Bloomfire CEO Philip Brittan Interview Transcript
Clint Betts
Phillip, thank you so much for coming on the show. You are the CEO of Bloomfire. You've had an impressive career even before that in a lot of different roles. You were a VP at Google, for example. You've been at Crux Informatics; you were the CEO of that. Yeah. You've had some incredible roles, and now you're at Bloomfire and you're in Montana. So, give us a sense of what Bloomfire does, how you got there, and I'll follow up with a question on Montana.
Philip Brittan
Okay, terrific. So Bloomfire has been, for a number of years, a leader in the knowledge management and enterprise search space, and it's used by large and small companies, including a bunch of name brands and then kind of across the spectrum to manage their knowledge and get the most out of their corporate knowledge, and we're industry-agnostic and we serve customers across a very wide range of industries. You can name an industry, and we've got customers in it because we solve really very core corporate needs that are common for any kind of business. As I said, the company's been around for a while and is considered a leader in the space, and regularly has won and continues to win. We just won two awards recently, industry awards, as recognized as a leader.
I got involved... As you say, I'm based in Montana. I grew up on this ranch where I now live, lived in New York for many years, and then came home about three and a half years ago full-time, and live here and have been involved with a number of different things, including working with a bunch of enterprise software companies or enterprise IT, generally speaking. And through that work, I got introduced to Bloomfire and was invited to join the board last fall. I think I started in October. We were having discussions over the whole course of the last year, and I formally joined the board last fall. And then early this year... And sorry, my role on the board was to be the independent board member, so not part of the investor group and not part of management, and also specifically to think about where the strategy for the firm goes in the next chapter.
It's a successful, growing, profitable firm doing very well, but the thinking was, "Hey, what does the next chapter look like?" And a lot's changing in the world, the advent of AI, obviously, I'm sure we'll talk about that, and just changing customer expectations, changing market conditions. You always have to keep evolving, and so I was tasked to think about our strategy for the next chapter, and I also think about the company's life in terms of chapters. It felt to the folks inside the firm, just at the time that I joined the board, that it was time to kind of think about where we go for this next chapter of how many years.
And so I spent a bunch of time really looking at the space and came up with this idea of enterprise intelligence as the natural convergence and evolution of knowledge management and enterprise search, which we were in both of those already, plus business intelligence, and I realized the commonality of these things. They may look different, but they're all about helping companies query their corporate knowledge and get the most out of it. And there's no reason for them to be separate; it's just historical, and so they could really come together. So I did a whole bunch of work, research, talked to a lot of folks, and put together this concept and a set of principles to drive enterprise intelligence, and I presented it as, let's say, the conceptual flag in the sand for Bloomfire to go after and really execute against for this upcoming chapter.
And then an opportunity early in the year came up for me to take over the CEO role, and the board asked if I'd be willing to do that, and after some soul-searching, I agreed to do so. I think it's a wonderful company filled with wonderful people. It's in a really exciting space, much needed by every business on the planet. It's an extremely, extremely common set of needs. And so I was really excited about that scale and about this strategic direction and being able to really flesh out enterprise intelligence, not just as a strategy, but also to be the one driving it in reality. So those were the drivers, and it's been terrific. I've been in the seat for about four months and obviously gotten to know the team very well and gotten to know many of our customers, and it's an exciting journey.
Clint Betts
How has AI accelerated Bloomfire? It seems like what you do with the data collection and the way you can offer all these insights into your business, it seems like AI just superpowers that, right?
Philip Brittan
Oh, it does. Absolutely. So AI has had an impact on the industry generally. At the very least, a lot of people are talking about it and running around and trying to do a lot of experiments and get their arms around what it means and what it's going to mean for the future of work, for employment, for skills in the future, and what's possible, productivity, et cetera. Definitely in our space, where we're taking knowledge and surfacing insights based on that knowledge, is directly applicable, because it is fundamentally an insight engine that sits on top of knowledge, on top of data or content of some kind.
