you see this role and customer success teams evolving, and where and how do the agents, play into that, evolution?

912
02:35:10.540 –> 02:35:19.830
Josh Schachter (Gainsight): Yeah, yeah, thank you, thank you for having me. I mean, it’s been a couple years now before the boom of Agentic that customer success has been under… we can call it a little strain, right?

913
02:35:20.320 –> 02:35:32.739
Josh Schachter (Gainsight): and kind of self-identity, and figuring out exactly where… I mean, CS knows where… how much value it adds. I think CS has always kind of been trying to prove itself, so that others really fully understand that value that it contributes.

914
02:35:33.200 –> 02:35:40.020
Josh Schachter (Gainsight): And I think that now in this world of Agentic, the role of the CSM doesn’t shrink, it actually elevates.

915
02:35:40.150 –> 02:35:58.479
Josh Schachter (Gainsight): Because, you know, we can now use the CS manager and the CS team to really be trust brokers to help deliver the outcomes. You know, CS has been going towards making sure that they can deliver verifiable outcomes and map to those outcomes very specifically. When we think about

916
02:35:58.680 –> 02:36:15.089
Josh Schachter (Gainsight): what’s happening in the world of SaaS right now, the quote that we latch onto, actually, is from about a month ago. Sam Altman, I think it was on X, he tweeted, or X’d, whatever they call it these days, that we’re entering the fast fashion of SaaS. Easy come, easy go.

917
02:36:15.090 –> 02:36:16.750
maryshea: I have not heard that.

918
02:36:16.750 –> 02:36:17.440
Josh Schachter (Gainsight): Yeah.

919
02:36:17.440 –> 02:36:19.809
maryshea: Like, it’s very, very catchy, isn’t it?

920
02:36:19.810 –> 02:36:33.590
Josh Schachter (Gainsight): Totally, and really appropriate, too, because you talked about… somebody earlier was mentioning how, you know, everybody can buy code, and every Y Combinator company is coming out with… it’s not the Uber of X anymore, it’s the agentic of X, right?

921
02:36:33.660 –> 02:36:35.780
Josh Schachter (Gainsight): So, whether it’s from…

922
02:36:35.830 –> 02:36:48.290
Josh Schachter (Gainsight): as SaaS companies, whether it’s this burgeoning or mushrooming, or crowded space of all these other up-and-comers, that are trying to kind of eat up incumbent market share.

923
02:36:48.290 –> 02:37:00.040
Josh Schachter (Gainsight): or whether it’s, you know, internal build programs, where coders can now use Claude Code and Cursor and these other tools to really kind of help them ramp things up, and then they can build internally.

924
02:37:00.080 –> 02:37:04.590
Josh Schachter (Gainsight): Whether it’s from external or internal, SaaS companies are under greater attack.

925
02:37:04.760 –> 02:37:08.119
Josh Schachter (Gainsight): And so, you know, what we believe is that

926
02:37:08.330 –> 02:37:20.029
Josh Schachter (Gainsight): you need to now treat every customer as if they’re your best customer. It’s no longer good enough to say, okay, we’re gonna bear hug the top 20% of customers that represent 80% of our revenue. You still have to do that.

927
02:37:20.210 –> 02:37:29.249
Josh Schachter (Gainsight): But you can’t assume anymore that your mid-tier and your long-tail are just gonna stay, you know, out of inertia and out of auto renewals. So, that’s where we believe that Agentic.

928
02:37:29.350 –> 02:37:44.169
Josh Schachter (Gainsight): and AI are most helpful. There’s this orchestrated approach, of course, that feeds back to, you know, our ecosystem and how we’re building our orchestration of AI at Gainsight, but there’s this approach of,

929
02:37:44.560 –> 02:38:00.680
Josh Schachter (Gainsight): you know, high touch, your most important customers, that’s really human-led. You’ve got your mid-touch, that’s the digital-led, which is amplified through AI, and then you’ve got your long tail, and that’s where previously you weren’t managing those customers, but now through Agentic, you can.

930
02:38:00.870 –> 02:38:07.930
Josh Schachter (Gainsight): So, one of the most popular use cases that now we’re working with customers on Is our renewal agent.

931
02:38:08.020 –> 02:38:15.320
Josh Schachter (Gainsight): Where previously, you couldn’t put bodies on the ability to, to, you know, to reach out.

