WEBVTT
00:00:17.600 --> 00:00:21.679
Hi everybody, welcome to this episode of Said It Differently.
00:00:21.760 --> 00:00:31.920
It's the podcast show brought to you by Troop, and it's focused on B2B messaging and storytelling and how you're bringing all of that to market and what you do with it.
00:00:32.159 --> 00:00:42.880
Today's topic is an interesting one for me because I'm seeing and hearing a lot in the market right now around how to bring in more market signal.
00:00:43.200 --> 00:00:51.600
And that is bringing even more data and intelligence to product marketing, who maybe was doing a lot of that DIY in the past.
00:00:51.759 --> 00:00:58.079
So we're going to be diving into like what do we do with all of these different sources of market signal?
00:00:58.240 --> 00:01:04.560
What are the different types of data sets that are out there now that we maybe didn't have at scale before?
00:01:04.719 --> 00:01:13.120
And joining me today for this conversation is longtime friend and colleague, um Constantine Limbarakis, who is with Liberist Consulting.
00:01:13.280 --> 00:01:19.200
He actually brought this topic to me because it was burning a hole in his brain, maybe Constantine.
00:01:19.680 --> 00:01:27.840
But he is an expert product marketer that I've met through the procurement and supply chain space, uh, who really has domain knowledge there.
00:01:28.000 --> 00:01:31.760
But Constantine, please introduce yourself before we dive into the topic.
00:01:32.159 --> 00:01:34.319
Thank you, Jennifer, for having me here again.
00:01:34.480 --> 00:01:36.799
Uh, again, my name is Constantine Limbarakis.
00:01:36.879 --> 00:01:41.760
I am the founder of a consulting firm called Liberus Consulting.
00:01:41.840 --> 00:02:11.280
And what we do is take a lot of our experiences that we've done as practitioners back in the past, as product marketers and as experts in the, as Jen said, procurement supply chain space, and apply that expertise as we've seen the evolution of these areas and these industries, and apply that to the technology providers that require insight to say, well, how do we go to market and how do we best put our foot forward for uh for focusing on what's what what changes lie ahead?
00:02:11.520 --> 00:02:12.240
Yeah, great.
00:02:12.400 --> 00:02:16.639
And I know you also get out there into the market a lot, Constantine.
00:02:16.719 --> 00:02:22.879
You're hosting and moderating a lot of conferences and events in your space between that and your client work.
00:02:23.039 --> 00:02:31.039
So you're certainly seeing and hearing a lot in the market around more tools and more mechanisms for bringing in more market signal, right?
00:02:31.199 --> 00:02:33.680
Like, what was the inspiration for you on this topic?
00:02:33.759 --> 00:02:38.639
Or what was like triggering for you that you felt like, yeah, we need to tackle this?
00:02:38.960 --> 00:02:47.520
Well, you know, one of the things that I just been seeing is just the number of technologies that are out there now that help everybody in B2B, right?
00:02:47.599 --> 00:02:54.960
Whether it's in a particular function or even in the go-to-market space and product marketing and sales automation and marketing automation.
00:02:55.120 --> 00:03:02.319
And, you know, interestingly enough, you know, uh analyst firms like the Gartners of the World are now really trying to hone in on that.
00:03:02.479 --> 00:03:07.759
And I listened to a few of the webinars that they had and they're talking about hype cycles and all this.
00:03:07.919 --> 00:03:16.159
And I thought what's really interesting, and they're saying there's so many of these dynamics of them talking about these different solutions that are out there.
00:03:16.319 --> 00:03:31.680
I can only imagine then when you're a practitioner, you're trying to figure out all these different solutions that are competing with each other, might where you might have sales have one set of solutions, marketing having another set, and then there's this crossover, and you're saying, well, what am I relying on?
00:03:31.840 --> 00:03:38.719
And then so you have, you know, the signal confusion that the product marketer is always trying to orchestrate, right?
00:03:38.879 --> 00:03:46.479
So this just adds to it when you have a lot of different technologies that may or be maybe working together, but maybe not.
00:03:46.639 --> 00:03:49.360
And that's that's kind of why it's a good conversation to have.
00:03:49.680 --> 00:03:50.400
Very timely.
00:03:50.479 --> 00:03:50.879
You're right.
00:03:50.960 --> 00:04:01.439
Yeah, Gartner did just recently unveil its first hype cycle for product marketing, just like late summer 2026 here for those that might be watching this later.
00:04:01.599 --> 00:04:09.360
Um, so yeah, there's a lot coming in, and I would say that a lot of that signal, the noise, um, is is getting noisier.
00:04:09.759 --> 00:04:23.439
I was just at a recent um CMO and PMM summit in San Francisco and just looking around the exhibit hall floor and seeing all of the different vendors that are doing using AI to collect signal.
00:04:23.600 --> 00:04:25.279
Uh, Troop is one of them.
00:04:25.600 --> 00:04:26.639
I will be fast.
00:04:27.199 --> 00:04:32.879
Um and we're focused on messaging and your story and how that signal is playing into things.
00:04:33.120 --> 00:04:49.040
Um, but there are a lot, and then there were a lot of sessions about people, you know, from different presenters and speakers talking about how they're tackling the data and and how they're actually using AI, DIY to get data that they want.
00:04:49.360 --> 00:04:52.240
Um it's uh it's a lot.
00:04:52.399 --> 00:05:06.399
Um so as we dive into this topic, you know, one of the things that I was thinking about is after we did our pre-interview is, you know, product marketing has never had a lack of opinions being thrown at them from all directions.
00:05:06.560 --> 00:05:11.439
And now that opinion is hopefully getting replaced by real data.
00:05:11.759 --> 00:05:22.240
So now they're they have not just opinions, but they also have a lot of data to support, you know, and sort through and to try to prioritize.
00:05:22.480 --> 00:05:38.240
So, you know, one of the things that I think we're gonna be talking about is how we're gonna be using what are good ways of using that data, knowing when to use the data, how to deal with those conflicts in different data, and you know, just bringing in the human element as well.
00:05:38.480 --> 00:05:48.319
So let's start with just the different sources of data that you and I are seeing and hearing out there, you know, in addition to Troop doing the messaging signal.
00:05:48.879 --> 00:06:05.439
Well, I mean, you have the all the different kinds of aspects on a star, if you will, of everything from sales, conversation, intelligence, traditional tools like gong and sales loft on what the buyers are saying and the questions that they're asking and objections raised.
00:06:05.519 --> 00:06:10.000
So that traditional tool, you know, we know a lot of people that use that technology.
00:06:10.160 --> 00:06:24.480
And then, of course, you have like the win-loss aspect that's helping you understand what why buyers chose you and what competitors are saying, and sometimes not doing anything at all, the classic problem of why they didn't choose a solution.
00:06:24.720 --> 00:06:27.839
So that win-loss aspect is a whole other echelon.
00:06:28.160 --> 00:06:40.959
And then, of course, kind of close to that, we've seen you know, technologies like competitive intelligence tools like a clue or a crane that continuously are tracking what's going on on a website and you're trying to bring that in.
00:06:41.120 --> 00:06:48.560
But then you're also trying to bring in your battle cards of what you've honed in on and try to make sense of that together.
00:06:48.720 --> 00:07:02.639
Um, you've also got tools, of course, like you know, like you know, traditional research platforms trying to look at approaches, tools like Pendo, obviously, for showing what customers are actually doing once they use your product.
00:07:02.800 --> 00:07:08.800
Um, you know, and then of course, you have what we're trying to do within product marketing with what you're doing here at Troops.
00:07:08.879 --> 00:07:24.160
So you got all these different aspects of how really what the product marketing role is, and around that are all those people in the go-to-market, I guess, grouping that have the technologies they're trying to use and tell you what needs to happen.
00:07:24.319 --> 00:07:29.920
So that's where I think you're getting all these different signals because they might be telling you something different.
00:07:30.000 --> 00:07:38.160
And so then, how do you then, as a product marketer, try to figure out well, what should our position be when we're hearing these different types of signals?
00:07:38.319 --> 00:07:41.680
And do we change or do we trust one over the other?
00:07:41.839 --> 00:07:43.439
And I think that's what we're trying to get at.
00:07:43.680 --> 00:07:44.160
Absolutely.
00:07:44.319 --> 00:07:47.439
I I think I plus one on all of the examples that you gave.
