Unlocking the B2B Ecosystem AI Data and the Future of Intent

(00:31) this a lot we'll get to dive in deeper. Bumbora really is so much of, you know, what I consider at least the gold standard or even some of the glue that that ties so many other platforms together. Cuz I I can't tell you how many pitches I've been on where they say that one of their, you know, one of their best features is that they have Bombora data built in. Right. It's almost like Yeah.

(01:02) And I think you've done a great job with this sort of Intel inside. Yeah. U version of it and I'm just very excited to learn way more about the this and and dive in here. And so, uh, Mike, welcome aboard. Yeah, David, thanks so much for having me. Appreciate that warm intro. Um, and kind of hits on some of the themes I think we were planning on covering and, um, excited to be here and and hello to everybody who's on.

(01:29) Uh, great. Well, um, yeah, I mean, feel free to kick things off. got the slides up here and uh yeah, great. Let's do it. Yeah. Yeah. I thought it it would be useful just to give some really fast context on the B2B ecosystem and kind of a little bit about the role that Bomba plays within it and then hopefully just kind of talk more conversationally around what we're seeing in the marketplace and how things are evolving, you know, certainly with AI, but just um kind of at large um you know, with AI being a component, but you know, kind of like these other broader shifts that we're seeing as well.

(02:06) Great. Um cool. So we could just build a little context like kind of exactly like David said uh we were really built around um kind of being this point of glue uh for a lot of the B2B ecosystem. So we think about things in terms of like the full B2B funnel. Uh right. So if marketer wants to run a top offunnel awareness branding type campaign, you know, we play a role there all the way down to the, you know, SDR trying to decide who to make a phone call to on a Monday morning inside of a sales use case. um we kind of uh were built to to make you

(02:42) know high quality intelligent B2B data available across the entire ecosystem across the entire funnel. Um so in the next slide here I'll just tell a little bit about how we do that. So um kind of in the same exact spirit we we built the business around a cooperative of publishers. Um so going back you know 10 years ago now we started working with a lot of you know vertical trade associations, trade publications, broader business focused websites. So think you know Wall Street Journal, Bloomberg, Forbes, Fortune etc.

(03:16) um events companies, lead genen companies our best effort to kind of build a proxy for the B2B internet and across that large B2B ecosystem we're kind of you know boiled down to two things. We're monitoring companies consume B2B research and get a good understanding of what they're interested in, what they might be gearing up to research and buy.

(03:39) Uh, and we're able to track uh, anonymous B2B professionals. Uh, and that data becomes, you know, kind of core to, uh, account-based and other B2B advertising that takes place uh, across the ecosystem. So, in the next slide, we get into kind of like the the different techniques that we have to be good at to do this. Well, um, right, we need to be able to use our tag across this ecosystem, understand what pages are being consumed, uh, how much time and effort people are spending on those pages, what companies the users work for that are consuming those pages. Um, and then build those all into models that give us

(04:17) a really good opinion on what companies are interested in and what what they might be gearing up to buy. So then finally we we take that raw material from the cooperative. We take those techniques that I just brushed past really quickly and we we kind of make these three things available. Uh company level intent data.

(04:37) Uh identity resolution. So this is the ability to say hey here are anonymous users on your website and everything we know about them. Um like what companies they work for and what job titles and job functions they have. Uh, and then like I mentioned, we're also able to to take this anonymous device level data and power a lot of B2B advertising.

(04:56) Um, so the all of that that kind of like great data that I described boils into these three different offerings that that Bumbora brings to market. And then yeah, last last thing here just for context is a lot of what what David said at the top. We're very very much an ecosystem business. So not a not really a platform that you would come and and kind of uh kind of control your entire go to market from but more of a data source that gets distributed out into all of the platforms and places where the actual you know use cases are

(05:28) enabled whether those are adtech advertising platforms or like I was saying all the way down into these sales intelligence um you know and sales tech type platforms and everything in between. So hopefully that just kind of acts to set the stage for kind of where Vombor is coming from um as we watch the market evolve and that type of thing.

(05:52) But um David, I don't know you could you could gut check me if if that all made sense and anything I should clarify. Yeah. Well well it all makes sense. I I'd imagine one of the the most uh frequent questions you get is how do you have this actionable anonymous data? What can you collect? What can't you? and uh and just getting a little bit more of a sense of how privacy works, especially as that Yeah.

(06:17) as the guard rails always seem to be moving on that one. Yeah. Yeah. I think a big part of it is having a direct relationship with the publisher. Mhm. There's ways in all B2B and B toc there's kind of like back doors and side doors where you can build data assets without having a direct relationship with a publisher or the end user. So because we have the direct relationship with the publisher kind of step one we're in their consent string that allows us to store consent from end users as as users are opting into the privacy policies of those publishers.