932
02:38:15.340 –> 02:38:33.529
Josh Schachter (Gainsight): in a very personalized way to the long-tail customers that each individually, maybe not… are not as valuable to you as a company, now Agenta can do that. You can put them on a phone call, an email sequence, make it very personalized, answer their questions, connect it to the CPQ, send them out, you know, the order form, close the loop, and be done with it.

933
02:38:33.540 –> 02:38:53.320
Josh Schachter (Gainsight): So, you know, that’s where we see Agentic as being most helpful, is in that long tail, but then even as you kind of go upward into the mid-tier segments and the more human-led segments, this is where AI and agents are helping, of course, to take the busy work away from CSMs, get them closer to, you know, to really

934
02:38:53.320 –> 02:38:56.440
Josh Schachter (Gainsight): Being fully strategic and fully outcomes-driven.

935
02:38:57.180 –> 02:39:18.139
Josh Schachter (Gainsight): You know, helping them do the work of prepping for their QBRs and EBRs in advance, making sure that they have the full handoff from the sales team, that when a new customer onboards, and they know exactly what the scoop is, and can be most strategic in that learning. You know, delivering to them all the different signals that are coming in about the customer, so they know at all times which customers need to, you know, have a watchful eye on them.

936
02:39:18.140 –> 02:39:30.919
Josh Schachter (Gainsight): So that’s where we, you know, that’s where I see, the most popular use cases around AI and Agentic, is both of the long-tail automating, but then at the higher touch segments, being able to really augment and elevate the role of the CSM.

937
02:39:31.510 –> 02:39:39.409
maryshea: Yeah, thank you, thank you so much for that. You know, I know we tend to have an executive-level audience here,

938
02:39:39.650 –> 02:39:40.929
maryshea: with this group.

939
02:39:40.980 –> 02:39:56.440
maryshea: But I spend a lot of time helping folks find jobs and get jobs. I’m at a certain point in my career where I have a very large network, and I feel the pain of folks that are churning out of different types of roles and all of the change that’s happening out there, and so I’m

940
02:39:56.440 –> 02:40:14.469
maryshea: really focused on helping as much as I can, and Meerkat will have a feature that will help seekers, too, in the not-too-distant future, so I’m excited to share that. But what kind of advice would you give CS professionals, you know, ICs, managers, directors.

941
02:40:14.470 –> 02:40:22.380
maryshea: To really make sure that, you know, they’re the ones that are thriving in the next… Three years.

942
02:40:22.930 –> 02:40:30.690
Josh Schachter (Gainsight): Yeah, yeah. Their roles are going to… they already are becoming, but they will continue to become more strategic.

943
02:40:30.710 –> 02:40:50.419
Josh Schachter (Gainsight): more revenue-focused, we know that that needs to happen, and more outcome-driven, right? So as a leader, you need to make sure that you’re… I just came from a conference yesterday, actually, and these are the… these are the kind of the tenets that everybody was talking about, is to set up your organization so that, you know, you are proving those outcomes, and getting as close to the money trail as possible.

944
02:40:50.420 –> 02:40:56.889
Josh Schachter (Gainsight): I think at a individual looking at, you know, job security and the future of my role perspective.

945
02:40:57.030 –> 02:40:59.120
Josh Schachter (Gainsight): Actually, whether you’re a leader or not.

946
02:40:59.520 –> 02:41:01.550
Josh Schachter (Gainsight): You’ve got to be AI native.

947
02:41:01.670 –> 02:41:08.869
Josh Schachter (Gainsight): You’re only gonna lose your job… you’re not gonna lose your job to the machine. You’re gonna lose your job to the person who knows how to use the machine better than you do.

948
02:41:08.870 –> 02:41:11.739
maryshea: We’ve been all saying that for so many years now, yep.

949
02:41:11.740 –> 02:41:16.919
Josh Schachter (Gainsight): Yeah, it hasn’t changed, it’s just a different, different technology, right, stack in some ways.

950
02:41:17.020 –> 02:41:19.250
Josh Schachter (Gainsight): And so,

951
02:41:19.300 –> 02:41:32.089
Josh Schachter (Gainsight): the… that’s what I see out there in the market when we speak to people, is that the folks who are most effective are the ones that are just diving in. They’re playing, it’s shadow… it’s no longer shadow IT, it’s shadow AI.