00:07:47.519 --> 00:07:49.439
And in fact, I just got an email this morning.
00:07:49.519 --> 00:07:52.160
I subscribed to a newsletter called RevOps Co-op.
00:07:52.399 --> 00:07:57.839
Um, I can't even recall how I got on that newsletter list, but I enjoy their content.
00:07:58.000 --> 00:08:14.480
And they were talking about signals that are coming in for those running ABM strategies, account-based marketing strategies, and like enrichment signals around account level behavior and buying team and buying committee behavior and signals that they're detecting there.
00:08:14.720 --> 00:08:20.480
You've got social media signals, and yeah, there's a there's a lot of options out there, you know.
00:08:20.720 --> 00:08:27.600
So a lot of it too is is how do we prioritize it and how do we trust it?
00:08:27.920 --> 00:08:35.600
I I I would argue trust is maybe not as big of a concern for me.
00:08:35.840 --> 00:08:43.200
Um, but I'm definitely gonna let you talk about why you want to kind of raise a flag about trust.
00:08:43.519 --> 00:08:52.240
For me, a lot of it is about getting really statistically significant data that's also in the right context, right?
00:08:52.480 --> 00:08:55.759
So let's take wind loss as an example.
00:08:55.919 --> 00:09:25.200
I remember, you know, before there was uh tools like closed and others who did wind loss intelligence, doing just the traditional you email somebody, hey, would you talk to me about why you chose us, why you didn't choose us, and you'd have a conversation, and that conversation would often, you know, you'd you'd gather the information from that, um, make some notes, you'd rely on what that person decided to tell you, and it was also reliant on what you decided to ask them.
00:09:25.360 --> 00:09:29.919
And then you would lob that, and that would then become data that you would pass along.
00:09:30.000 --> 00:09:37.360
That now that didn't scale very well, and um, you also were at the mercy of how that conversation went.
00:09:37.600 --> 00:09:43.279
So that's an example of like, okay, maybe that plays into your thoughts around trust, Constantine.
00:09:43.519 --> 00:09:47.600
Do we trust that we got enough information and the right data out of that?
00:09:47.840 --> 00:09:50.480
Now we have tools that are doing this at scale.
00:09:50.559 --> 00:10:01.679
So I I think the good news is that because you have the tool uh and you have the scale, that that that tribal knowledge and maybe bias might be less.
00:10:01.840 --> 00:10:06.000
But again, it goes back to I think three things that I think that we should try to look at.
00:10:06.159 --> 00:10:08.320
One is the proximity.
00:10:08.559 --> 00:10:12.960
How close is that data to the behavior or decision that I'm trying to understand?
00:10:13.120 --> 00:10:13.279
Right.
00:10:13.600 --> 00:10:14.559
That's a great point.
00:10:14.879 --> 00:10:20.799
So understand, like when I'm looking at the data for customers, do we truly understand the customer behavior?
00:10:20.960 --> 00:10:22.240
Okay, so check that.
00:10:22.399 --> 00:10:31.759
If I'm looking at leads in a funnel and I'm going to a conference and I'm trying to understand why they might be downloading something, what's that behavior that's behind that?
00:10:32.000 --> 00:10:39.120
Or in a traditional late stage deal, what's the proximity of us trying to get deals closed?
00:10:39.279 --> 00:10:42.399
What's the behavior behind that and a decision I'm trying to understand?
00:10:42.559 --> 00:10:56.799
So still understanding the nature of our corporation, our company, our sales organization and a go-to-market organization, how we've worked, understanding that behavior is important to understanding then what that data is bringing us.
00:10:56.960 --> 00:10:58.639
And then to your point, the scale.
00:10:58.879 --> 00:11:11.120
So, you know, if I'm hearing three customers say this and those three customers make up a majority of our business, you know, that's gonna be very different than if I've got tons of customers and I'm trying to understand what that means.
00:11:11.279 --> 00:11:21.200
Is there a pattern that I'm seeing that other, you know, customers might be doing and they might be flexing or they might be adjusting to the way they want to use our technology?
00:11:21.360 --> 00:11:28.399
And then to your to your last point about, again, the trust in the context, do I understand what that what that signal is?
00:11:28.639 --> 00:11:31.440
Do I know the context under how I got that data?
00:11:31.600 --> 00:11:48.159
So back in the day when we may have had kind of more of that relying on a salesperson to give us their honest view of what's going on versus understanding, you know, how much more of that data I have to say, well, this is this is why we think this might be more true than not.
00:11:48.320 --> 00:11:56.399
And then so then trying to triangulate those three things, which is not easy, is I think the way then where the human judgment comes in.
00:11:56.639 --> 00:12:04.559
So, how do we try to bring those three elements together, which I said proximity of the data, the scale of that data, and the context of that data.
00:12:04.639 --> 00:12:06.159
So I love that framework.
00:12:06.399 --> 00:12:10.159
I'm I'm gonna definitely highlight that as we summarize this session.
00:12:10.240 --> 00:12:12.559
I think that's those are really important points.
00:12:12.720 --> 00:12:14.960
The the proximity one definitely jumps out.
00:12:15.279 --> 00:12:17.440
Actually, all three of those things jump out.
00:12:17.600 --> 00:12:24.559
Um, getting that signal in the moment and having it be timely, having it be relevant is really important.
00:12:25.120 --> 00:12:38.240
Um, you know, just going back to win-loss, it's not as useful to interview somebody three months later than you know, knowing what happened, you know, as shortly after it happened or as it happened.
00:12:38.480 --> 00:12:46.559
So talking about how we can navigate all of this using AI as a tool, um, it gives us a lot of optionality.
00:12:46.799 --> 00:12:54.799
There are lots of products out there that are AI native or AI, you know, moving to bringing AI into existing SaaS products.
00:12:55.039 --> 00:13:00.320
We can DIY AI agents to help us get some of this data and this signal.
00:13:00.720 --> 00:13:07.840
Uh but once that's all happening, then how do we prioritize where to focus, right?
00:13:08.080 --> 00:13:14.000
So, you know, if you have all these different options on the table, you laid them out very nicely at the beginning.
00:13:14.399 --> 00:13:16.080
Where do you start, right?
00:13:16.240 --> 00:13:18.559
Like, how do you know where to focus?
00:13:18.799 --> 00:13:21.840
How what would you recommend to folks, Constantine?
00:13:22.159 --> 00:13:31.919
I I think the AI question becomes one depending upon what the technologies are that you're using that may already have the AI embedded.
00:13:32.080 --> 00:13:37.279
It also goes back to how comfortable you know the organization is in using that.
00:13:37.440 --> 00:13:44.320
So I think AI can dramatically improve the scale and that triangulation that we were talking about.
00:13:44.559 --> 00:14:12.879
And the context judgment still remains that harder part because just like we all have had experience working in our own environments, if we're working with a ChatGPT or Claude or Gemini or whatever, we all kind of go through those motions of experimenting and playing with that information to see how does that, how do we best use those tools to get what we think we need rather than just getting overly inundated by what the AI is telling us.
00:14:13.039 --> 00:14:14.960
So definitely the scale's there.
00:14:15.039 --> 00:14:26.399
But then I think where the AI can help is help us identify themes or objections or patterns or language and changes, that that's where that scale comes in, right?
00:14:26.559 --> 00:14:38.799
Where when we were on our own and we look at two or three or four or five scenarios, or we went to a sales meeting or go to market call and we had a couple examples, well, the AI can really accelerate that.
00:14:38.960 --> 00:14:41.840
So I think that's one of the benefits of using it.
00:14:42.000 --> 00:14:46.159
And I also think that it can also help us find contradictions, right?
00:14:46.240 --> 00:14:59.919
So if you're looking at the AI and you're looking through, and depending upon how you're interacting with it, we may be able to use that tool to help us say, well, are there things that we could find here that are not consistent?
00:15:00.080 --> 00:15:03.919
So it's it goes back to that pattern finding where the scale comes into play.
00:15:04.080 --> 00:15:12.000
And then also, I think what's cool about using the AI is it can also help us with continuous intelligence that we just didn't have before.
00:15:12.240 --> 00:15:22.799
So, you know, if we're in a project process or in a quarterly basis and we're trying to do some kind of positioning exercise or we're trying to do some kind of win-loss study every six months.