(06:48) That's really really important. Um so then it like then you can put the B2B specific lens on it and we think through things very much on a company level. So we're not so much interested in understanding that this is John Doe at Boeing. We're really interested in understanding that this is an anonymous user who works at Boeing.

(07:10) And then we're interested in understanding of all of those users that we've tagged at Boeing and think of them as a big kind of like swarm of bees. What are they interested in now more than they normally are, but this is where B2B kind of has a little bit of an advantage. Uh we're not interested in people as much.

(07:28) Um, so between, you know, solid consent and and relationships with the publishers and kind of B2B in general being more of a company level, uh, proposition, um, that's kind of our our POV or where where we sit from a privacy perspective. And and so when you say what people are interested in, like how much of this is the content they're consuming on these publishers or what signals ladder up to interest? Yeah.

(07:54) Yeah. So it's like anything that we can see being consumed across our ecosystem. So it might be a user read an article on let's say hybrid cloud computing. Um but we might find out that that article the user actually landed there directly from a search engine. We'll take that into account. We might find out that um you know they were only on that page for a moment. So we'll throw out that interaction.

(08:18) uh we might found out that they downloaded a white paper on on that topic. So we can wait and score that differently. Um and this is all contemplating the individual interactions but B2B not so much interested in the individual interactions.

(08:38) is we want to roll all of that up to a company level understand normal company level behavior right so to keep using the same example how much does Boeing normally care about hybrid cloud computing so that notice when there's a big spike so we think about it in those two altitudes like being really good at understanding the individual interactions kind of like the grains of sand on the beach but then using all of those to get a really good picture of a company's behavior as it compares to their normal baseline if that if that made sense.

(09:03) Yeah, it it does. I mean, is there a point like you take a a company like Boeing where then it where then you drill down at least some degree, you know, location, division, you know, some kind of discipline. So, it's not just trying to boil the ocean of Boeing and try to see where this is even coming from. 100%.

(09:28) So, and again, this is much most useful in like a sales use case where an SDR or saleserson needs to, you know, really like hunt into a specific signal. And so, yeah, for those use cases, we do we get down to like a geo region. Um, which in the United States can be into like a like a city, you know, like area um or internationally it could be like a province. So, it helps that salesperson kind of do the do the hunting.

(09:55) a little less useful from like a marketing or a campaigns perspective where you know probably want to cast a wider net and reach multiple stakeholders and and that type of thing. Mhm. And uh yeah cuz uh it it's I mean I'm I'm also really curious right now.

(10:17) I mean you've you've been at this for a while, right? you co-founded this company, a ton of experience here building this. Um, like over the past decade, it's easy to say, you know, if I ask you what's changed, like of course AI has changed, but but more specifically, like what is really different for especially your customers and your publishers right now than maybe was pressing for them even a couple of years ago? Yeah, it is a great question.

(10:49) I think one of the things that we're seeing is kind of like the reward around more flexible business models. Whereas, you know, over the years there's been a little bit more and this goes beyond data. This is kind of like just the vendor sphere in general. Um this idea of like, hey, if you're a marketer, you're a sales and marketing organization, we as a vendor, we're going to ask you to make a really big bet. uh you're going to have to pay up for something.

(11:16) Uh it might be expensive, then you got to figure out how to adopt it, how to get value from it, how to do change management around it. Um and it feels like and just kind of maybe as a coincidence or on a parallel path to AI or maybe AI started to change um you know, perceptions on the way things should work.

(11:35) Uh it feels like now what's rewarded is flexibility, smaller bets, things that that kind of like invite these incremental um gains as opposed to like you know asking organizations to make these major points of adoption and and kind of like go through these big change management exercises. Mhm.

(11:59) Um, so in B2B we see like a couple different ways that that like the ecosystem can evolve to kind of like solve for that, but it's it's a pretty stark change I'd say. So does that also mean that there are changes in the kinds of tech companies that are uh that are looking to go and incorporate Bomba data? I think so.

(12:25) Well, I you know over the years it's always been very heavy in B2B tech. Mhm. I think they tend to be the earlier adopters. Um so large you know enterprises like Oracle, SAP, Cisco, Salesforce, Adobe etc etc. Like big big B2B tech companies tend to be ahead on this stuff in our experience. uh but we are starting to see you know much more traction in finance and manufacturing and other verticals as well which I think is a pretty normal evolution now like as get to these more kind of modular flexible business models I think that'll help everyone adopt things faster um because the the tech companies have learned to be a little bit more nimble

(13:03) uh whereas I think you know these newer approaches to bringing things to market can can you know help others be similarly nimble Uh, one thing I'm so curious about your perspective on is that like I'm hearing a couple things from you that sound a little counterintuitive, if not countercultural in the business sense right now.