And so we recognized, and before my time, the firm recognized that AI was really a powerful additive capability to what we were doing, and quickly moved to adopt it and build AI capabilities into the product. So previously it was very much about managing, curating content, authoring content, and then searching through the content, and with AI, we have in addition to all of that, also the ability to talk to your content and ask a question, not necessarily just get back a document or a set of documents that match the actual answer. So we have that in Bloomfire, and we've had that for a while, and we continue to make that better and better. We're very excited about all kinds of ways that we might be able to use AI in novel, in very practical, very, very grounded ways, not breathless about it, but we understand deeply what it is.
And I have a pretty deep background in AI myself. It's what I studied in college almost 40 years ago, believe it or not. So we think there are all kinds of applications, certainly for surfacing insights, both reactive, customer asking a question, and proactive, say, "Hey, given what you do and given the kinds of things you're interested in, here's some information or here's an insight that may be relevant to you." So we think that the proactive aspect is under-leveraged in the market today and is much needed. The unknown unknowns are as important as the known unknowns. [inaudible 00:08:14].
Clint Betts
Yeah, I think that might've been like a Dick Cheney quote. That's cool.
Philip Brittan
Yeah, or Donald Rumsfeld. Yeah.
Clint Betts
Yeah, yeah. That's actually a more coherent way to use it than he did, if I remember right.
Philip Brittan
Yeah, and then we also think, look, AI can be used not just for insights, but also we have a big focus on cleaning data and having a self-healing knowledge base, and we think we can use AI, very specialized models to go after specialized kinds of tasks like identifying... We're already starting to do some of this. Identifying when you have documents in your knowledge base that are really redundant, that are 90% overlap, and should be brought together, or when two documents directly conflict with each other semantically. It's very hard to do if you don't use AI to understand what the documents understand in some sense, what the documents are saying, be able to semantically look at, "Oh, these are saying the same thing," or, "Oh, these are saying the opposite things," or, "These are saying different things and that's fine. They're complementary." So we think there's a lot of opportunity to be using AI models and really kind of a collection of models that are very specialized to do all kinds of tasks, all collaborating together.
Clint Betts
As you think about AI and having studied it, where do you think we're going here? I mean, you've got all these doomsdayers, you've got people who think it's going to be the greatest thing ever, and then they've got... It seems like the practical application, at least currently, is what you are doing with it, which is really kind of fine-tuning it and having it do very specific and powerful tasks. Where are we at here? What's going to happen?
Philip Brittan
So I think it's early days is what I think, and that's the easy answer, but I really believe that, and I think some of the evidence we see of that is that there's pretty early and up on the peak of inflated expectations in the classic hype cycle, and a lot of folks are jumping on it and saying, "Oh my gosh, AI is a future," and there's a lot of prognostication obviously in the industry about, "Oh my God, AI changes everything," or, "Oh my God, everyone's going to be unemployed," or all these sort of... Yeah, I don't know. They're doomsayers or just people with a melodramatic interpretation, let's say. But my belief is that nobody really knows it's too early. And some very specific evidence we see is that all kinds of companies have been rushing out, understandably, to do a lot of experiments and figure out AI, and hire people who know what they're doing with AI to try to understand.
Nobody wants to get left behind, and people, I think, see the promise... In some conceptual sense, they understand it's valuable if they can figure out how to use it, yet you've probably seen some of these headlines. A huge percentage of these projects are being abandoned, and some of that is fine. It's early days, people are doing experimentation, and when you experiment, you expect to abandon a bunch of experiments and kind of move forward with the things that are working. But I think a number of firms have been kind of surprised from what I've read and what I've heard in the industry by how not magical it's turned out to be, and so I think we're still in a phase of, there's a huge amount of promise and people are starting to figure out some areas where it's very useful.
We certainly do within Bloomfire. I do, I use AI models every day to help me in various ways. I think they're wonderful tools, but you have to understand that they are tools, not magic, and you have to use them like any tool; you have to use them appropriately. If you try to use a chainsaw to nail a hammer and a nail, it's not going to work. So you've got to use it the right way, and you've got to use it for the right job. So I think companies are starting to come to grips with that, and I think they've had some issue of understanding that AI has a tremendous amount of potential and can do a lot of things, but there's been, from what I understand, some struggle to really connect the capabilities with the use cases, with the value, like where is it actually useful, where is it really valuable, because it doesn't determine that for you on its own.