00:15:23.039 --> 00:15:35.519
Nowadays, with some of these tools that are bringing all this information together and with help with the AI, we're almost kind of doing this more in a continuous manner that helps us uh, you know, be stay on top.
00:15:35.600 --> 00:15:44.399
And I don't know if I want to say real time, but it helps us stay on top of things more in a way that is not something that was just possible before.
00:15:44.559 --> 00:15:50.639
And then using that way to interact with the AI to say, let's just kind of take a point in time and say, well, what's going on?
00:15:50.799 --> 00:15:52.320
Is there something I should be seeing?
00:15:52.399 --> 00:15:54.000
Yeah, and start asking that question.
00:15:54.080 --> 00:16:00.399
So using the AI as a tool to as a feedback mechanism that you can keep going back and forth with.
00:16:00.559 --> 00:16:05.759
Um, and so that that's where I think we have the advantage now uh that we just didn't have before.
00:16:05.919 --> 00:16:13.600
You're actually able to interact with the the AI and not just you know do analysis on your own and assume you're going down the right path.
00:16:13.759 --> 00:16:16.720
So it's kind of doing a checkpoint for you in some ways.
00:16:17.039 --> 00:16:18.159
Yeah, definitely.
00:16:19.279 --> 00:16:19.840
For sure.
00:16:20.000 --> 00:16:22.559
I I I totally agree with all of that.
00:16:22.720 --> 00:16:33.519
I think that one of the beauties of it is that it can keep this sort of um, it's it's almost like a a watchtower.
00:16:33.679 --> 00:16:41.919
Like if if it's constantly ingesting data and looking for these patterns, it can be this eye in the sky.
00:16:42.080 --> 00:16:44.080
I'm sorry, that's a terrible reference.
00:16:44.240 --> 00:16:46.480
But it's little Alan Parsons there for us.
00:16:47.679 --> 00:16:58.000
Yeah, it can alert you when there's something you do need to pay attention to, which is great, but it also can combine data sets in an interesting way.
00:16:58.159 --> 00:17:08.240
Uh, an example that I thought of um was, you know, let's say and the human element is here too, but here's an example.
00:17:08.400 --> 00:17:15.759
Let's say we have identified that we have maybe an issue with customer retention and maybe also expansion.
00:17:16.160 --> 00:17:18.799
Um, all right, so we've identified this as an issue.
00:17:18.960 --> 00:17:25.920
If that is a business challenge currently, then that obviously dictates some prioritization, right?
00:17:26.000 --> 00:17:47.599
And product marketing in combination with some of the other members of the team, customer success, sales, even product, product services, whoever may be involved with that challenge of customer retention can now say, okay, what data sets do we have available to us that can help us understand this better and figure out an action plan, right?
00:17:47.680 --> 00:17:56.079
So you can take data out of uh Pendo, Pendo Novis can tell you what's happening with product behavior specific to your customers.
00:17:56.559 --> 00:18:02.079
Um and it can even segment it and give you some real specific specificity around that.
00:18:02.319 --> 00:18:12.160
And you can take that data and perhaps combine it with truth, which is maybe looking at what type of messages and talk tracks are we using with our customers.
00:18:12.400 --> 00:18:14.559
How are they, how are we messaging to them?
00:18:14.640 --> 00:18:18.000
What happened leading up to before they became a customer?
00:18:18.240 --> 00:18:21.440
Meaning what was their expectation from a messaging standpoint?
00:18:21.599 --> 00:18:25.839
And maybe also how are we messaging to them after they've become a customer?
00:18:26.720 --> 00:18:30.079
How does that line up with how they're using the product?
00:18:30.319 --> 00:18:37.839
Like and ask AI to do some pattern analysis on those two data sets and say, what are we detecting?
00:18:38.000 --> 00:18:39.519
Are we seeing something there?
00:18:39.680 --> 00:18:43.920
I think that's an example of like how you can bring some of the signal together.
00:18:45.039 --> 00:18:55.359
I mean, even taking that step further, maybe seeing the types of questions that the customers are asking as that product has rolled out and understanding that uh element.
00:18:55.599 --> 00:19:05.519
And is that something that's also now coming up more frequently in the questions that are being asked about what our product can do in trying to close a deal?
00:19:05.680 --> 00:19:18.720
So those, again, that relationship between customer and prospect and those in the in the sales cycle versus those that are just kind of, you know, we're going out and trying to go to a conference and understand what we should be seeing.
00:19:18.799 --> 00:19:33.359
All those, I think those are all tying and it's much easier to do that because you have this information that you're trying to still experiment with and play with to see if you can detect those patterns, but you know, test assumptions much more quickly, right?
00:19:33.839 --> 00:19:43.759
Um, and I think part of that too is, you know, dare I say, this kind of Socratic method that you sometimes play with with the, you know, with the AI, right?
00:19:43.839 --> 00:19:45.200
So you kind of throw out an idea.
00:19:45.440 --> 00:19:50.000
You say, I'm gonna make an idea, I have a claim here, I think that this is what's going on.
00:19:50.160 --> 00:19:59.759
We could take a look and start probing into that more with the data that we have, and then try to get some clarity around that to say, is that, is that, is there some truth to what we're saying?
00:20:00.000 --> 00:20:03.119
And being able to use the AI in and more in that nature.
00:20:03.200 --> 00:20:33.279
So that's why I'm saying being able to interact with the with a platform in that much more in a way that's it's in a way that it responds more quickly in a human way than it was just plainly looking at data and looking at the numbers, which might have been a little bit more kind of esoteric to say, well, I see this pattern, let's take an action on that, but yeah, interacting with the technology to be able to maybe get insights that you you just weren't able to before uh because you didn't have that, the AI being able to help you that way.
00:20:33.680 --> 00:20:34.319
Agreed.
00:20:34.480 --> 00:20:35.599
Yeah, for sure.
00:20:35.759 --> 00:20:40.640
I thought you were gonna go somewhere else with the Socratic method, but but I like where you went with it.
00:20:40.720 --> 00:20:49.680
But when you when you threw that out there, I thought about how different people across different teams might be bringing their data together into a meeting.
00:20:49.759 --> 00:20:53.119
And and I think there's a Socratic method there too, right?
00:20:53.200 --> 00:20:59.680
Like if you have people with different data sets or different views on the same challenge, let's sort through that.
00:20:59.839 --> 00:21:00.640
Let's get through that.
00:21:00.799 --> 00:21:03.519
Let's figure out well, what data is more meaningful?
00:21:03.680 --> 00:21:05.359
What data do we care about?
00:21:05.519 --> 00:21:11.599
Um, so there's the Socratic method with EAI, but there's also the human element of it, too, right?
00:21:11.680 --> 00:21:14.319
We need to bring this all together as a team.
00:21:14.559 --> 00:21:29.920
Yeah, in the end, and probing into that and then kind of coming up with, you know, exposing what we think is true or what's not, and then getting some clarity there to say, yeah, let's all agree that this is the way we should, this is the approach we should take because now we we all have some an agreement on on what the approach should be.
00:21:30.000 --> 00:21:38.640
So yeah, so yeah, it definitely and more of a human side of it with the data that we're we are collecting more uh as well as with the yeah, absolutely.
00:21:38.960 --> 00:21:57.359
Yeah, and I think you know, your your trio uh framework um should also is a good one to help be a vetting process for other data sets within the organization as you're bringing those together and determining which signals do we trust more than others to help us make a decision.
00:21:57.519 --> 00:22:10.880
The humans at the end of the day, hopefully you're Still making a lot of the strategic decisions, you know, maybe not the tactical automated decisions, but we're still directionally making the important big rock decisions, right?
00:22:11.200 --> 00:22:12.319
Yeah, absolutely.
00:22:12.480 --> 00:22:16.880
Yeah, the proximity, the scale, the context, you know, that that those three things.
00:22:17.039 --> 00:22:21.039
I I think again, uh, I like to use that as as a method to help.
00:22:21.119 --> 00:22:24.400
Again, so it just gives you a better perspective on an approach.
00:22:24.559 --> 00:22:31.759
It's a simpler approach, but it it also gives you a way to just kind of like hone in on how am I gonna manage this?
00:22:31.920 --> 00:22:35.440
Because it sometimes becomes it might become too much, right?
00:22:35.519 --> 00:22:41.440
And you're saying, well, how do I make a decision uh because I have different, I have those different signals that are coming up.