(13:31) Um because at the at the LLM level there's a lot of often doom and gloom prognostication right now that the that they've hit the limits of available data and you know and then it's like then you just start training it on its own you know on data that's upgraded through their own systems and and so then does that lead to some of the dumbing down of what happens there.

(13:56) But there's also at the same time it's like what other sources of untapped data out there and and it and it is eerily like this whole data is the new oil mantra. Some of what you're saying seems to go against the grain if I'm hearing right and I'm curious it uh feel free to correct me on this too that more data isn't always better. It isn't always necessary either.

(14:22) and that like you can kind of get by with some fewer signals and you don't need every like to actually, you know, do the job. Yeah. No, I I think No, I think we would say that with with where everything is going that more data is still certainly good. Um I think the way that that data gets transacted on, the way it gets made available is what is what we're seeing evolve.

(14:47) Like ideally, you know, synthesizing lots of different go-to market data points is difficult. Large enterprises are okay at it, but still not great. Like all these different data elements that go into who might be ready for a phone call or who might be ready for a sales process. Um, more is better there still always like the ICP. You need to know who's coming to your website. You need to know who's installed the right technologies that make them a fit for your solution.

(15:14) You need to know who has intent. You need to know who's race counting. You need to know who's that executive changes. There's a litany of B2B data that you need and more of it is better, but the way that customers have been asked to kind of like transact and adopt that stuff that I think is what is changing, right? Like you can either buy kind of a big slew of it inside of one platform.

(15:38) Um, and there's pros and cons with that. like naturally like the individual sources themselves won't be quite as good because you're this big conglomerated source you're going to have less control that type of thing or you can go buy it all individually do these big data licenses with lots and lots of like best and breed data providers so painful um and so I think what's changing will change over the next three years is like hey how do you get access to best and breed data more of an outcomebased model like hey I want to

(16:09) pay for what I eat I want to pay for the volume that I need. I want to pay when I get a good outcome. That type of thing. But yeah, if we can if we can bridge access to all the data a customer needs, make it the best available type of that data.

(16:28) Check the box, say you have it, but it's actually the best form of that data, but also give it to them in a way that makes sense from an adoption perspective. That that's where we see things going. Mhm. So, so then with that like uh there's also one thing that may never change is this constant push and pull of fragmentation versus consolidation. Uh and so and so we've seen this with big tech repeatedly that there's some you know rich get richer but there's a lot of uh there are a lot of niche players that are be able to go and launch quickly and like you know and and solve some very specific needs. you see this like like if you you could probably break down the B2B sales cycle, you

(17:14) know, to hundreds of micro steps and you see all these entrance coming in at at different parts there. Uh like are I'm curious what you're seeing as far as uh are we on a certain end of this side of this pendulum and and where is this now? Yeah, I think what what's coming next will be like that synthesis layer because like the actual like I and again I'll come at this from a data provider perspective.

(17:44) I just rattled off all those different data points that you probably need to go to to go to market effectively. I think the like there's still room for solutions around synthesizing all of that data making it highly actionable for a particular end user. Mhm. Um so like you think you can think of that on one hand as like well that's just one part of the value chain. You need all those different points of data.

(18:07) Um and you need a bunch of different ways to organize it and all of those things. Um so in a way this is just another category that'll start to emerge and we see lots of like newer partners in this space. Um, but yeah, I think that's the the next one where new players will evolve and like, hey, who can be best at making prescriptions uh based on the synthesis of lots and lots of different data points. Yeah, it's it's so funny.

(18:34) I was actually just talking to someone on the the GEO side and, you know, generative engine optimization uh and and I was telling them I'm like, I love what you're aggregating here. like when am I going to start getting these like just alerts in my language that say here's what to do with this and and and now I think that expectations also rising right it's not just that like you should be able to at least if nothing more than plugging in some kind of LLM translator on top of what you're doing like then let alone you know building something more sophisticated actually deliver what's more and more becoming like that Yeah.

(19:15) Insight. Yes. Yeah. And I saw like something in the chat as well. And I've heard you talk about this, David, with with some of your stuff, too. It's like we're not quite it's it's we're getting there, but I don't know that like let's say a salesperson would would, you know, work with a tool that could synthesize all the information that's on the internet plus a bunch of important data about an account, get a prescription, and be like, "Wow, all right. That's going to cut through the noise." like stuff that's coming back is still just a

(19:45) little general. It's like, well, yeah, I'm going to talk to Boeing about benefits for an aerospace company. Like, yeah, yeah, they're in aerospace. Um, but but we're getting there. And like I think I've heard you talk about this before. It's like that that period where it could be easy to get disillusioned, but those that like follow all the way through with with the cycle will see that, you know, we're getting very close to where these prescriptions can be very accurate and and timely.

