eGain Revisited

Enterprise AI has moved beyond experimentation. The challenge now is building systems that deliver answers companies can trust—especially in highly regulated industries where accuracy, governance, and compliance are nonnegotiable. 

In this episode, Gary and Scott welcome Evan Siegel of eGain back to the Macro AI Podcast. Drawing on his experience in financial services, customer experience, and large-scale contact center operations, Evan explains how organizations are moving from AI pilots toward practical, measurable deployment.  

The conversation explores eGain’s expanding focus on banking and healthcare, why enterprise knowledge has become foundational infrastructure for AI, and how companies can reduce hallucinations by connecting AI systems to accurate, governed, and continuously maintained information. 

They also discuss: 

• What has changed most in enterprise AI over the past year  
• The unique AI challenges facing banking and healthcare  
• Why knowledge architecture may matter more than the latest foundation model  
• How organizations can build accurate, explainable, and compliant AI systems  
• The business metrics that demonstrate real AI value  
• Whether enterprises will use one foundation model or orchestrate several  
• The most common mistakes companies make when beginning their AI journey  
• How AI agents could reshape customer service over the next three to five years  

For business and technology leaders, this episode provides a practical look at what it takes to move from AI enthusiasm to trusted, governed, and measurable execution. 

Featured guest: Evan Siegel, eGain 

Follow the Macro AI Podcast for practical conversations about artificial intelligence, enterprise technology, and the strategies business leaders need to understand what comes next. 

eGain

https://www.egain.com/

About your AI Guides

Gary Sloper

https://www.linkedin.com/in/gsloper/

Scott Bryan

https://www.linkedin.com/in/scottjbryan/

Macro AI Website: 

https://www.macroaipodcast.com/

Macro AI LinkedIn Page:  

https://www.linkedin.com/company/macro-ai-podcast/

Gary's Free AI Readiness Assessment:

https://macronetservices.com/events/the-comprehensive-guide-to-ai-readiness

Scott's Content & Blog

https://www.macronomics.ai/blog

Transcript

00:00
Welcome to the Macro AI Podcast,  where your expert guides Gary Sloper and Scott Bryan navigate the ever-evolving world of artificial intelligence.  Step into the future with us  as we uncover how AI is revolutionizing the global business landscape  from nimble startups to Fortune 500 giants.  Whether you're a seasoned executive,  an ambitious entrepreneur,

00:27
or simply eager to harness AI's potential,  we've got you covered.  Expect actionable insights,  conversations with industry trailblazers  and service providers,  and proven strategies to keep you ahead in a world being shaped rapidly by innovation.  Gary and Scott are here to decode the complexities of AI  and to bring forward ideas that can transform cutting-edge technology  into real-world business success.

00:57
So join us,  let's explore, learn  and lead together.  Welcome back to the Macro AI Podcast. I'm Gary Sloper joined as always by my cohost, Scott Bryan. And this week we're excited to welcome back Evan Siegel from E-gain.  Evan joined us last April on episode 13  and we've had a great conversation about customer experience,  financial services, knowledge management and how AI was beginning to shape the contact center.

01:25
But since that last episode, and since we spoke with Evan last, a lot has changed since then. Over the past year, as many of you are aware, AI has moved from experimentation to execution in the workplace.  Enterprises are no longer just asking whether AI can  answer questions. They're asking whether  the questions can be trusted and governed and connected to the right knowledge for the organization and  safely deployed in regulated industries like banking, insurance, and healthcare.

01:55
ah That is exactly where E-Gain has been focused for years, AI powered knowledge, customer engagement, and helping companies deliver accurate, consistent answers across agents, customers, and employees. ah Evan specifically brings valuable perspective to the conversation. If you remember our last episode, he has a deep experience in financial services, customer experience, and large scale contact center operations, including his time at Wells Fargo.

02:23
So today we want to catch up on what has changed, ah what enterprises are getting right and wrong, ah and how AI agents may reshape customer service over the next few years. So Evan, welcome back to the Macro AI podcast. Since we last spoke, uh what do you think has changed the most in how enterprises are approaching AI?

02:44
So Gary, first of all, thanks a lot for having me back.  going for the, I'm gonna date myself a little bit here. I'm going to be, I'm trying to be  the Don Rickles to your Johnny Carson, be that sort of frequent  guest on your show.  I'm a of Rickles.  What's changed? Well, first of all, let me kind of tell you a little bit what we've seen. ah Go back 12 months. ah What  you'd  see was in the early days, people would be like, hey,

03:13
AI is kind of magic. Let's uh just get a pilot going. We know we've got our knowledge, our policies and procedures are in multiple places. uh There's a lot of crap there, but hey, AI is kind of magic. It will sort it out. And they run a lot of pilots  and, know, uh spoiler alert, those pilots fail because the underlying knowledge  is, uh you  know, lot of errors in it, a lot of... uh