00:22:41.680 --> 00:22:44.160
Yeah, so definitely that's a great recommendation.
00:22:44.319 --> 00:22:46.960
Getting into other recommendations that that we would have.
00:22:47.119 --> 00:23:02.799
I certainly would say if it gets if it's getting really noisy, think about where the pain and challenge is for the business right now and and kind of because you do ultimately have to prioritize, right?
00:23:02.880 --> 00:23:07.200
You can't be addressing every signal all the time.
00:23:07.359 --> 00:23:09.680
You know, that's a bit difficult, right?
00:23:09.839 --> 00:23:16.720
So certain things are gonna raise to be more of a project priority at the moment or more of an initiative at the moment.
00:23:16.960 --> 00:23:20.960
And then once you address that, you can kind of let it ride, as you as you mentioned.
00:23:21.039 --> 00:23:25.200
The the AI can then just be the detection mechanism on what's happening.
00:23:25.519 --> 00:23:30.000
But I would say, you know, make sure you're anchored in what's a priority for the business.
00:23:30.240 --> 00:23:36.880
And that requires domain knowledge and domain expertise that the human product marketers bring.
00:23:36.960 --> 00:23:41.119
And I know you and I talked about that in a previous podcast episode as well.
00:23:41.680 --> 00:23:42.559
Absolutely.
00:23:42.960 --> 00:23:56.000
I also think that this is where the familiarity with the platform that you're using becomes really critical because then you could you could use those tools, those AI elements that might be in the tool to help identify what's going on.
00:23:56.160 --> 00:24:03.759
So, you know, we hear a lot about agents, and I know Jen, you mentioned me speaking at a lot of these conferences, we're talking about agents in procurement or supply chain.
00:24:03.920 --> 00:24:25.519
But when you're using some of the agents and the platform and being able to interact, that allows you then to see, let's let's test some of these things that we're trying to do, where you have an agent doing some recurring process to collect the information in the way that you might be looking for, having something pop up in your inbox, the way you want it to be researched, or using some of those elements to your advantage.
00:24:25.599 --> 00:24:33.680
And I think those things, again, just automating some of those processes so that you can then get that clearer picture is something we should just always be doing.
00:24:33.839 --> 00:24:37.440
So, you know, how do I continuously monitor my competitors?
00:24:37.599 --> 00:24:50.319
How do I continuously monitor monitor what our customers are saying and then bringing that together and and then being able to show some pattern that we're trying to recognize with our with our team to say, yeah, this is I think the direction we should be going.
00:24:50.880 --> 00:24:56.559
Yeah, and I I uh, you know, said differently on some of this stuff.
00:24:56.640 --> 00:25:00.559
I I would say I get really excited.
00:25:00.640 --> 00:25:06.880
I'm a data junkie, so I get excited about a lot of this stuff, but sometimes data can just be for data's sake.
00:25:07.119 --> 00:25:11.680
And you do still have to sort through the data that is more meaningful.
00:25:11.839 --> 00:25:19.519
I see a lot of um marketing language out there around we turn data into action or we turn insights into action.
00:25:19.920 --> 00:25:26.640
And yes, that's true, that's important, but somebody also you know some of that can be automated, that action can be automated.
00:25:27.039 --> 00:25:36.559
Um, but product marketers, I think, should in particular should really embrace having a lot of data knowledge that they now that they previously didn't have access to.
00:25:36.720 --> 00:25:42.160
And it's sort of their job to sift through what matters and what is important.
00:25:42.319 --> 00:25:50.400
There, there I think product marketing can have a real moment here to use this to become even more strategic to the business.
00:25:50.640 --> 00:25:50.880
Yeah.
00:25:51.279 --> 00:25:52.400
Agreed or disagree?
00:25:52.640 --> 00:25:54.240
No, I I agree, I agree.
00:25:54.319 --> 00:26:03.839
And I think it goes back to something you and I have spoken about before, where is the challenge of the product marketer and trying to let's just say herd the cat sometimes, right?
00:26:04.000 --> 00:26:11.839
Where they're bringing in a lot of these elements and showing some proof where they have to have a win to show this is why we're doing it this way.
00:26:11.920 --> 00:26:21.039
And there's a proof point that they then get buy-in with the other teams, because again, they might be relying on the tools that they have disposal, you know, at their disposal.
00:26:21.200 --> 00:26:37.039
So if you can show some of that insight that you're being able to bring with the tools that they have, or even the tools that you're using to show, yeah, I'm I'm gonna tell you that this is the direction I think we should go because of all these things that I'm hearing from all the different parts of the go-to-market team.
00:26:37.119 --> 00:26:59.359
So it helps put product marketing on a bigger pedestal because then they understand how to use the technology to show to the rest of the team what they should be doing or what they should be backing off of, particularly in their focus of coming up with the proper positioning, whether it's for the product or you know, it's or our product line or even just, you know, just directionally of where the where the company should be going.
00:26:59.680 --> 00:27:00.640
Yeah, definitely.
00:27:00.799 --> 00:27:02.160
I I really love this.
00:27:02.319 --> 00:27:17.599
Um I the the conversations, the this topic has, as I think back on a lot of the episodes we've had on this podcast show, has really kind of weaved through a lot of those conversations without directly addressing this like you and I have today.
00:27:17.839 --> 00:27:36.880
We just a few episodes ago, I was talking with Kevin Chan, who's a head of product marketing um for uh locust, who was formerly with Fledio, and he was talking about you know competitive uh intelligence as a signal and to how not to overreact to every competitive signal that you get.
00:27:37.039 --> 00:27:50.960
And so, and it kind of goes back to your thesis for this episode, Constantine, which is that not every signal is worth addressing, not every signal is necessarily trustworthy, and not every signal is a priority.
00:27:51.279 --> 00:27:58.319
Um so, you know, this is this is a moment for all of us to figure out how to make sense of all of this data signal.
00:27:58.799 --> 00:28:00.160
Yeah, yeah, absolutely.
00:28:00.319 --> 00:28:24.240
Yeah, and and I think in the end, again, using technology as a tool, using using these as elements, never forgetting about the human element and understanding, like we were saying before, of having that intuition of trying to think of what we should be doing, and then using that technology to your advantage, where again, having all the the data is net is overflowing.
00:28:24.400 --> 00:28:28.400
Let's just make sure you're looking at the right data and not trying to boil the ocean with it, right?
00:28:28.559 --> 00:28:28.799
Yeah.
00:28:29.359 --> 00:28:30.480
Yeah, exactly.
00:28:30.640 --> 00:28:37.440
Well, how to here how to wrap up, um, Constantine, uh, how does this relate to the work that you're doing with clients?
00:28:37.599 --> 00:28:46.640
Certainly you're you're um working with clients who are dealing with a lot of data sets, new data sets that they've not had before or at a scale that's higher.
00:28:46.799 --> 00:28:48.799
How does this relate to your work with them?
00:28:49.200 --> 00:28:56.880
Well, uh, you know, we we know that our our clients are trying to use a lot of the tools that are out there uh to try to make sense of things.
00:28:56.960 --> 00:29:07.440
And it goes back to knowing what kind of data are we are we looking at, what do they want to hone in on, what is ultimately their strategy and how they should be tuning into that.
00:29:07.599 --> 00:29:15.680
So we help guide them with just knowing which which technologies they're using, whether you're using something homegrown or they're trying to decide on what to buy.
00:29:15.839 --> 00:29:39.920
But at the same time, uh going back to knowing who they are as a company, what they're trying to say, what they're trying to do in the market, and using that knowledge of the market itself, of that history, experience, or even the insight of why was this company founded, what directionally is changing, and helping them uh put that together.
00:29:40.079 --> 00:29:55.279
So leveraging the data that we get from that technology is something helpful, but it helps us then when we're doing some of these workshops and talking to them about what we think is happening based on all this information that they're giving us and what we anticipate what they should be doing differently.
00:29:55.599 --> 00:29:56.160
Wonderful.
00:29:56.319 --> 00:29:56.640
Okay.
00:29:56.720 --> 00:29:58.000
Well, thank you, Constantine.
00:29:58.079 --> 00:29:59.279
I always love talking with you.
00:29:59.359 --> 00:30:03.839
Thanks for a great topic and uh good luck to everybody out there with all your signals.
00:30:04.240 --> 00:30:06.880
Thank you, Jen, and thank you, everybody, for joining us.