(20:14) So, so then I mean, uh Earl Richards Jr. has taken this uh even a step further than I did as he often does uh and uh talks about you know he says we need to work on helping clients turn their data into information into insights and uh and insights into business impact which is you know is such an important step like what role do you want to play in all of this? Yeah, I mean definitely like the way upstream we're a data company first only so many things we can be good at. So we have to be really good at understanding having a strong opinion at

(20:53) what a company's interested in and we need to be really good at having a pool of anonymous devices that are targetable for B2B advertising like first things first. Um, from there like we can have a strong opinion as to where those data assets add value and how they should be applied.

(21:20) Downstream of that, we rely on the ecosystem and lots of partnerships to make the data actually like go to the last mile and make those business impacts. Mhm. Um, so if we were to say to ourselves, hey, let's build a whole platform around this and be the best at synthesizing multiple data sets, like we'd just be giving ourselves too much homework to be good at at everything.

(21:39) Well, and so, uh, that's why you built the ecosystem that Yeah. And and that also I think relates to Paul's question here in terms of options for small business. Is it really just a matter of the partners that are tapping into Bomba data and if they work with small businesses? Yeah, and it always depend for us blessing and a curse. Vast surface area. It depends on your use case as well.

(22:10) So it's an SDR type use case. We've got a bunch of sales intelligence platforms, you know, like Cognisum, Sales Intel, Lucia, Apollo, many, many others. Um, and then if it's an advertising, you're looking to launch some campaigns and get them into market.

(22:27) We work with lots of partners in adte and ad platforms like Reddit and others. So, it it kind of depends on what you want to accomplish. Um, but typically with a smaller company, you would work through one of our partners in one of those categories. Yeah. And and they in turn want to serve in all these Yeah. all the different levels of the market.

(22:52) And so uh uh where it really just becomes a big marketplace for that. Yeah, that's right. And we certainly have, you know, lots of relationships with large enterprise companies that that buy the data directly and bring it into their own environment and do their own, you know, modeling and and their data science and analytics teams work with the data directly. So we have we have both.

(23:13) And and on the publisher front, like how do you determine what might be the right kind of publisher for you? What might be maybe too niche or too small or how do you filter all that? Yeah, great question. Um, you know, from a high level, like if if you're a publisher and you deal in you're serving a B2B professional, helping a B2B professional do their job, creating content about somebody doing their job.

(23:41) Mhm. You're good for us. Um, we have a lot of coverage across all verticals. So, it's not like that there's a b there wouldn't be any B2B content that we wouldn't want to add into our mix because we're really good at synthesizing it. We're really good at waiting it and scoring it and um you know rebaselining companies based on the addition of new sources.

(24:03) That's kind of like a core strength. So, if you're in B2B or you're business focused, we've kind of like have a really good engine for bringing that content consumption into the models and and making good use of it. Um, and even broader content, it helps us from an identity perspective.

(24:22) It gives us more kind of like grist for the mill of getting better and better at making predictions as to what company a user works for. Mhm. We're that that's a core strength is like we're we're good at synthesizing data. We wouldn't want like something that's purely B to C. That would be a little noisy, but otherwise we can we can take it all.

(24:42) Are do you wind up in scouting mode at all? Like are there areas where you're like like oh man there's some new trends emerging and publish like we want to make sure that this is part of our set like our and our partners sometimes asking you for certain kinds of publishers even specific ones. Yeah. Always. We have a whole team that just runs the co-op.

(25:02) Um, so they're constantly trying to figure out, you know, who are the next block of publishers that we want to try to bring into the fold. And we have had large customers say like, "Hey, yeah, it looks like you have coverage, but we'd love you to work with this publisher." And they help make those introductions. And more and more we're like helping to bridge these use cases together.

(25:22) Like our our audience footprint can be used to do advertising direct to publisher. Mhm. More and more we're trying to connect those dots for our publishing partners. Like we might be creating an audience for a big advertiser and be like, "Hey, if we're sending it, let's say to the trade desk, there's no reason why we can't send it to Forbes and Bloomberg and Wall Street Journal as well.

(25:45) " And those advertisers are doing direct to brand advertising anyway. So, we're just again like kind of like the glue that that's helping to connect the the ecosystem. It's great. And and by the way, as there are other questions from attendees here, uh feel free to enter them in the chat or even raise your hand if you want to ask live.