You have to... Again, it's a tool. You have to use it for something. And I think some firms are still casting around to figure out, "Okay, where is it useful?" And the other is just getting the basics, and again, it's a tool, not magic, so you have to use it in appropriate ways, and a big part of what we are focused on at Bloomfire is accepting and understanding that AI is only useful on top of some kind of content base. It is fundamentally... These tools, even generative AI, are statistical analysis tools. They're very, very complex and very sophisticated, really amazing. They're amazing, but they are statistical analysis tools. They take big piles of training data or in context data, and they look at statistical relationships between the elements in that, and then based on what you're asking me, they give you an answer or some kind of insight that's based on statistical relationships, and while the answer may be novel, it is based on this content.
And so the content has to be good. A model with no content doesn't do a thing; it does zero. So this is a big focus for us, saying you have to connect to all your content. If you have it just operate on this little bit or this little bit or this little bit, you're going to be missing the full value or miss the whole story, and if you don't clean, as we were talking about previously, if you don't clean up that data, make sure it's not conflicting, make sure it's consistent and harmonious and doesn't have holes or things in it, you're going to get unreliable outputs.
Back to your question, I think companies are struggling because they haven't taken that piece of it seriously enough, and there's a lot of folks saying, "Oh. Wow. We've got a giant data mess. Let's just throw AI on top of it and that'll solve it," and it doesn't. Garbage in, garbage out has been true of computer science since the very beginning, because computer science fundamentally is all about processing data and performing some kind of calculations on top of data, and AI is exactly the same. It's not different from everything that's come before at that very high conceptual level of anything else in computer science. And so garbage in, garbage out is as true now as it's ever been. AI does not change that, and in fact, it may be more pernicious because these AI models sound so credible, no matter what you feed into them, they come back with a very convincing and very confident answer, and you may not spot the problem.
If you look at a chart and you've got spikes or discontinuities in your chart, let's say you're looking at a typical business intelligence dashboard, you say, "Ooh, I see the data is messy. My chart goes like this, and then it goes like this, and then it goes like this. Something's going on. It's worth a look," but if you ask an AI model that's sitting on top of really messy content, it comes back with a really good-sounding answer what they're all about, and it won't be obvious to you that there's something wrong underneath unless you really dig. So that makes it even more important than ever to make sure that what you're feeding into it is clean, because it's going to be even harder to spot later.
So, just saying, "Oh, I'll just throw an AI model on top of our mess," we believe is not a good solution and just won't get you very far, and you'll run into problems. And we hear a customer say, "Yeah, we did this thing, we used this vendor, or we built this thing in-house, and we're getting hallucinations, and we don't know what's right, or we can't trust it. We have to check everything, and there's more effort than if we'd just done it ourselves." And so there's a little bit of that, the trough of disillusionment starting to appear at least in some ways, and again in the hype cycle because of some of those early findings. But it's just a question of using it in the right way and having the right set of expectations, like anything else.
Clint Betts
What does a typical day look like for you as CEO of Bloomfire? And you're still new to the role and all that type of stuff, but give us a sense of what your day looks like.
Philip Brittan
It varies a lot depending on the needs of the day. I work from home for the most part. My office is on the ranch. Like Jeff Bezos, I believe in getting up early and then having some putter around time first thing in the morning. I don't know if you read that. Jeff loves to spend an hour in the morning just puttering around and doing things, and I do think that is helpful, to get ready for the day. I will often just sort of work on a few different things, and I love to exercise. I exercise pretty much every morning, and so that's an important part of my life.
And then my day will have a variety of different things. Obviously, we've got internal meetings, I'm talking to customers, I'm talking to folks like industry experts like yourself in different ways. I spend a bunch of time thinking, analyzing, and reading. I think that's very important to be continually curious and trying to absorb new things I've been using. Back to the previous question, I do use AI to help me do research, particularly initial like, "Oh, go tell me what's going on over here," and then I can dive into things, and that's very useful for that. I live on a ranch, so sometimes I get to go out and feed the horses or do things, and I very much enjoy having my day a little bit. It can get a little crazy sometimes, just time-wise, but I enjoy having a few different things.