03:42
duplicates, et cetera, causes the AI to fail.  And so I think  one thing that has changed recently is people now realize that  unstructured data is critical to AI success. And let me give you a proof point. So  as you mentioned, I focus a lot on the banking vertical.  I was at an ABA conference three weeks ago. Every speaker, I kid you not, every speaker, whether they were a consultant or a bank on stage,

04:11
made a big point about, hey, unstructured data is getting that right and accurate is critical to AI success. A lot of the consulting firms, their whole pitch was we'll do an AI assessment particularly focused on unstructured data. So  I'd say that's a realization of that's changed in the marketplace. I think people know they can't just do an AI pilot and have  the  magic AI sorted out the  large language model.

04:41
But  what I'm also seeing  is the next hurdle that folks are underestimating. And that is the difficulty of scaling from a pilot to a full-scale implementation. know, in a pilot, you're going to have fewer use cases, you're going to have less knowledge powering it, less data. Then when you start to expand, you get a lot more complexity. And then if you haven't meticulously tended to your knowledge,

05:10
If you don't have all kinds of learning loops and reporting and analytics and data,  when you scale, it's going to blow up. And I think people sort of underestimate that step. Yeah. And all the edge use cases too. Yeah. think  when I was hoping to jump into kind of right out of the gate is I think you guys have some, since we last spoke over a year ago, I can't believe it's already been over a year.  You guys have come up with some new offerings kind of  focused on the banking and healthcare.

05:39
verticals. And I was just wanting to get into what problems were those solutions designed for?  were there areas that uh some of the other solutions that are out there weren't addressing that EGain was trying to focus in on? Yeah. So this is my sweet spot. I lead the banking vertical at EGain. So if I wax a little long-winded here,  excuse me, because I'm very passionate about this,  but there's a lot to say. So I'll be long-winded because there's a lot to say. So let's...

06:08
based on my banking experience, plus I talk to lot of banks as uh a representative of E-Gain, let me tell you, no surprise, what's on the mind of the C-suite. So what I'm going to do is you tell you what's on the mind of the C-suite, and then I'll tell you how  our banking suite  of  solutions addresses those priorities. Because that's where you want to be, right? You as a software provider want to be in that C-suite of priorities. So what they're all talking about is we want to acquire new customers.

06:37
for the customers that we have, we want to increase share of wallet. We want to do upsell, cross-sell. uh There's a lot of value to be had  by your customers. That's the  old Wells Fargo grade eight.  Of course, cost reduction is always there. uh We want to make sure the branch, which is where a lot of the sales conversations happen, is a great experience. And oh, by the way, retention is still really important. Once we have a customer, we can't lose them because of the lifetime value there. And oh, by the way, double asterisks,

07:07
we got to stay compliant all the way through, right? We're regulated, the government's on top of us, and  a compliance issue is uh a brand  destroying event. So let's kind of go through how eGainSolution solves those priorities. So our Knowledge Hub, single source of truth, is really all about enabling AI, enabling  cost reduction, et cetera. We'll get a lot more into that Knowledge Hub. That's our core product.

07:35
What we've done in the banking suite is we've added some functionality around some of our other capabilities. So on top of our  knowledge hub, have AI agents, both for self-service in the digital channel and the contact center. So  in the contact center, think of it as an agent assist. So let me give you a use case of where we've supplemented that to achieve that business goal of upsell, cross-sell. So we're going to walk the contact center agent through VR knowledge

08:04
govern, clear, no errors in it, no duplicates, et cetera, how to resolve an issue. Say this, do this, very deterministic error-proof guidance. What we can do, however, now is we can also consume data on customer profitability. So let's say it's a customer who's highly profitable who's calling up for a fee reversal,  super common situation in banking.  What we can do is say, hey, our policy

08:33
for fee reversal says, there's a table, it says, profitability above X, no questions asked fee reversal. So then we can guide the contact center agent to  do that, right?  And that's retention, because you don't want to mess with your highest profitable customers.  Then once the issue is resolved, we can do what every bank wants to do, a highly tailored next best product. Hey, Evan.

08:59
you qualify as one of our most valuable customers, you qualify for premier checking. So then you're executing that upsell cross-sell. It's right in the flow of what that contact center agent needs to do. Okay, so that's one enhanced capability we have. Now,  let's go to that acquire new customers, that branch experience. If you walk into any bank, if they're not an EGANG client, and you tell that branch banker or MSR in a credit union, hey,

09:28
I'm about to do a home remodel. What product should I have? Every bank executive will tell you most of their bankers will be like a deer in the headlights. That's a difficult conversation to have. These branch bankers are entry level. They don't know. They're not familiar with that many products.  We've built  a solution we call Sales Advisor, which uses AI to  guide a needs-based conversation.