(26:06) Uh and and we already have an offer. So uh uh Jamie, go for it. Hi, thanks for uh letting me ask. um just in that area where you end or or or start in terms of the synthesizing of data, do you ever get predictable about things? And what I mean by that is go back to Boeing.

(26:34) We're noticing them becoming more nimble or we're noticing them becoming more um uh want their their want to change as a as a culture of the company. Is there is there an overarching culture or personality that that uh you kind of label in some way, shape or form about the companies uh to help with uh predictive models of how they might react in in different ways to different stimulus like AI coming in and are they going to react quickly or not or the economy is going to do something how are they going to react? No, great question and I appreciate you joining my Boeing example. Spent 10 years using Boeing as

(27:11) as an example for tracking account behavior. Um, so yeah, one of the core strengths is this historical kind of highly structured historical data set. So we publish the intent data week over week going back many years. And so this allows us to look at any outcome.

(27:35) let's say um a set of companies that um did a bunch of work around ESG uh and we can look at the behavior of those companies over time and to your point then find other companies that are behaving very similarly. Some of those behaviors might be really obvious like they have these spikes in research on topics directly related to ESG.

(27:56) Some might be, you know, weird corlaries like, hey, companies that tend to invest a bunch in HR and employee relations also tend to do really well with ESG. Um, so yeah, like anybody, we do this on behalf of customers and customers certainly do it themselves on their own. Like anytime you're looking to make a prediction around a specific outcome, you can build that model using the historical data.

(28:21) Great. Thanks Jamie. Uh Earl's got some questions too. What's the typical data maturity of your publishers that you work with? It's a mix. Um like the ones that you know the publishers we work with are in all different businesses. They have different revenue models. So the large kind of business focused publishers I talked about, they're in the advertising business.

(28:46) uh and they're pretty good at applying data into their advertising campaigns from a targeting perspective and from an analytics perspective. I put them on the higher end of maturity all the way down through these like longer tail niche B2B vertical publications mostly in the lead genen business sometimes in the in the events business.

(29:07) So they're using data to better promote campaigns or to better do lead scoring and things like that. So some of them are quite good at that. you might think of it intuitively as like a little less sophisticated. Um, but it's really just a different way to apply data that's not adtech focused.

(29:25) Um, so it's a mix and it certainly depends on kind of like what revenue model you have as a publisher. Right. And then and then which teams from your publishers do you usually work with like audience strategy, bisdev and sales, marketing, analytics, etc. Yeah. Um, also a mix I'd say on the larger publisher side that are mostly in the advertising business, it's a lot of ad ops from a day-to-day perspective, right? Are the ones launching campaigns, providing reporting and analytics against campaigns, that type of thing.

(29:56) And then like down in the the longer tail of the B2B niche trade publication, we're working with executive leadership um to get buy in to the cooperative and then to start to like connect us into the other teams that can make use of the data. I'd say like you can imagine um kind of the more niche you are a little bit more closely you'll hold your data. Mhm. Uh right.

(30:20) Like you have this really specific audience of um banking professionals in the United States. Very few publishers have this kind of like condensed access to those users. So you're going to be a little bit more kind of like thoughtful about how you enter into a proposition that shares that data. Um and we have an amazing co-op team that's that's you know taken our kind of approach and our ethos as a business and like done a really good job evangelizing that. uh across the publisher community.

(30:52) So I you know I'd be remiss not to ask a bit about some of the challenges overall that that are happening and I'm see a lot tied to B2B data. Um one of them is is that there are a lot of players out there that that you know are all about quantity and not about quality at all. Right.

(31:21) Um, and that uh that I mean my fear is that the problem of noise and especially noise to signal and some talked about that signal to noise ratio in the chat is just it it's you know it's like the tragedy of the commons. It's make things worse for everyone because there are a lot of bad actors out there.

(31:43) Um uh and so how much quality control does Bombora need to do with the partners you're working with in the ecosystem since you are some of that glue out there? Oh, interesting. So like when we think about companies that we will partner with. Yeah, that's that's interesting. Like I think like you know we want to position hopefully rightfully as like the gold standard when it comes to B2B intent data. Mhm.

(32:08) So the partners that are working with us, they're making a larger investment. Mhm. Work with Bomba, make it available to customers. So right away we're kind of like align because there are ways to your point, there's they're we live in a, you know, a competitive landscape where there's other flavors of intent data. U but it's kind of like self-qualifying.

(32:25) And the partners we're working with are making that investment and they're using this data set that has, you know, we know where it's coming from, consent from end users, true baseline so that we understand when there's a real spike in interest in research. Um, so hopefully it we're taking care of that by, you know, finding customers willing to make that investment.