00:00:17.600 --> 00:00:21.679
Hi everybody, welcome to this episode of Said It Differently.
00:00:21.760 --> 00:00:31.920
It's the podcast show brought to you by Troop, and it's focused on B2B messaging and storytelling and how you're bringing all of that to market and what you do with it.
00:00:32.159 --> 00:00:42.880
Today's topic is an interesting one for me because I'm seeing and hearing a lot in the market right now around how to bring in more market signal.
00:00:43.200 --> 00:00:51.600
And that is bringing even more data and intelligence to product marketing, who maybe was doing a lot of that DIY in the past.
00:00:51.759 --> 00:00:58.079
So we're going to be diving into like what do we do with all of these different sources of market signal?
00:00:58.240 --> 00:01:04.560
What are the different types of data sets that are out there now that we maybe didn't have at scale before?
00:01:04.719 --> 00:01:13.120
And joining me today for this conversation is longtime friend and colleague, um Constantine Limbarakis, who is with Liberist Consulting.
00:01:13.280 --> 00:01:19.200
He actually brought this topic to me because it was burning a hole in his brain, maybe Constantine.
00:01:19.680 --> 00:01:27.840
But he is an expert product marketer that I've met through the procurement and supply chain space, uh, who really has domain knowledge there.
00:01:28.000 --> 00:01:31.760
But Constantine, please introduce yourself before we dive into the topic.
00:01:32.159 --> 00:01:34.319
Thank you, Jennifer, for having me here again.
00:01:34.480 --> 00:01:36.799
Uh, again, my name is Constantine Limbarakis.
00:01:36.879 --> 00:01:41.760
I am the founder of a consulting firm called Liberus Consulting.
00:01:41.840 --> 00:02:11.280
And what we do is take a lot of our experiences that we've done as practitioners back in the past, as product marketers and as experts in the, as Jen said, procurement supply chain space, and apply that expertise as we've seen the evolution of these areas and these industries, and apply that to the technology providers that require insight to say, well, how do we go to market and how do we best put our foot forward for uh for focusing on what's what what changes lie ahead?
00:02:11.520 --> 00:02:12.240
Yeah, great.
00:02:12.400 --> 00:02:16.639
And I know you also get out there into the market a lot, Constantine.
00:02:16.719 --> 00:02:22.879
You're hosting and moderating a lot of conferences and events in your space between that and your client work.
00:02:23.039 --> 00:02:31.039
So you're certainly seeing and hearing a lot in the market around more tools and more mechanisms for bringing in more market signal, right?
00:02:31.199 --> 00:02:33.680
Like, what was the inspiration for you on this topic?
00:02:33.759 --> 00:02:38.639
Or what was like triggering for you that you felt like, yeah, we need to tackle this?
00:02:38.960 --> 00:02:47.520
Well, you know, one of the things that I just been seeing is just the number of technologies that are out there now that help everybody in B2B, right?
00:02:47.599 --> 00:02:54.960
Whether it's in a particular function or even in the go-to-market space and product marketing and sales automation and marketing automation.
00:02:55.120 --> 00:03:02.319
And, you know, interestingly enough, you know, uh analyst firms like the Gartners of the World are now really trying to hone in on that.
00:03:02.479 --> 00:03:07.759
And I listened to a few of the webinars that they had and they're talking about hype cycles and all this.
00:03:07.919 --> 00:03:16.159
And I thought what's really interesting, and they're saying there's so many of these dynamics of them talking about these different solutions that are out there.
00:03:16.319 --> 00:03:31.680
I can only imagine then when you're a practitioner, you're trying to figure out all these different solutions that are competing with each other, might where you might have sales have one set of solutions, marketing having another set, and then there's this crossover, and you're saying, well, what am I relying on?
00:03:31.840 --> 00:03:38.719
And then so you have, you know, the signal confusion that the product marketer is always trying to orchestrate, right?
00:03:38.879 --> 00:03:46.479
So this just adds to it when you have a lot of different technologies that may or be maybe working together, but maybe not.
00:03:46.639 --> 00:03:49.360
And that's that's kind of why it's a good conversation to have.
00:03:49.680 --> 00:03:50.400
Very timely.
00:03:50.479 --> 00:03:50.879
You're right.
00:03:50.960 --> 00:04:01.439
Yeah, Gartner did just recently unveil its first hype cycle for product marketing, just like late summer 2026 here for those that might be watching this later.
00:04:01.599 --> 00:04:09.360
Um, so yeah, there's a lot coming in, and I would say that a lot of that signal, the noise, um, is is getting noisier.
00:04:09.759 --> 00:04:23.439
I was just at a recent um CMO and PMM summit in San Francisco and just looking around the exhibit hall floor and seeing all of the different vendors that are doing using AI to collect signal.
00:04:23.600 --> 00:04:25.279
Uh, Troop is one of them.
00:04:25.600 --> 00:04:26.639
I will be fast.
00:04:27.199 --> 00:04:32.879
Um and we're focused on messaging and your story and how that signal is playing into things.
00:04:33.120 --> 00:04:49.040
Um, but there are a lot, and then there were a lot of sessions about people, you know, from different presenters and speakers talking about how they're tackling the data and and how they're actually using AI, DIY to get data that they want.
00:04:49.360 --> 00:04:52.240
Um it's uh it's a lot.
00:04:52.399 --> 00:05:06.399
Um so as we dive into this topic, you know, one of the things that I was thinking about is after we did our pre-interview is, you know, product marketing has never had a lack of opinions being thrown at them from all directions.
00:05:06.560 --> 00:05:11.439
And now that opinion is hopefully getting replaced by real data.
00:05:11.759 --> 00:05:22.240
So now they're they have not just opinions, but they also have a lot of data to support, you know, and sort through and to try to prioritize.
00:05:22.480 --> 00:05:38.240
So, you know, one of the things that I think we're gonna be talking about is how we're gonna be using what are good ways of using that data, knowing when to use the data, how to deal with those conflicts in different data, and you know, just bringing in the human element as well.
00:05:38.480 --> 00:05:48.319
So let's start with just the different sources of data that you and I are seeing and hearing out there, you know, in addition to Troop doing the messaging signal.
00:05:48.879 --> 00:06:05.439
Well, I mean, you have the all the different kinds of aspects on a star, if you will, of everything from sales, conversation, intelligence, traditional tools like gong and sales loft on what the buyers are saying and the questions that they're asking and objections raised.
00:06:05.519 --> 00:06:10.000
So that traditional tool, you know, we know a lot of people that use that technology.
00:06:10.160 --> 00:06:24.480
And then, of course, you have like the win-loss aspect that's helping you understand what why buyers chose you and what competitors are saying, and sometimes not doing anything at all, the classic problem of why they didn't choose a solution.
00:06:24.720 --> 00:06:27.839
So that win-loss aspect is a whole other echelon.
00:06:28.160 --> 00:06:40.959
And then, of course, kind of close to that, we've seen you know, technologies like competitive intelligence tools like a clue or a crane that continuously are tracking what's going on on a website and you're trying to bring that in.
00:06:41.120 --> 00:06:48.560
But then you're also trying to bring in your battle cards of what you've honed in on and try to make sense of that together.
00:06:48.720 --> 00:07:02.639
Um, you've also got tools, of course, like you know, like you know, traditional research platforms trying to look at approaches, tools like Pendo, obviously, for showing what customers are actually doing once they use your product.
00:07:02.800 --> 00:07:08.800
Um, you know, and then of course, you have what we're trying to do within product marketing with what you're doing here at Troops.
00:07:08.879 --> 00:07:24.160
So you got all these different aspects of how really what the product marketing role is, and around that are all those people in the go-to-market, I guess, grouping that have the technologies they're trying to use and tell you what needs to happen.
00:07:24.319 --> 00:07:29.920
So that's where I think you're getting all these different signals because they might be telling you something different.
00:07:30.000 --> 00:07:38.160
And so then, how do you then, as a product marketer, try to figure out well, what should our position be when we're hearing these different types of signals?
00:07:38.319 --> 00:07:41.680
And do we change or do we trust one over the other?
00:07:41.839 --> 00:07:43.439
And I think that's what we're trying to get at.
00:07:43.680 --> 00:07:44.160
Absolutely.
00:07:44.319 --> 00:07:47.439
I I think I plus one on all of the examples that you gave.