952
02:41:32.090 –> 02:41:50.029
Josh Schachter (Gainsight): Right? But they’re building up those workflows on their own, asking for forgiveness, not for permission. Leaders are doing the same thing. By the way, leaders themselves, you know, even at the CCO level, I spoke recently to Sophia Barbosa, she’s the CCO of BMC, 5 to 10,000 person company, major enterprise.

953
02:41:50.510 –> 02:41:51.380
Josh Schachter (Gainsight): And…

954
02:41:51.580 –> 02:42:10.669
Josh Schachter (Gainsight): she took that role recently, there was a split in the org, she had basically, you know, 80% of the same amount of work, 50% of the talent, because it was split. And so, she devoted her efforts on building an agentic platform within the organization that could do all of these different things, you know, that she needed to be done to save her team time and efficiency.

955
02:42:10.740 –> 02:42:22.660
Josh Schachter (Gainsight): And so, even there, you know, like, she didn’t have to take that effort to really kind of, you know, prioritize this, but she was forward-thinking enough to do that. And I speak to leaders at smaller companies, Series A company, you know, types of folks.

956
02:42:22.670 –> 02:42:39.899
Josh Schachter (Gainsight): And they carve out, you know, their Friday afternoons, or whatever the case might be, to play around with AI, to build their own agents, to coach their agents. Like, one leader told me, you know, it’s like having a one-on-one, a standing one-on-one with your agent, just to fine-tune it in many ways.

957
02:42:40.360 –> 02:42:40.790
Josh Schachter (Gainsight): Right?

958
02:42:41.180 –> 02:42:44.589
Josh Schachter (Gainsight): I think that leaders themselves, you’re going to see this in CS.

959
02:42:44.590 –> 02:42:48.340
maryshea: No, I do it, I do it already. I mean, it’s unbelievable.

960
02:42:48.680 –> 02:43:00.500
Josh Schachter (Gainsight): Yeah, you have to, you have to. You have to keep the wheels greased. And, and leaders are… I’ll just… I’ll end on this final point, is that, you know, so we know that ICs need to embrace it, but leaders,

961
02:43:01.260 –> 02:43:13.239
Josh Schachter (Gainsight): they’re soon… they’re going to be managing not only people, but managing their agents, looking at a dashboard that compares the performance across their agents, the performance of their people, right? They might be doing two different things.

962
02:43:13.240 –> 02:43:22.479
Josh Schachter (Gainsight): But that’s going to be part of their control tower, their bird’s eye view, is understanding how to architect and orchestrate the entire human plus agentic motion.

963
02:43:22.930 –> 02:43:38.450
maryshea: Yep, yep, I completely agree. I do have one more question for you, and then I want to hit… we’ve got a couple of audience questions, so I want to get there, and I’m definitely getting back to the PLG, so don’t… don’t worry about that. But, you know.

964
02:43:38.450 –> 02:43:49.939
maryshea: We’ve all seen the MIT report that came out, and I was talking to my friend Seth Mars the other day about, the lack of ROI, and so on and so forth, and my personal theory is that

965
02:43:50.080 –> 02:43:54.750
maryshea: The reason we’re seeing, sort of, negative ROIs right now

966
02:43:54.750 –> 02:44:10.330
maryshea: There’s a couple reasons. One is that the technology is maturing faster than we as a society, both on the work front as well as the personal front, can adopt and ingest it. So that’s one thing. But I think there’s another thing at play, which is…

967
02:44:10.820 –> 02:44:22.369
maryshea: Some people, maybe in some of the larger and enterprise companies where you’re seeing slower adoption, because the adoption at startup levels at my company, it’s like, we are so heavily using it, but,

968
02:44:22.840 –> 02:44:39.039
maryshea: Is it maybe because people are seeing this as a zero-sum game, which is, like, either the AI’s gonna take my job, or I’m gonna block the AI out of my company, and I’m gonna hug my job for another 2 or 3 years, and, you know, like, there’s this jockeying right now, where there’s a power play.

969
02:44:39.210 –> 02:44:47.439
maryshea: Is that what’s causing some of the adoption issues, or what causes adoption on the CS side?

970
02:44:47.440 –> 02:44:51.889
Josh Schachter (Gainsight): I actually don’t see it that way, to be honest with you. I… Okay. I see,

971
02:44:52.340 –> 02:44:59.480
Josh Schachter (Gainsight): ICCS teams, CSMs, fully embracing AI. Cool. I don’t yet see them embracing agentic.