(32:48) So this is fantastic because, you know, data is the new oil and all that. Uh my question is going to make uh Burkowitz roll his eyes. It's about blockchainbased solutions around and I don't want to get techy but around the uh decentralization method just to make that data more reliable. So the middleman is out of it and anyone with any intent is kind of out of it.

(33:11) How are you distinguishing your proof uh using blockchain or any of those kinds of digital uh DT services um ledger stuff like that? Yeah, I I wouldn't say that we're actively like in that right now, but I'd love to hear and not to turn it back on you an example from what you know of our business of like how that could apply in a customer context or otherwise.

(33:36) I think we have lots of ways that we do validation. Um, but measurement and validation is probably like the most important thing that that we can do for customers. So if there are better, smarter ways to approach that, I'd love to hear with I I'll I'll talk to you about it later, but one thing that might be a relevant topic for everyone as well is with uh the nano banana kind of thing.

(34:00) It's just like the latest flavor, no pun intended, and all of a sudden the line is everything you see on the internet is now for sure not something that you can rely truth. I mean, it's just so good. Um data is the same thing. It's not just imagery and things like that. How are you tackling that from a differentiation standpoint so you can sell guarantee your your your information? Yeah.

(34:26) So we we touched on it a little bit before but it starts with like if if we can if we can be provided a truth set then we can compare the behavior of that truth set to a control group and find a distinct behavioral pattern. Right? So let's say we use Boeing uh uh we know that um Boeing was selling uh something and a 100 customers bought it, right? We know we we have that truth set.

(34:51) We know the dates that that that it was purchased. We can find a distinct behavioral pattern from those 100 accounts, a research pattern that we saw across our cooperative that differs from another 100 accounts of about the same size. And we can go deeper into specifically the how we do the control group creation, which is really important.

(35:09) But once we can do that, we basically know the behavioral pattern and we can be pretty good at at identifying other companies that are behaving the same way. It's like the the Facebook mirroring thing. It's just pretty obvious. Yeah. Yeah, that's right.

(35:26) What's funny, one trick of it is let's say we get these I use this number of 100 closed one accounts from a customer. um we can't just compare it to a control group of some other hundred accounts. We need to make sure that those other hundred accounts are about the same size and that we see them the same amount, right? Because if we just take another 100 random accounts that are maybe smaller or that we don't have the same visibility into, it's hard to make an applesto apples comparison. Yeah.

(35:54) So the fact is you have all that comparative data means you can match apples to apples. Yeah. And it's also a big advantage of this idea of kind of like a closed ecosystem that we can observe across. So like other kind of more temporal data sets, they're typically able to make observations across kind of like a moving target.

(36:16) Um like okay, I see these behavioral interactions on the internet, but where on the internet is always changing. The access I have to the internet is always changing. So the reliability of any like kind of trend data or historical data that comes from that is very low. Uh but we have this kind of like controllable ecosystem that we can keep like kind of like tuning and relying upon and that like allows that historical capability to be have a lot more fidelity.

(36:43) So one one last point here's a radical idea talking about it in a in an MIT think tank last weekend and is how do we make this advertising supported future that we in this group would wouldn't mind. um uh how do we actually get the bad guys out and that means maybe the company's not making a billion dollars opening it up like Zuckerberg did and said everyone's welcome come on and use the system to a you know what we're going to vet you we're going to make sure you are who you are and uh make sure that that the ecosystem advertisements aren't going to you know rip you off there's certain criteria uh is that an association element is that something

(37:14) that AMG could actually champion is it uh interactive advertising bureau I mean who would police that and actually have a good housekeeping seal of approval of here's a a marketplace where we've we've kind of assessed and we just haven't just opened the back door. Yeah.

(37:41) And and then like how do we deal with some of the fundamental challenges like like uh one of my perhaps less popular opinions is that it's just too easy to buy domains. Yeah. Uh, and so you just see how how how some of the access and and ease and all of these seemingly really good things have created some intractable problems. So register your domain and your gun. I hear you got it. There you go.

(38:06) The and there was yet another David in the chat uh Mike just to as a kind of clarifying point is saying so exclusivity to the data de facto establishes veracity which becomes the competitive differentiator. Yeah, that that's definitely a big part of it, right? Like there's um I think it's 86% of all the interactions that we're able to see uh on a daily basis are exclusive to us.

(38:30) Um so that that certainly helps. Um now a lot of the publishers that we work with um their data is very difficult to get to for another provider anyway because like they don't make their inventory available and the bidstream is just not kind of out there. Um but that you know having exclusivity over the other chunk helps as well.