00:07:47.519 --> 00:07:49.439
And in fact, I just got an email this morning.
00:07:49.519 --> 00:07:52.160
I subscribed to a newsletter called RevOps Co-op.
00:07:52.399 --> 00:07:57.839
Um, I can't even recall how I got on that newsletter list, but I enjoy their content.
00:07:58.000 --> 00:08:14.480
And they were talking about signals that are coming in for those running ABM strategies, account-based marketing strategies, and like enrichment signals around account level behavior and buying team and buying committee behavior and signals that they're detecting there.
00:08:14.720 --> 00:08:20.480
You've got social media signals, and yeah, there's a there's a lot of options out there, you know.
00:08:20.720 --> 00:08:27.600
So a lot of it too is is how do we prioritize it and how do we trust it?
00:08:27.920 --> 00:08:35.600
I I I would argue trust is maybe not as big of a concern for me.
00:08:35.840 --> 00:08:43.200
Um, but I'm definitely gonna let you talk about why you want to kind of raise a flag about trust.
00:08:43.519 --> 00:08:52.240
For me, a lot of it is about getting really statistically significant data that's also in the right context, right?
00:08:52.480 --> 00:08:55.759
So let's take wind loss as an example.
00:08:55.919 --> 00:09:25.200
I remember, you know, before there was uh tools like closed and others who did wind loss intelligence, doing just the traditional you email somebody, hey, would you talk to me about why you chose us, why you didn't choose us, and you'd have a conversation, and that conversation would often, you know, you'd you'd gather the information from that, um, make some notes, you'd rely on what that person decided to tell you, and it was also reliant on what you decided to ask them.
00:09:25.360 --> 00:09:29.919
And then you would lob that, and that would then become data that you would pass along.
00:09:30.000 --> 00:09:37.360
That now that didn't scale very well, and um, you also were at the mercy of how that conversation went.
00:09:37.600 --> 00:09:43.279
So that's an example of like, okay, maybe that plays into your thoughts around trust, Constantine.
00:09:43.519 --> 00:09:47.600
Do we trust that we got enough information and the right data out of that?
00:09:47.840 --> 00:09:50.480
Now we have tools that are doing this at scale.
00:09:50.559 --> 00:10:01.679
So I I think the good news is that because you have the tool uh and you have the scale, that that that tribal knowledge and maybe bias might be less.
00:10:01.840 --> 00:10:06.000
But again, it goes back to I think three things that I think that we should try to look at.
00:10:06.159 --> 00:10:08.320
One is the proximity.
00:10:08.559 --> 00:10:12.960
How close is that data to the behavior or decision that I'm trying to understand?
00:10:13.120 --> 00:10:13.279
Right.
00:10:13.600 --> 00:10:14.559
That's a great point.
00:10:14.879 --> 00:10:20.799
So understand, like when I'm looking at the data for customers, do we truly understand the customer behavior?
00:10:20.960 --> 00:10:22.240
Okay, so check that.
00:10:22.399 --> 00:10:31.759
If I'm looking at leads in a funnel and I'm going to a conference and I'm trying to understand why they might be downloading something, what's that behavior that's behind that?
00:10:32.000 --> 00:10:39.120
Or in a traditional late stage deal, what's the proximity of us trying to get deals closed?
00:10:39.279 --> 00:10:42.399
What's the behavior behind that and a decision I'm trying to understand?
00:10:42.559 --> 00:10:56.799
So still understanding the nature of our corporation, our company, our sales organization and a go-to-market organization, how we've worked, understanding that behavior is important to understanding then what that data is bringing us.
00:10:56.960 --> 00:10:58.639
And then to your point, the scale.
00:10:58.879 --> 00:11:11.120
So, you know, if I'm hearing three customers say this and those three customers make up a majority of our business, you know, that's gonna be very different than if I've got tons of customers and I'm trying to understand what that means.
00:11:11.279 --> 00:11:21.200
Is there a pattern that I'm seeing that other, you know, customers might be doing and they might be flexing or they might be adjusting to the way they want to use our technology?
00:11:21.360 --> 00:11:28.399
And then to your to your last point about, again, the trust in the context, do I understand what that what that signal is?
00:11:28.639 --> 00:11:31.440
Do I know the context under how I got that data?
00:11:31.600 --> 00:11:48.159
So back in the day when we may have had kind of more of that relying on a salesperson to give us their honest view of what's going on versus understanding, you know, how much more of that data I have to say, well, this is this is why we think this might be more true than not.
00:11:48.320 --> 00:11:56.399
And then so then trying to triangulate those three things, which is not easy, is I think the way then where the human judgment comes in.
00:11:56.639 --> 00:12:04.559
So, how do we try to bring those three elements together, which I said proximity of the data, the scale of that data, and the context of that data.
00:12:04.639 --> 00:12:06.159
So I love that framework.
00:12:06.399 --> 00:12:10.159
I'm I'm gonna definitely highlight that as we summarize this session.
00:12:10.240 --> 00:12:12.559
I think that's those are really important points.
00:12:12.720 --> 00:12:14.960
The the proximity one definitely jumps out.
00:12:15.279 --> 00:12:17.440
Actually, all three of those things jump out.
00:12:17.600 --> 00:12:24.559
Um, getting that signal in the moment and having it be timely, having it be relevant is really important.
00:12:25.120 --> 00:12:38.240
Um, you know, just going back to win-loss, it's not as useful to interview somebody three months later than you know, knowing what happened, you know, as shortly after it happened or as it happened.
00:12:38.480 --> 00:12:46.559
So talking about how we can navigate all of this using AI as a tool, um, it gives us a lot of optionality.
00:12:46.799 --> 00:12:54.799
There are lots of products out there that are AI native or AI, you know, moving to bringing AI into existing SaaS products.
00:12:55.039 --> 00:13:00.320
We can DIY AI agents to help us get some of this data and this signal.
00:13:00.720 --> 00:13:07.840
Uh but once that's all happening, then how do we prioritize where to focus, right?
00:13:08.080 --> 00:13:14.000
So, you know, if you have all these different options on the table, you laid them out very nicely at the beginning.
00:13:14.399 --> 00:13:16.080
Where do you start, right?
00:13:16.240 --> 00:13:18.559
Like, how do you know where to focus?
00:13:18.799 --> 00:13:21.840
How what would you recommend to folks, Constantine?
00:13:22.159 --> 00:13:31.919
I I think the AI question becomes one depending upon what the technologies are that you're using that may already have the AI embedded.
00:13:32.080 --> 00:13:37.279
It also goes back to how comfortable you know the organization is in using that.
00:13:37.440 --> 00:13:44.320
So I think AI can dramatically improve the scale and that triangulation that we were talking about.
00:13:44.559 --> 00:14:12.879
And the context judgment still remains that harder part because just like we all have had experience working in our own environments, if we're working with a ChatGPT or Claude or Gemini or whatever, we all kind of go through those motions of experimenting and playing with that information to see how does that, how do we best use those tools to get what we think we need rather than just getting overly inundated by what the AI is telling us.
00:14:13.039 --> 00:14:14.960
So definitely the scale's there.
00:14:15.039 --> 00:14:26.399
But then I think where the AI can help is help us identify themes or objections or patterns or language and changes, that that's where that scale comes in, right?
00:14:26.559 --> 00:14:38.799
Where when we were on our own and we look at two or three or four or five scenarios, or we went to a sales meeting or go to market call and we had a couple examples, well, the AI can really accelerate that.
00:14:38.960 --> 00:14:41.840
So I think that's one of the benefits of using it.
00:14:42.000 --> 00:14:46.159
And I also think that it can also help us find contradictions, right?
00:14:46.240 --> 00:14:59.919
So if you're looking at the AI and you're looking through, and depending upon how you're interacting with it, we may be able to use that tool to help us say, well, are there things that we could find here that are not consistent?
00:15:00.080 --> 00:15:03.919
So it's it goes back to that pattern finding where the scale comes into play.
00:15:04.080 --> 00:15:12.000
And then also, I think what's cool about using the AI is it can also help us with continuous intelligence that we just didn't have before.
00:15:12.240 --> 00:15:22.799
So, you know, if we're in a project process or in a quarterly basis and we're trying to do some kind of positioning exercise or we're trying to do some kind of win-loss study every six months.