972
02:44:59.670 –> 02:45:10.160
Josh Schachter (Gainsight): I think that’s still new. I spoke at the CS Summit in San Francisco two days ago, and asked for a show of hands of who knows… actually, it’s interesting.

973
02:45:10.160 –> 02:45:20.790
Josh Schachter (Gainsight): I asked for a show of hands of, who… who’s heard of Vibe coding? Who knows about Vibe coding? And this is in… this is at the Hyatt at San Francisco airport, so, like, we’re in the hub of tech.

974
02:45:20.860 –> 02:45:23.530
Josh Schachter (Gainsight): And I got, like, 15% hands raised.

975
02:45:24.040 –> 02:45:25.059
maryshea: That’s shocking!

976
02:45:25.060 –> 02:45:30.219
Josh Schachter (Gainsight): Yeah, and it killed my follow-up questions, because it was going to be a double-click and projecting, but there was, like, no sample size left, right?

977
02:45:30.220 –> 02:45:31.429
maryshea: There’s nowhere to go.

978
02:45:31.430 –> 02:45:44.719
Josh Schachter (Gainsight): There’s nowhere to go, yeah. So we’re not there with Agentic yet, we’re there with AI. Everybody, you know, a year ago, it was half the org that embraced it, but now I think everybody is. And I don’t think… I think people… there was a little bit of a fear in what they didn’t know.

979
02:45:44.720 –> 02:45:52.739
Josh Schachter (Gainsight): With AI, but now, at least in the CS org, I don’t see that same fear. When… when… when we see that there’s speed bumps in

980
02:45:52.950 –> 02:46:04.200
Josh Schachter (Gainsight): in AI deployment and integration and getting the full ROI, it’s honestly usually because of data availability. In order to make the most of Agentic and AI, you know, it feeds on

981
02:46:04.390 –> 02:46:19.089
Josh Schachter (Gainsight): the oil that is the data available through all the different systems out there. And when you get to larger organizations, that’s harder to do in a very clean, consistent way, and so that, I think, is something that’s maybe kind of slowed down organizations a little bit.

982
02:46:19.090 –> 02:46:24.690
Josh Schachter (Gainsight): The other thing is, and I don’t know if this has slowed folks down or not, but this is more of a caveat, is…

983
02:46:24.690 –> 02:46:39.579
Josh Schachter (Gainsight): again, at least in customer success, when you are thinking about mapping out an agentic workflow, which can be hugely valuable, make sure that you have the process, as a human would do it, documented, right? Because the agent can really mimic your best human in many ways.

984
02:46:39.590 –> 02:46:50.410
Josh Schachter (Gainsight): And so, that’s something that we work… like, when we first, you know, start working with our customers, it’s, you know, let’s whip out a Miro board, and let’s do a workshop together to make sure it’s fully mapped out.

985
02:46:51.010 –> 02:47:14.070
maryshea: Thank you for that really thoughtful answer, and I completely agree that it’s, you know, especially with some of these larger companies, we’re… and I found this when I was at outreach, where if you don’t have your workflows buttoned down, you can buy, you know, the Cadillac of execution tech, and it’s still not really going to work. So, I think that’s very fair, in addition to

986
02:47:14.070 –> 02:47:25.459
maryshea: maybe some of the social, visceral implications that I was bringing on earlier, so we’ll have to see, right? But Julia, is this a good time to switch to one of the audience questions?

987
02:47:25.460 –> 02:47:29.880
Julia Nimchinski: Yeah, definitely such an amazing panel. Do you want me to read it, Mary, or…

988
02:47:29.880 –> 02:47:31.189
maryshea: Yeah, go for it!

989
02:47:31.190 –> 02:47:44.189
Julia Nimchinski: Yeah, so our audience is asking about your thoughts, all of our panelists, about the LLMs and monetization of visibility, both on the human side and the AI bot side.

990
02:47:47.520 –> 02:47:49.040
Julia Nimchinski: What are your thoughts, Cindy?

991
02:47:51.580 –> 02:47:55.580
Sandy Diao: Yeah, I’ll kind of jump in with my own hypotheses. You know, obviously I don’t want to be.