Clint Betts
Living on a ranch as a CEO is a great escape. I mean, you're like in business all the time, and then you get to go out in nature, feed the horses, maybe get some fly-fishing in. I don't know if you fly fish. You should if you're in Montana.
Philip Brittan
I do. I fly fish and we've got a beautiful trout stream just right next to the [inaudible 00:19:33]. Well, I can very easily grab my rod, jump out there, fish for 20 minutes, and come back in, and that's a doable thing.
Clint Betts
That's so awesome. You have the perfect setup for that. Give us a sense for Montana, how it's grown as a tech ecosystem, how it's growing as a state just in general, and where is it? I mean, I don't think a lot of people think of Montana as a tech hub, and you probably don't want them to.
Philip Brittan
It's a two-edged sword, and so the tech ecosystem is very much alive and well, and it's been growing fast. And as you know, Clint, I'm the chair of our main nonprofit for encouraging the tech industry in Montana. I've been on that board for years. Yet another thing I do, and I'm currently the chair. It's called Scaling Montana. And we engage, encourage, educate, and connect basically people in the tech industry here, and we do a variety of different things. We have a monthly speaker series, we do an annual summit, we do mentorship for startups, Montana-based startups, and we do a lot of connecting of different people and resources around the state. And we have a mix here of a bunch of startups, at an earlier stage. Some of them are starting to get pretty far along, a few really great success stories of Montana-based technology companies, and then larger companies increasingly have a presence here.
So when I first moved back to Montana three and a half years ago, I was working for Google at the time, and I was working here remotely for Google. And all the major tech companies have some presence either with an office or a handful or more than a handful of remote employees. So increasingly we're seeing this presence, and the great thing about that is, like any ecosystem, it's that set of connections between all the people and you meet like-minded people in the bar or there's a meetup or there's a conference or whatever, and when you share ideas, you collaborate, you encourage each other, you egg each other on so to speak, you inspire each other, and all of that is really good for the tech ecosystem. So we're seeing that. I would say it's earlyish days in Montana, but not super early.
We're getting there and we have a couple of... We've got a lot of software here. We also have a very strong focus on photonics and quantum computing here. We have one of only two full quantum test beds in the country. One of them is located here. A lot of that is coming out of work being done at Montana State University, where there's a lot of really neat, particularly in material sciences and physics, and a lot of that is more hardware-oriented, but obviously it has a huge impact on software. So there's some really, really neat stuff.
To your question about, yeah, do we want to encourage it or not? We certainly want to encourage economic growth in the state, and that's part of my mission to help the state and give back. I'm also on the board of a public policy think tank here called the Wheeler Center as part of that same sort of mission of wanting to help the state. We're kind of divided here between even within individuals, not just the population divided, but people flip back and forth between we want more people to fuel growth, but no, block the border. Don't let anybody move to Montana; all these people from other states are coming here. I think we've been a growing state, obviously from day one, as the whole West, the whole country has been, and that continues, and it seems to be fairly healthy, and people bring...
People bring different ideas, different experiences, that sort of diversity of perspective, et cetera, which is good and that's healthy for the community and the ecosystem, but I also find very encouragingly that people who move here from other states want to be Montanans and they want to fit in and they want the genealogy of Montanans are just very nice and polite, a little soft-spoken, but very polite, and I find people, even when they come from rather let's say more brusque parts of the country come here and they want to... Everyone waves to each other on the street, and people are generally quite polite. So I'm very hardened by that, that people come here to join the culture, not to change it.
Clint Betts
Yeah, they call it the last best place for a reason, man. It's beautiful up there. It's just-
Philip Brittan
I feel very fortunate to have grown... I grew up here on this ranch, as I mentioned, and I feel very fortunate to have had that experience and then extremely fortunate to have the ability to come back and continue to live on the ranch.
Clint Betts
How do you think about culture and maintaining culture, particularly as most days you're working remotely? Is the whole team remote?