(38:52) And then and then also relating to some of this as far as like yeah you talked about kind of normalizing data uh and and Earl had another good question. So uh have your existing clients shared their numbers with your team so your teams can better weight the value and impact of your data that they're using and uh for data source scoring like what data do you get back? Yeah.

(39:23) So that's like we mentioned it a couple times but if a customer you know we're not a platform so it's not like we're constantly birectionally reading our customers data changing our algorithms and sending it back. We have partners that that kind of get there. But yeah, we have lots of customers that'll periodically three or four times a year send us some closed one data and allow us to like retune the intent signals that we're providing to them. Mh.

(39:48) So we might find that hey, you know, customers that close one, they tend to spike on this set of topics at this time uh at this kind of like intensity level. And we can kind of like retune their signals accordingly. M and and when you're talking about retuning signals, how much of that winds up being specific versus general? Um well, it's always very specific in that it's, you know, based on their closed one data.

(40:16) It's for that specific customer. Now, like kind of by design, we do live in a world of a set structured taxonomy. Um, and there's a bunch of good reasons for that, but so it's always going to be within that kind of like um intentional limitation of like we can only return signals back against this like set of topics that you know we monitor for across the co-op.

(40:41) Does that did that answer your question? Yeah. Yeah, for sure. And so um now looking forward like yeah what else is it uh can you share anything more about like where the road map is or or other things that are like coming up for you? Yeah, we talked a lot in amazing engagement.

(41:11) Thank you on this call about uh our intent data business, but we're also uh innovating quite a bit around our audience business. So I mentioned it earlier in the overview. Uh as part of the cooperative, we have anonymous uh B2B professionals and we're able to tie B2B attributes to those devices and make those addressible for advertising. Mhm. And that part of our business is uh you know growing very quickly and we we have a bunch of newer things that we're working on for that kind of theater of our business. Um I'll mention a couple of them. One is ABM measurement. So a

(41:41) big chunk of B2B advertising is targeting a set of accounts. Mh. So IBM wants to reach a set of a thousand key accounts. We have the ability to take that list of accounts, turn it into an addressable audience. Um we we've always been able to do is make that addressable and targetable.

(42:00) Uh but where we've spent a lot of time over the last couple of years is also being able to deliver back to IBM what accounts they reached uh at what frequencies, what job titles, job functions, etc. So providing the reporting back to IBM in this example. Um so that they can pull that into their data environments and build that into their models as well.

(42:20) And so, and is that something that it sounds like things I've heard from some of your partners having an unknown version of that? Yes. So, this goes back to this idea of kind of like in a more open ecosystem, more flexible business models and our partners have amazing, you know, offerings in their own right. That's a model.

(42:46) That's one way to do ABM advertising. You buy into a platform that has a DSP. that DSP comes with reporting. Great. Uh there's another way that we see the ecosystem working, which is I already work with the trade desk or I already work with DB3. I want to do my ABM advertising on those large DSPs, but I don't want to have to sacrifice reporting.

(43:06) I want to be able to get ABM reporting there as well. Um and so that's that's one of the areas where we've innovated and be able to provide that reporting kind of in a more open ecosystem model, if that makes sense. Yeah. Yeah, for sure. And and if there are one area within the audience space, um another is like starting to work with other account level data, I'll say brethren who have really interesting account level data attributes and helping them to translate their account level data into targetable audiences as well. Um so the number of those partnerships coming out where it's like, hey, I want to do good ABM or good B2B

(43:46) advertising. I love Bombora data, but I'd love to be able to have some of these other again best and breed premium branded data sets that I can also target with advertising. Um, so it's another way that we're looking to kind of democratize access to, you know, high quality data for use cases in a flexible way. If that if that makes sense. It it does. And and I'm glad you also brought up ABM.

(44:13) I mean this was where when I first was encountering Bumbor in the wild. It was from ABM platforms that were all saying they integrated with you and and I had to dive deeper and understand well wait like why is this data so good and powerful like like why like you know why why are they touting this right? Yeah.

(44:44) and we we work with all the ABM platforms and they're great partners and there's just there's different customers want different models. Um you know for some customers that platform model works great and they have big businesses and successful part of the ecosystem and then other other customers they might operate in a little bit more of kind of like a open environment where they use different platforms for different use cases and they want data that can connect through to that kind of more disperate stack.

(45:08) I mean, do you get a sense if there's any shift and and I know this is a bit removed from what you do specifically, but from what you're seeing out there in terms of how people are acting on this where there's some more of like the mass targeting everyone at, you know, everyone at Boeing, right? Uh or everyone in a certain geography department. Yeah. Something like that.

(45:35) versus trying to take this and then see like, okay, we've got these five contacts at Boeing. We're going to now go deep in these onetoone outreach and uh programs here. Are are you seeing any shifts in like appetite for doing one more than the other? Is that changed at all? Yeah, I think it's use casebased.