992
02:47:58.870 –> 02:47:59.440
Julia Nimchinski: Oop.

993
02:47:59.730 –> 02:48:00.730
Julia Nimchinski: A live TV!

994
02:48:00.730 –> 02:48:05.800
maryshea: may have lost Sandy for a minute, so, yeah, anyone else want to jump in?

995
02:48:11.740 –> 02:48:15.710
Mario Moscatiello: I’m happy to take it. I mean, if you look at Alphabet.

996
02:48:15.780 –> 02:48:35.560
Mario Moscatiello: I think, like, something like 70% of their revenue is ads. And I think… I don’t know if there is a better way, necessarily. I think in the same way, there is gonna be… I think, like, the excuse for people to get, like, a better, you know, ChatGPT for free is gonna be, well, we’re gonna show you ads, and so on and so forth. I think what’s different is that, there is gonna be…

997
02:48:35.560 –> 02:48:40.740
Mario Moscatiello: like, it’s gonna be harder to compete, because there is less real estate available. In that sense. If you think about a page.

998
02:48:41.060 –> 02:48:56.750
Mario Moscatiello: like, Google’s starting, like, with some queries, you have, like, two or three, like, you have one ad or two, and now you have, like, five or six, and then, like, you have a few organic results, and now you have ads below those organic results. I think the real estate in the chat experience is way less, and so at the same time, I think it’s got to be very hard.

999
02:48:56.750 –> 02:48:57.710
Sandy Diao: I’d be very helpful.

1000
02:48:57.710 –> 02:49:19.380
Mario Moscatiello: for these companies, sorry, I’m hearing myself speak. But at the same time, it’s gonna be harder for companies to compete, so I think the quality and the bar is gonna be higher, and so I think the algorithms are just gonna be able… are just gonna be better at figuring out, whether the content that they’re promoting is actually high quality or not, and whether the users are finding success. So I think it’s… it’s gonna…

1001
02:49:19.380 –> 02:49:42.780
Mario Moscatiello: I think it’s still gonna be in the realm of ads. When it comes to bot interaction, we’re already seeing something, you know, Cloudflare announcing features for, you know, so you can get… you can… bots, like, essentially you can get paid, like, you can have an API where if a bot is visiting your website, like, they’re paying for croning the pages and so on and so forth. So, like, there are definitely, like, some hints at monetization when it comes to bots. When it comes to human, I still think it’s gonna be a lot of ads.

1002
02:49:45.020 –> 02:49:55.169
maryshea: Got it. Thanks for that, Mario. And Sandy, we lost you for a moment, in the excitement of the question. Do you want to weigh in here as well?

1003
02:49:55.600 –> 02:50:18.299
Sandy Diao: Oh, yeah, just gonna quickly say, I was gonna put on my monetization hat, kind of having observed how a lot of these social apps have monetized the attention economy, essentially. I think there are, you know, sort of 3 possibilities in my mind, or probably many, but, you know, 3 likely possibilities. One of them, actually, that we’re already kind of starting to see a lot of people care about are the rankings that we see and some of these responses, so when you ask

1004
02:50:18.300 –> 02:50:38.570
Sandy Diao: one of these LLMs for solutions or recommendations, they generally give you a list, right? They don’t give you one response. They tell you, here’s 5 or 10 different options, and I do think that, you know, LLMs are going to sort of figure out, are there opportunities here where certain brands or certain, companies can be mentioned for some of these top slots? That’s one.

1005
02:50:38.570 –> 02:50:43.680
Sandy Diao: I think the second one that, you know, I think every LLM and every player is going to be, sort of.

1006
02:50:43.680 –> 02:50:57.240
Sandy Diao: figuring out and sifting through is just broadly, you know, which identities are most verifiable and trustable. And so, identity management, I think, could be another where you have a verifiable presence, because LLMs, you know, sometimes do have somewhat limited signal around

1007
02:50:57.240 –> 02:51:17.350
Sandy Diao: we see these things referenced on Reddit or Quora or Wikipedia, but, you know, a lot of it as, you know, community or user-contributed, how do we sort of add another layer of stamp and credibility? You see a lot of these companies, especially, you know, sort of broader social platforms, trying to create verifiable identities. Probably the last one, not sure if LMs will ever go down this path, but…

1008
02:51:17.510 –> 02:51:36.560
Sandy Diao: I think that, probably these LLM players are realizing that they’re sending a lot of traffic, and there’s this obvious potential here to potentially charge for links to commercial partners, you know, e-commerce is probably a hot area for this, but, you know, more broadly, I do think that these players are going to be very slow about

1009
02:51:36.770 –> 02:51:57.460
Sandy Diao: launching some of these monetization mechanisms, because that’s going to erode trust really fast, plus the fact that these tools are not just user-facing, but they’re also APIs, and a lot of other companies are relying on their services and outputs as well. I think everyone’s going to be pretty, I think, you know, cautious about testing their way into broad-scale monetization. So that’s… that’s my prediction.