Philip Brittan
The whole team is remote. The whole team is remote. We're officially based in Austin, and we have a small office there, but it's just a couple of people. We have been remote for a number of years, and it works very well, and I think... Honestly, COVID changed my view on all this. For years and years, I was very much an in-office person, and I wanted people in the office. I believed being next to each other in front of the white board or having those fly-by conversations in the hallway or the micro-kitchen or whatever, and I still believe in all that, but I think when COVID happened and all of us and I was running my fourth startup at that time, all of us had to go home, it opened my eyes to, "Oh. Actually, this can work if you're intentional about it."
And of course, what COVID did that made this work is that everyone was home, not some people were home, and most people were in the office. That's harder. But when everyone is home, we all as a society have to practice it at the same time. We figured out that it actually works and people can be very productive, and we've seen there are a lot of stats around that, and there are a lot of benefits to it. Now, you bring a very good point to the table about culture, maintaining company culture. I think one, interestingly... We've had video conferencing for a long time, but it wasn't much used as you probably know before. Before COVID, it was around, and we had Cisco and a few things for video; mostly, we'd meet in person or we'd talk on the phone.
When I was at Thomson Reuters, I was managing a team of 6,000 people, and they were scattered all over the world. They were all in offices, so we didn't really think about it, but it was highly distributed. I'm not sure what having three people in an office out of a huge team like that is versus working from home. Looking back, I now think that's kind of an academic difference. So I think we were operating more... Certainly in larger firms like Thomson Reuters, we were operating a more distributed kind of work style than we realized already, and so we were maybe a little more practiced at work from home than we thought. But interestingly, we had VC capabilities. We used them, but not that much. Mostly, people met in person or talked on the phone, and COVID just made it okay; now, everyone's going to use video conferencing.
And I think that was a good thing, and I think we've learned how to connect with each other, and video conferencing obviously brings in higher bandwidth communication. Not as high... If we're in person, we have the highest bandwidth communication, so to speak. We can hear, we can see, we can get a sense of people. On video conferencing, we can hear and we can see people at least to an extent, a little bit flattened, but still we can see people and we can pick up their body language, and I think we've... Versus the phone, all we can do is hear them. We don't know, are they rolling their eyes? Are they drifting off? Are they multitasking? I don't know.
So I think the VC innovation has been helpful for companies to connect. And I've had this experience over the last number of years since the beginning of COVID where I've only talked to somebody over video conferencing and then met them, and was sure when I met them live that I had met them before, and they're like, "No, we've only seen each other over video conferencing," which is interesting though. I didn't feel like it was so wild. Unless somebody is really tall, then occasionally they're like, "Whoa. I've only seen you sitting down," for example. But generally, you get a good sense. Now, I do think it is important for people to get together and all firms that are distributed like ours, and I know now there are quite a few firms that are highly distributed, and there are some real benefits to it, but you do have to work at it. You need to have...
Leadership is very important to drive that culture across the firm, regardless of whether you're in one place or not. It's a little harder to just pick it up from osmosis, so you have to be a little more intentional about it. But again, that's good leadership. And two, it is important from time to time that people get together, and they do meet in small groups or larger groups. And you probably had this experience too, if you meet somebody while you maybe spend two days at an off-site or something together, and you get to know them in the flesh, then when you go back to interacting over video conferencing, it's already a richer experience because you know them, you know what they're like. You have a more multi-dimensional sense of that person.
So you don't have to be together all the time, but you have to spend some time with each other, you have to get to know each other, and you will take that back and that will live on. That's like an ongoing asset, so to speak. There are certain things it's great, like off-sites have always been. Even just all being in the office isn't super productive, like let's all go somewhere, let's all lock ourselves in a room away from all the distractions of the office and just focus on a thing, coming up with a new strategy or designing a product, or whatever it is, or we're doing a bunch of exercises to just work together better as a team. And so I think all of those are still really important.
Clint Betts
Yeah. How do you think about the economic outlook for the rest of the year? How's the software doing? How's tech doing? SaaS? What's your sense of things?