(45:56) So certainly sales use case, precision, precision, precision. Mhm. I think inside of like an advertising use case, you can kind of like um reach some diminishing returns, all right, congratulations. You found the the right 10 people at Boeing, but it's like, hey, there's a bigger committee. There's like it's a long sales cycle.

(46:21) If it's a top offunnel campaign, you want to build much more of a chorus. So, it really depends on the on the use case. That sounds a bit like what Mark Pritchard at PNG had said early on about just targeting on Facebook that they found that like once they segmented the audience too narrowly then they were missing a lot of that opportunity. Yeah.

(46:45) That and it's like it for a lot of platforms uh we won't mention any specifically but some of them like it knocks their algorithms out of whack. So like you're not getting any more. you're actually pay then you now you have to pay a lot more to reach this small group of people could probably have paid the same amount still reach those people cast a wider net so depending on the platform it doesn't microtargeting can hurt you sorry guys I uh put my video on because I'm in the islands and Caribbean I know you're jealous this is my tan um yeah my

(47:17) question was just to get clarity on working with a platform like Oracle depending on the clients obviously they have a different text stack depending their level of maturity and sophistication we get that so I'm curious how does Oracle integrate into their platform to the data because you're adding additional data for their target accounts their audience segments so I'm just kind of curious how the plumbing works essentially yeah and you mean if you're a customer that's like doing go to market work on the Oracle stack like Unity Eloqua that

(47:47) type of thing yeah well more like if you are providing the data for whatever whatever initiative that they have whether it's B TOC or B2B how do how would a client go about integrating your data into their stack so they can use it got it so in in this case Oracle is a just kind of like a a hypothetical customer of ours you know Oracle is the vendor providing the data to a potential SMB or agency or tech solution where does Oracle play a part I guess what's their role I guess yeah so one one way we would work with a customer that might be like built around Oracle tools um like in Eloqua or Unity

(48:26) their CDP and so we would be able to say like okay we curate these intent data signals kind of upstream and then we're able to push those into really any environment Oracle being one of them u and make that data actionable for different use cases um that space of like what used to be more of like the marketing automation space I think is evolving into a CDP type space um is a good example of like Those are great places to to integrate the data because they can then push downstream into sales use cases or lead scoring use cases or other kind of internal systems um that

(49:02) that can you know get customers to outcomes. Got it. And obviously or Oracle and the partnerships you have your own tech solutions to CDPs for examples and marketing automation like Eloqua. So I was just kind of wondering how that works into clients who have their own solutions in place and that kind of conversations.

(49:21) That's usually what I went into. It's always like we work to get as elegant and like you know we want the data to be ubiquitous. So kind of like one end of the spectrum is like no we don't have a direct integration but we have a team that can get the data into that particular CDP.

(49:41) Maybe it's you know a CD not a lot of button used or something like that all the way down through like yep we have elegant you know out of the box integration with CDPA and like uh or Adobe CDP is a good example of that where a customer but were to buy access to the bombore intent data we can turn it on you know immediately in the Adobe CDP because there's a pre-built integration. Got it. Thank you.

(50:04) Yeah, appreciate Mike. Any uh further thoughts like what's ahead for you? Things that should also be top of mind for folks as you know as marketers and others in our space just try to make the most of our access to data and what we do with it. Yeah. Yeah.

(50:29) I think like it try to like tie it together that the themes that we we've covered is like think about things through the lens of specific use cases and outcomes. Mhm. How um a company like Bomba or others can like kind of in a more localized precise way get you to that that outcome um in a flexible way so that you don't have to boil the ocean and think about you know a complete change management project or you know totally changing your sales and marketing culture in one swoop.

(50:59) uh but how you can get premium best in breed data for specific outcomes uh in a way that that is nimble uh and allows you to move quickly. Great. Well, well, this is tremendous. I mean, it's it's so great. Yeah. just again like coming across Bombor so much in the wild and uh and yeah, having been an indirect customer of yours many times over than uh then getting to understand so much more about how it works and and how this whole ecosystem comes together in a way that that I think a lot of us might not fully appreciate and and I think allows so many members here to just ask smarter questions and and look

(51:35) to do just better work with all of this because you know without great B2B data we're all yeah none of us could do our job so so appreciate what you're doing to not just in the ecosystem but to help connect it of course yeah no I appreciate you having me and all the great engagement and questions and you know very uh detailed granular uh appreciate the thoughtfulness excellent well uh you know come by again anytime it's great to connect with you here and thanks to you and your team uh for making this happen. Great.

(52:13) Thanks everyone. See you. Bye.