1010
02:51:58.310 –> 02:52:11.419
Vanessa Schneider: I want to jump on to that, if that’s okay, because I really agree with what Sandy just said, and just, like, doing a little A plus B. A moment ago, we were talking with Josh about this sort of trust and adoption headwind that

1011
02:52:11.420 –> 02:52:18.609
Vanessa Schneider: AI enigentic tools face because of the FUD in the conversation. So that’s one…

1012
02:52:18.650 –> 02:52:31.579
Vanessa Schneider: headwind to monetization, and the other is what we were just talking about with, the sort of bad taste in everyone’s mouth about page 1 results on Google. So, if the LLMs are smart to consider these sort of

1013
02:52:31.750 –> 02:52:54.189
Vanessa Schneider: cultural or social forces that are sort of exhausting people. They’re going to have to do exactly what Sandy says, and be incredibly cautious and thoughtful. I think that is all exacerbated by the strange intimacy that I think we can all experience when being in collaboration with an agent over multiple sessions and conversations. It’s just like a more delicate

1014
02:52:54.190 –> 02:53:00.820
Vanessa Schneider: UI? And so I think they… they’re gonna have to be a little more clever than ads.

1015
02:53:02.390 –> 02:53:19.220
maryshea: Yeah, I like that strange intimacy. I think you’ve summed it up. It definitely feels that way, doesn’t it? We’re so reliant on these agents once we see what’s possible. So I’d love to circle back to the PLG topic, and we’ve talked and covered a

1016
02:53:19.410 –> 02:53:33.970
maryshea: a lot of things today. One thing I found is when you’re a founder, you get a lot of sales pitches. So I’ve gotten more sales pitches than I’ve ever gotten, in my life, professionally, and 90% of those pitches are from

1017
02:53:33.970 –> 02:53:39.950
maryshea: Mario, believe it or not, SDR companies looking to help me with lead gen and book meetings, and I’m like.

1018
02:53:39.970 –> 02:53:47.180
maryshea: folks, I’m not booking meetings, I’m trying to, you know, drive user activation, you know, one user at a time. So…

1019
02:53:47.180 –> 02:54:00.999
maryshea: anyone want to jump in and say, how do you think about some of these themes and topics and strategies that we’ve talked about here and modify them for, at least an early-stage PLG company like Meerkat?

1020
02:54:03.530 –> 02:54:05.190
maryshea: Anyone want to jump in?

1021
02:54:05.900 –> 02:54:09.989
maryshea: Including Julia, we’ll put you… we’ll let you jump in, too, if you want.

1022
02:54:16.470 –> 02:54:31.329
Vanessa Schneider: I’ll jump in. Not surprisingly, I’m gonna say video has to be part of your PLG strategy. I think the pressure here with self-serve growth is that you kind of have to be everywhere, and as many of the group has mentioned.

1023
02:54:31.330 –> 02:54:42.910
Vanessa Schneider: The expectation for, personalization and, like, deep, deep relevance to your use case, to your persona, to your motivation, to your pain point is growing. It, it’s not…

1024
02:54:42.920 –> 02:54:56.529
Vanessa Schneider: just for SDRs to know the buyer considerations deeply, your whole PLG motion has to sort of reckon with that complexity. Good thing we all have this agentic tool stack in order to achieve that kind of

1025
02:54:56.530 –> 02:55:04.520
Vanessa Schneider: fragmentation and scale. But I think that, you know, one of the things that we all have to reckon with on the PLG side is

1026
02:55:04.850 –> 02:55:11.530
Vanessa Schneider: Buyers are not interested in hearing from brands. There’s just great suspicion and hesitation around corporate voice.