Philip Brittan
So I think I'd say uncertainty is the watchword. The first part of the year, we saw it and the whole industry saw it, just tremendous economic uncertainty, and I think potentially some governmental policies around the economic direction of the country and of the world, frankly, to some extent, have people unsure about where things are going. It could be amazing, it could be terrible, it could be something in between. Not known. I think as time has gone on, we've seen, and others, again... We see it ourselves, talking to our customers, and we see it in the headlines, and talking to other firms, other enterprise SaaS companies, that some of that uncertainty, at least some of the fear aspect of the uncertainty, has gone down. So I think there's still uncertainty, but people are a little less trepidatious about the uncertainty. They're like, "Okay, we can roll with the uncertainty." I feel like that has definitely happened.
And so I think the outlook for the back half of the year has improved generally, and I think a lot of it has to do with this, okay, people are just feeling okay about rolling with the uncertainty, and the sky isn't falling, and we can deal with things. I do think there are other areas of uncertainty. So there's sort of general economic uncertainty. I think there's a lot of, again, prognostication and some of it's fairly melodramatic, some of it incendiary about us, the end of SaaS, but I don't see that at all. Certainly, it's not the end of enterprise technology. And things take different forms, things evolve over time. I haven't seen any reason why SaaS, which is a very broad category, really changes in the short term, and certainly, the fundamental need isn't there for something really different. Customer expectations constantly evolve, enabling technologies to constantly evolve and allow for new things, like AI.
But again, back to our earlier part of the conversation, I don't think people really fully understand yet how AI exactly is going to help their business. Yes, here and there, and a lot of firms are using it for very specific things, but I think the larger scale cataclysmic kind of changes, not sure. They may not happen, they may take far longer to sort of happen. They may happen in a slower evolution than in a sort of cliffy way than some of the prognosticators have been prophesying. So we'll see. So I think that's another area of uncertainty, okay, what is not just the economic situation going to look like, but what's the future of software look like? How do we best serve our customers? And I think there are all kinds of forces going on at the moment, including different business models and different ways to look at things and to delight customers. And there's no real, not yet, a truly clear, confident direction. There's still a lot of exploration, a lot of speculation, and discussion and debate going on in the industry.
So those levels of uncertainty are in the background, but I don't think they're really affecting most people's... They may be affecting some, but I think most businesses are not yet really directly affected by that level of uncertainty, compared to the economic uncertainty, which was much sharper, but that again, seems to be receding a little, at least a bit. Hopefully it continues to... Hopefully, economic uncertainty continues to recede.
Clint Betts
Finally, we end every interview with the same question, and that is at ceo.com, we believe the chances one gives are just as important as the chances one takes. When you hear that, who gave you a chance to get you to where you are today?
Philip Brittan
I mean, so many people. So many people. I would say probably fundamentally, and it's probably a common answer, my family was extremely supportive, a loving family. Again, I'm just sheerly lucky. I can take no credit for it. And they're very supportive, and they taught me a lot. And growing up on a ranch taught me a lot about making things happen and being responsible. A ranch is basically a collection of little family businesses, and different parts of the family would do things. So my folks would have the cattle business and the horse breeding operation, my sister ran the egg chicken business, and I ran the pig business. We had all these different things. Obviously, we all worked together, but we had areas of responsibility that we were responsible for. My parents were very clear that we were responsible. We had to do the record-keeping financially. We paid for the cost of feed and made money from that business, and so we understood what running a little business felt like, and then learned all kinds of amazing things and had a great education.
My folks sent me off to Harvard and enabled that to happen at multiple levels, and I'm forever grateful. So I think fundamentally, I'd say my parents. You can say, well, helping your child, now it isn't really taking a chance, but you are always hoping for the best, and you're giving people a chance at success. So I think that's foundational. But I will add... I don't want to [inaudible 00:37:22], but I will add that so many people in my life, folks who've hired me, investors who've backed me, employees who've joined teams, they've all taken real risks, and they may not have known me very well at the time, but they backed me and allowed me to do something successful that allowed me to give back, but that allowed me to grow, and I'm forever grateful and I approach every day with gratitude about that.
Clint Betts
That's incredible. Philip, thank you so much for coming on. Congratulations on everything. Good luck up there in Montana. I'm sure we'll be talking to you about some of the things you're doing there since we're neighbors, and best of luck.
Philip Brittan
Absolutely. Thank you, Clint. Really appreciate it.
Clint Betts
Likewise. Thanks, my friend.
Edited for readability.