00:15:23.039 --> 00:15:35.519
Nowadays, with some of these tools that are bringing all this information together and with help with the AI, we're almost kind of doing this more in a continuous manner that helps us uh, you know, be stay on top.
00:15:35.600 --> 00:15:44.399
And I don't know if I want to say real time, but it helps us stay on top of things more in a way that is not something that was just possible before.
00:15:44.559 --> 00:15:50.639
And then using that way to interact with the AI to say, let's just kind of take a point in time and say, well, what's going on?
00:15:50.799 --> 00:15:52.320
Is there something I should be seeing?
00:15:52.399 --> 00:15:54.000
Yeah, and start asking that question.
00:15:54.080 --> 00:16:00.399
So using the AI as a tool to as a feedback mechanism that you can keep going back and forth with.
00:16:00.559 --> 00:16:05.759
Um, and so that that's where I think we have the advantage now uh that we just didn't have before.
00:16:05.919 --> 00:16:13.600
You're actually able to interact with the the AI and not just you know do analysis on your own and assume you're going down the right path.
00:16:13.759 --> 00:16:16.720
So it's kind of doing a checkpoint for you in some ways.
00:16:17.039 --> 00:16:18.159
Yeah, definitely.
00:16:19.279 --> 00:16:19.840
For sure.
00:16:20.000 --> 00:16:22.559
I I I totally agree with all of that.
00:16:22.720 --> 00:16:33.519
I think that one of the beauties of it is that it can keep this sort of um, it's it's almost like a a watchtower.
00:16:33.679 --> 00:16:41.919
Like if if it's constantly ingesting data and looking for these patterns, it can be this eye in the sky.
00:16:42.080 --> 00:16:44.080
I'm sorry, that's a terrible reference.
00:16:44.240 --> 00:16:46.480
But it's little Alan Parsons there for us.
00:16:47.679 --> 00:16:58.000
Yeah, it can alert you when there's something you do need to pay attention to, which is great, but it also can combine data sets in an interesting way.
00:16:58.159 --> 00:17:08.240
Uh, an example that I thought of um was, you know, let's say and the human element is here too, but here's an example.
00:17:08.400 --> 00:17:15.759
Let's say we have identified that we have maybe an issue with customer retention and maybe also expansion.
00:17:16.160 --> 00:17:18.799
Um, all right, so we've identified this as an issue.
00:17:18.960 --> 00:17:25.920
If that is a business challenge currently, then that obviously dictates some prioritization, right?
00:17:26.000 --> 00:17:47.599
And product marketing in combination with some of the other members of the team, customer success, sales, even product, product services, whoever may be involved with that challenge of customer retention can now say, okay, what data sets do we have available to us that can help us understand this better and figure out an action plan, right?
00:17:47.680 --> 00:17:56.079
So you can take data out of uh Pendo, Pendo Novis can tell you what's happening with product behavior specific to your customers.
00:17:56.559 --> 00:18:02.079
Um and it can even segment it and give you some real specific specificity around that.
00:18:02.319 --> 00:18:12.160
And you can take that data and perhaps combine it with truth, which is maybe looking at what type of messages and talk tracks are we using with our customers.
00:18:12.400 --> 00:18:14.559
How are they, how are we messaging to them?
00:18:14.640 --> 00:18:18.000
What happened leading up to before they became a customer?
00:18:18.240 --> 00:18:21.440
Meaning what was their expectation from a messaging standpoint?
00:18:21.599 --> 00:18:25.839
And maybe also how are we messaging to them after they've become a customer?
00:18:26.720 --> 00:18:30.079
How does that line up with how they're using the product?
00:18:30.319 --> 00:18:37.839
Like and ask AI to do some pattern analysis on those two data sets and say, what are we detecting?
00:18:38.000 --> 00:18:39.519
Are we seeing something there?
00:18:39.680 --> 00:18:43.920
I think that's an example of like how you can bring some of the signal together.
00:18:45.039 --> 00:18:55.359
I mean, even taking that step further, maybe seeing the types of questions that the customers are asking as that product has rolled out and understanding that uh element.
00:18:55.599 --> 00:19:05.519
And is that something that's also now coming up more frequently in the questions that are being asked about what our product can do in trying to close a deal?
00:19:05.680 --> 00:19:18.720
So those, again, that relationship between customer and prospect and those in the in the sales cycle versus those that are just kind of, you know, we're going out and trying to go to a conference and understand what we should be seeing.
00:19:18.799 --> 00:19:33.359
All those, I think those are all tying and it's much easier to do that because you have this information that you're trying to still experiment with and play with to see if you can detect those patterns, but you know, test assumptions much more quickly, right?
00:19:33.839 --> 00:19:43.759
Um, and I think part of that too is, you know, dare I say, this kind of Socratic method that you sometimes play with with the, you know, with the AI, right?
00:19:43.839 --> 00:19:45.200
So you kind of throw out an idea.
00:19:45.440 --> 00:19:50.000
You say, I'm gonna make an idea, I have a claim here, I think that this is what's going on.
00:19:50.160 --> 00:19:59.759
We could take a look and start probing into that more with the data that we have, and then try to get some clarity around that to say, is that, is that, is there some truth to what we're saying?
00:20:00.000 --> 00:20:03.119
And being able to use the AI in and more in that nature.
00:20:03.200 --> 00:20:33.279
So that's why I'm saying being able to interact with the with a platform in that much more in a way that's it's in a way that it responds more quickly in a human way than it was just plainly looking at data and looking at the numbers, which might have been a little bit more kind of esoteric to say, well, I see this pattern, let's take an action on that, but yeah, interacting with the technology to be able to maybe get insights that you you just weren't able to before uh because you didn't have that, the AI being able to help you that way.
00:20:33.680 --> 00:20:34.319
Agreed.
00:20:34.480 --> 00:20:35.599
Yeah, for sure.
00:20:35.759 --> 00:20:40.640
I thought you were gonna go somewhere else with the Socratic method, but but I like where you went with it.
00:20:40.720 --> 00:20:49.680
But when you when you threw that out there, I thought about how different people across different teams might be bringing their data together into a meeting.
00:20:49.759 --> 00:20:53.119
And and I think there's a Socratic method there too, right?
00:20:53.200 --> 00:20:59.680
Like if you have people with different data sets or different views on the same challenge, let's sort through that.
00:20:59.839 --> 00:21:00.640
Let's get through that.
00:21:00.799 --> 00:21:03.519
Let's figure out well, what data is more meaningful?
00:21:03.680 --> 00:21:05.359
What data do we care about?
00:21:05.519 --> 00:21:11.599
Um, so there's the Socratic method with EAI, but there's also the human element of it, too, right?
00:21:11.680 --> 00:21:14.319
We need to bring this all together as a team.
00:21:14.559 --> 00:21:29.920
Yeah, in the end, and probing into that and then kind of coming up with, you know, exposing what we think is true or what's not, and then getting some clarity there to say, yeah, let's all agree that this is the way we should, this is the approach we should take because now we we all have some an agreement on on what the approach should be.
00:21:30.000 --> 00:21:38.640
So yeah, so yeah, it definitely and more of a human side of it with the data that we're we are collecting more uh as well as with the yeah, absolutely.
00:21:38.960 --> 00:21:57.359
Yeah, and I think you know, your your trio uh framework um should also is a good one to help be a vetting process for other data sets within the organization as you're bringing those together and determining which signals do we trust more than others to help us make a decision.
00:21:57.519 --> 00:22:10.880
The humans at the end of the day, hopefully you're Still making a lot of the strategic decisions, you know, maybe not the tactical automated decisions, but we're still directionally making the important big rock decisions, right?
00:22:11.200 --> 00:22:12.319
Yeah, absolutely.
00:22:12.480 --> 00:22:16.880
Yeah, the proximity, the scale, the context, you know, that that those three things.
00:22:17.039 --> 00:22:21.039
I I think again, uh, I like to use that as as a method to help.
00:22:21.119 --> 00:22:24.400
Again, so it just gives you a better perspective on an approach.
00:22:24.559 --> 00:22:31.759
It's a simpler approach, but it it also gives you a way to just kind of like hone in on how am I gonna manage this?
00:22:31.920 --> 00:22:35.440
Because it sometimes becomes it might become too much, right?
00:22:35.519 --> 00:22:41.440
And you're saying, well, how do I make a decision uh because I have different, I have those different signals that are coming up.