1027
02:55:11.540 –> 02:55:35.509
Vanessa Schneider: This is why affiliates and influencers are so effective. This is why community-led growth is so effective. And so what we as marketers have to do is sort of teach the teacher, speak to the spokespeople, and give them the tools to tell your brand story on your behalf in a way that speaks to the audience they’ve cultivated. It might be a small audience, so you gotta do a lot more operations and block and tackle out there.

1028
02:55:35.510 –> 02:55:40.470
Vanessa Schneider: But that relationship that they have is gonna be your most potent entry point.

1029
02:55:40.900 –> 02:55:54.940
maryshea: Yeah, I love that, and thank you, because that definitely is confirmatory of my strategy, but… so I feel really good about hearing that, but where I do get overwhelmed, and this may be a session for another day, or one-to-one with any of you separately, is…

1030
02:55:54.940 –> 02:56:08.860
maryshea: there’s so many channels, like, how do I start to figure out where to, you know, where to spend my time on the channels? And I guess the automation in, in, around distribution of those channels is absolutely crucial, because you can’t hit everything efficiently.

1031
02:56:08.860 –> 02:56:18.500
maryshea: So, thank you for that, and Julia just threw me a softball, so I’m gonna end with this, and then we’ll wrap up, with some closing thoughts, but she said.

1032
02:56:18.610 –> 02:56:34.139
maryshea: where is, Meerkat going to be in 2027? And, you know, thank you for that question. The truth of the matter is, we have no idea. We hope it’s a supernova and not a shooting star, that’s what we’re going for.

1033
02:56:34.140 –> 02:56:59.130
maryshea: Meerkat is part of this emergent category that I’m calling, and we’re all calling, AI teammates. Right now, our ICP is really focused on, fast-moving, distributed teams, community leaders, accelerator leaders, founders, startups, those sorts, so anyone that you’re not going to see Microsoft going after, so to speak. The other thing is, we are also ingesting proprietary data into

1034
02:56:59.130 –> 02:57:12.210
maryshea: our LLM into Meerkat. And so, we, in addition to PLG, we expect to have some large enterprise clients, and we just got our first deal yesterday, our first commercial deal, so I can now say we have

1035
02:57:12.420 –> 02:57:35.799
maryshea: 4,500 in ARR. I never thought I’d be so excited about such a small number, but it’s indicative of, you know, getting us off the ground and getting a right start, so we will be continuing to bring in proprietary data sets to provide more and more differentiation for, large enterprise clients. We have a messaging app component, so,

1036
02:57:35.800 –> 02:57:58.650
maryshea: I think there’s professionals and personal people out there who would say, hey, I don’t always want to use WhatsApp for whatever reason, or I don’t want to be tied into big tech exclusively. I’d love to use a company that is focused on bringing more joy into the workforce, more fun. Vanessa and others talked about that. You’re going to see

1037
02:57:58.650 –> 02:58:21.900
maryshea: lots of iconography and videos and ways to actually communicate and do it in a way where all of your data is safe in a cloud behind our firewall, and you’re having a little bit more of the fun. So, the answer is I don’t really know, but I’m very hopeful that, it’s gonna be a really exciting future. So, thank you all for this amazing, panel.

1038
02:58:21.900 –> 02:58:39.169
maryshea: Julia, you do an amazing job of curating. I think this is a perfect group, here, and I want to reach out to several of you independently, one-to-one, because I think there’s, some opportunities that you could help us with our mission, in addition to probably many others who are listening today. So, thank you all for the opportunity.

1039
02:58:39.740 –> 02:58:50.439
Julia Nimchinski: Thank you for the fascinating panel. There were a lot of comments in our HSC Slack, and all of you watching, please join, and join the conversation and ask questions. We will make sure to address them live.

1040
02:58:50.440 –> 02:58:59.930
Julia Nimchinski: If we can just spend one minute, each of you, just promoting your company, your newsletters, whatever it is, just let’s do a quick roundup.

1041
02:59:00.170 –> 02:59:03.310
Julia Nimchinski: Shameless plugs, Josh.

1042
02:59:05.030 –> 02:59:23.040
Josh Schachter (Gainsight): Well, so, we’re building the future of Agentic in customer success and post-sales, and we’re building… we have an adoption agent, a renewal agent, a different constellation of agents, so I’ll promote that. For anybody that wants to learn more, come visit us. I will also promote our podcast.

1043
02:59:23.190 –> 02:59:25.849
Josh Schachter (Gainsight): called Unchurned, which is the top podcast.

1044
02:59:26.580 –> 02:59:34.560
Josh Schachter (Gainsight): success, and we have some really cool guests on the program talking about Agentic and how they’re going through AI transformation for their customer success orgs. Unchurned.

1045
02:59:36.420 –> 02:59:39.310
maryshea: Awesome, I love that. I’m gonna tune in for sure.

1046
02:59:41.070 –> 02:59:41.820
Julia Nimchinski: That’s fun.

1047
02:59:41.820 –> 02:59:43.610
Mario Moscatiello: Oh, sorry, go ahead.

1048
02:59:43.610 –> 02:59:44.090
Julia Nimchinski: Sorry, go ahead.

1049
02:59:44.090 –> 02:59:48.070
Mario Moscatiello: So I’m just gonna have to jump on back-to-back, but…

1050
02:59:48.070 –> 03:00:11.629
Mario Moscatiello: Love to promote AirByte, we’re, you know, data movement platform, we really help companies make sure their first-party data is safe and secure when they train AI models. And we have a big, huge announcement, biggest product announcement we’ve ever done coming up next week, so if you’re in the data space, or if you care about data infrastructure, or it can help solve any of your use cases, like, on the 24th, we’re announcing something really, really big in the market, so I’m super excited about that.

1051
03:00:13.780 –> 03:00:14.500
maryshea: Awesome.

1052
03:00:16.160 –> 03:00:16.980
Julia Nimchinski: Who’s next?

1053
03:00:18.040 –> 03:00:35.690
Ashley Stepien: I’ll jump in. So, currently I lead marketing at Hex, and, we’re on a mission to transform and maybe kill BI as we know it. Bi has been notorious for just static data and, frustrating workflows.

1054
03:00:35.690 –> 03:00:46.689
Ashley Stepien: with your data team to get what you actually need. So, HEX is gonna make everyone a data person, using AI. You’ll be able to access your data, get all the information you need without a static dashboard.

1055
03:00:47.040 –> 03:00:48.080
Ashley Stepien: Check it out!

1056
03:00:50.570 –> 03:01:05.270
maryshea: Fantastic. I’ll jump in, Julia, you’ve given me enough time, so I’m actually just gonna plug my thought leadership series, which is called The Ways We Work, and Meerkat publishes a blog every week, and it’s really designed to focus on

1057
03:01:05.270 –> 03:01:12.100
maryshea: professionals at all levels of their journey, around how AI is shaping the future of work, and so,

1058
03:01:12.100 –> 03:01:19.290
maryshea: what we’re looking for is people who are actually turning in rather than leaning out, and if you want to be a guest writer, DM me.

1059
03:01:19.290 –> 03:01:40.379
maryshea: please check out TryMeerkat.ai, our website, and sign up for our newsletter, which I’ve been told recently is very important to our growth strategy. So that will be coming out in October. But follow us on LinkedIn, we have a company page, and check out us at trymearcat.ai, and let me know if you want to write a blog.

1060
03:01:42.730 –> 03:01:46.229
Julia Nimchinski: Awesome, Vanessa signed Cindy, and let’s wrap it up.

1061
03:01:46.540 –> 03:01:55.659
Vanessa Schneider: Great, we’ve talked a lot about Descript, so all I’m gonna say is, you can make a video, really, you can. And what I want you to do is find me on LinkedIn and send me the first video you make.

1062
03:01:57.220 –> 03:01:58.210
maryshea: It’s coming.

1063
03:02:00.350 –> 03:02:23.619
Sandy Diao: And I will close out here. I’m Sandy, it’s been great to be part of this. I’m gonna plug my sub-stack at Growth Notes, where I write long-form comprehensive essays about a lot of the topics we talked about here today, from AEO and GEO to PLG, to prosumers to go-to-market, and an upcoming piece, actually, on how to use AI agents for some of these growth channels as well, so you can look out for that.

1064
03:02:23.620 –> 03:02:25.830
Sandy Diao: sandydow.subspec.com.

1065
03:02:26.460 –> 03:02:35.819
Julia Nimchinski: Awesome. Thank you so much. Thanks again, Mary. Very excited about what you’re building and evangelizing, and we are transitioning to our next demo.