00:22:41.680 --> 00:22:44.160
Yeah, so definitely that's a great recommendation.
00:22:44.319 --> 00:22:46.960
Getting into other recommendations that that we would have.
00:22:47.119 --> 00:23:02.799
I certainly would say if it gets if it's getting really noisy, think about where the pain and challenge is for the business right now and and kind of because you do ultimately have to prioritize, right?
00:23:02.880 --> 00:23:07.200
You can't be addressing every signal all the time.
00:23:07.359 --> 00:23:09.680
You know, that's a bit difficult, right?
00:23:09.839 --> 00:23:16.720
So certain things are gonna raise to be more of a project priority at the moment or more of an initiative at the moment.
00:23:16.960 --> 00:23:20.960
And then once you address that, you can kind of let it ride, as you as you mentioned.
00:23:21.039 --> 00:23:25.200
The the AI can then just be the detection mechanism on what's happening.
00:23:25.519 --> 00:23:30.000
But I would say, you know, make sure you're anchored in what's a priority for the business.
00:23:30.240 --> 00:23:36.880
And that requires domain knowledge and domain expertise that the human product marketers bring.
00:23:36.960 --> 00:23:41.119
And I know you and I talked about that in a previous podcast episode as well.
00:23:41.680 --> 00:23:42.559
Absolutely.
00:23:42.960 --> 00:23:56.000
I also think that this is where the familiarity with the platform that you're using becomes really critical because then you could you could use those tools, those AI elements that might be in the tool to help identify what's going on.
00:23:56.160 --> 00:24:03.759
So, you know, we hear a lot about agents, and I know Jen, you mentioned me speaking at a lot of these conferences, we're talking about agents in procurement or supply chain.
00:24:03.920 --> 00:24:25.519
But when you're using some of the agents and the platform and being able to interact, that allows you then to see, let's let's test some of these things that we're trying to do, where you have an agent doing some recurring process to collect the information in the way that you might be looking for, having something pop up in your inbox, the way you want it to be researched, or using some of those elements to your advantage.
00:24:25.599 --> 00:24:33.680
And I think those things, again, just automating some of those processes so that you can then get that clearer picture is something we should just always be doing.
00:24:33.839 --> 00:24:37.440
So, you know, how do I continuously monitor my competitors?
00:24:37.599 --> 00:24:50.319
How do I continuously monitor monitor what our customers are saying and then bringing that together and and then being able to show some pattern that we're trying to recognize with our with our team to say, yeah, this is I think the direction we should be going.
00:24:50.880 --> 00:24:56.559
Yeah, and I I uh, you know, said differently on some of this stuff.
00:24:56.640 --> 00:25:00.559
I I would say I get really excited.
00:25:00.640 --> 00:25:06.880
I'm a data junkie, so I get excited about a lot of this stuff, but sometimes data can just be for data's sake.
00:25:07.119 --> 00:25:11.680
And you do still have to sort through the data that is more meaningful.
00:25:11.839 --> 00:25:19.519
I see a lot of um marketing language out there around we turn data into action or we turn insights into action.
00:25:19.920 --> 00:25:26.640
And yes, that's true, that's important, but somebody also you know some of that can be automated, that action can be automated.
00:25:27.039 --> 00:25:36.559
Um, but product marketers, I think, should in particular should really embrace having a lot of data knowledge that they now that they previously didn't have access to.
00:25:36.720 --> 00:25:42.160
And it's sort of their job to sift through what matters and what is important.
00:25:42.319 --> 00:25:50.400
There, there I think product marketing can have a real moment here to use this to become even more strategic to the business.
00:25:50.640 --> 00:25:50.880
Yeah.
00:25:51.279 --> 00:25:52.400
Agreed or disagree?
00:25:52.640 --> 00:25:54.240
No, I I agree, I agree.
00:25:54.319 --> 00:26:03.839
And I think it goes back to something you and I have spoken about before, where is the challenge of the product marketer and trying to let's just say herd the cat sometimes, right?
00:26:04.000 --> 00:26:11.839
Where they're bringing in a lot of these elements and showing some proof where they have to have a win to show this is why we're doing it this way.
00:26:11.920 --> 00:26:21.039
And there's a proof point that they then get buy-in with the other teams, because again, they might be relying on the tools that they have disposal, you know, at their disposal.
00:26:21.200 --> 00:26:37.039
So if you can show some of that insight that you're being able to bring with the tools that they have, or even the tools that you're using to show, yeah, I'm I'm gonna tell you that this is the direction I think we should go because of all these things that I'm hearing from all the different parts of the go-to-market team.
00:26:37.119 --> 00:26:59.359
So it helps put product marketing on a bigger pedestal because then they understand how to use the technology to show to the rest of the team what they should be doing or what they should be backing off of, particularly in their focus of coming up with the proper positioning, whether it's for the product or you know, it's or our product line or even just, you know, just directionally of where the where the company should be going.
00:26:59.680 --> 00:27:00.640
Yeah, definitely.
00:27:00.799 --> 00:27:02.160
I I really love this.
00:27:02.319 --> 00:27:17.599
Um I the the conversations, the this topic has, as I think back on a lot of the episodes we've had on this podcast show, has really kind of weaved through a lot of those conversations without directly addressing this like you and I have today.
00:27:17.839 --> 00:27:36.880
We just a few episodes ago, I was talking with Kevin Chan, who's a head of product marketing um for uh locust, who was formerly with Fledio, and he was talking about you know competitive uh intelligence as a signal and to how not to overreact to every competitive signal that you get.
00:27:37.039 --> 00:27:50.960
And so, and it kind of goes back to your thesis for this episode, Constantine, which is that not every signal is worth addressing, not every signal is necessarily trustworthy, and not every signal is a priority.
00:27:51.279 --> 00:27:58.319
Um so, you know, this is this is a moment for all of us to figure out how to make sense of all of this data signal.
00:27:58.799 --> 00:28:00.160
Yeah, yeah, absolutely.
00:28:00.319 --> 00:28:24.240
Yeah, and and I think in the end, again, using technology as a tool, using using these as elements, never forgetting about the human element and understanding, like we were saying before, of having that intuition of trying to think of what we should be doing, and then using that technology to your advantage, where again, having all the the data is net is overflowing.
00:28:24.400 --> 00:28:28.400
Let's just make sure you're looking at the right data and not trying to boil the ocean with it, right?
00:28:28.559 --> 00:28:28.799
Yeah.
00:28:29.359 --> 00:28:30.480
Yeah, exactly.
00:28:30.640 --> 00:28:37.440
Well, how to here how to wrap up, um, Constantine, uh, how does this relate to the work that you're doing with clients?
00:28:37.599 --> 00:28:46.640
Certainly you're you're um working with clients who are dealing with a lot of data sets, new data sets that they've not had before or at a scale that's higher.
00:28:46.799 --> 00:28:48.799
How does this relate to your work with them?
00:28:49.200 --> 00:28:56.880
Well, uh, you know, we we know that our our clients are trying to use a lot of the tools that are out there uh to try to make sense of things.
00:28:56.960 --> 00:29:07.440
And it goes back to knowing what kind of data are we are we looking at, what do they want to hone in on, what is ultimately their strategy and how they should be tuning into that.
00:29:07.599 --> 00:29:15.680
So we help guide them with just knowing which which technologies they're using, whether you're using something homegrown or they're trying to decide on what to buy.
00:29:15.839 --> 00:29:39.920
But at the same time, uh going back to knowing who they are as a company, what they're trying to say, what they're trying to do in the market, and using that knowledge of the market itself, of that history, experience, or even the insight of why was this company founded, what directionally is changing, and helping them uh put that together.
00:29:40.079 --> 00:29:55.279
So leveraging the data that we get from that technology is something helpful, but it helps us then when we're doing some of these workshops and talking to them about what we think is happening based on all this information that they're giving us and what we anticipate what they should be doing differently.
00:29:55.599 --> 00:29:56.160
Wonderful.
00:29:56.319 --> 00:29:56.640
Okay.
00:29:56.720 --> 00:29:58.000
Well, thank you, Constantine.
00:29:58.079 --> 00:29:59.279
I always love talking with you.
00:29:59.359 --> 00:30:03.839
Thanks for a great topic and uh good luck to everybody out there with all your signals.
00:30:04.240 --> 00:30:06.880
Thank you, Jen, and thank you, everybody, for joining us.