09:55
Hey, Evan, what brought you into the branch? Depending on my answer, what's the next best question so I can suss out  all the unmet needs, not just what specifically I came in. Let's say I came in for a credit card. The questions will also suss out, um do I have credit cards elsewhere? Do I have balances on those credit cards elsewhere? That's additional sales for like, let's say the balance transfer. Then you can also ask, hey, what's your financial priority?  And

10:24
based on their answer, go deeper, assess additional needs around financial priority, then you're unearthing additional sales opportunities. That is the holy grail of growing revenue in banking, a really great needs-based capability. So let me just pause there, a couple more capabilities, but that's really some of the key core ones. So thoughts or reactions to those. And I have two more to say after that. Paul. Yeah, I was just going to say probably nine out of 10 branch representatives don't even think to ask those questions.

10:53
obviously to just tease up those upsell opportunities. It puts it right in there. We had one client benchmark in an AB, a very robust AB test. That particular client wanted to improve loan sales, which is harder, right? Lending products are a little more complex. They saw a statistically significant 30 % lift in credit card and lending sales, which is huge. That's the most profitable products. One more dimension. So we talked about compliance.

11:23
We're busy building out an ecosystem with a bunch of partners  and through our ecosystem, we are able in the compliance space, our partner  will tell any bank or credit union, what are all the federal, state and local rules you're governed by? They will monitor those. When one of those change, they will alert that institution and show them the before and after. How has the rule changed?

11:51
they will pass that data to eGain. eGain then, this is the part that without this type of an automated compliance updating system takes months. eGain will then  note  all the policies and procedures that tie back to that regulation. We use our AI to do that. And then we have other tools to uh create first drafts of a new policy and procedure to reflect and comply with that new regulation.

12:21
And then finally workflow to sort of make sure all the legal and compliance folks at E2 review it. And then it's pushed out to the front line. So your, your time to be exposed when you're not compliant with a new rule of regulation is dramatically reduced. And your omnichannel availability of this new rule is there because we're, walking people through whether in the digital channel or the contact center, step-by-step how to resolve an issue. That's really interesting.

12:51
Yeah, I was just going to say that kind of piggybacks on another question I had is it's really important in those industries to build trust with your clients. And banking and healthcare are really two of the most regulated industries in the world, really. And it sounds like you have a system in place to help kind of build that so that your system remains accurate, it's explainable, and it's compliant. Is that really the direction you're going to make sure that you meet those regulations in those industries? Yeah.

13:20
Absolutely. It uh kind of gets back to the point I made earlier where  knowledge  is the key infrastructure for AI. Same thing, right? Let's go through some of what we do for that. there's no silver bullet here. It's really the collection of everything that we do. ah So hundreds of best practices. I'll just give a few, um and then we can kind of go off from there. uh

13:49
So one is we have a capability called knowledge intelligence, where we're scanning  the knowledge base on a multiple times daily basis. We're looking for things like  duplicates, outdated information, does the information answer the most common questions, et cetera. And then  where it fails, we flag that for the knowledge team. We very much believe that you need to have a human in the loop to kind of keep this all running.

14:19
and the human will fix it. Let's give an example of what happens if you have duplicate knowledge. Let's say address change for a service person moving overseas, kind of a high risk transaction a bank might have. If one  hiding some corner of your knowledge base is some old policy that says you take steps ABC, then there's an updated policy that really the people want you to use that says you take steps XYZ.

14:46
AI can't sort that out. It just sees two policies with the same steps. It might mix and match those.  The people say, the bank executives say that's a hallucination. It's really a bad data problem. We call out those duplicates to the folks, the knowledge and  then fix it and then trust goes up. Right. And let me just state how hard this problem is. So, you know, there's sort of four

15:10
common things about data accuracy. There's the currency of it, how up to date is it,  the duplicates which I touched on, there's coverage and there's precision, how accurate is  the answer, information to answer the question. If all of those four dynamics, all those four benchmarks are 90 % accurate, that means you're only gonna have a 65 %...

15:34
on average accurate answer, because you're taking 90 % times 90 % times 90%, right? The chance for error is dramatic. And back to your original question, in the world where you have to have trust, you have to have accurate answers, 65 % is not acceptable in an accurate answer.

15:56
And let me say more. that's just, I talked about knowledge intelligence. We have a bunch more, a bunch more. I mean, could really fill up an hour talking about all these best practices. Let me just talk about a few more, because it's cool stuff and it's the level of detail you need to have. I know a lot of your listeners are always thinking about make-buy and why this is why it's hard to make a really great knowledge management system. So we also have reporting.

16:23
We will tell the team, the knowledge management team, what questions are not being  answered  so they can go in and figure out why and create um better answers for that. We will tell them which uh generative answers were  followed by another question, which means that generative answer was not great. We also have a tool that we use to tune AI agents, because even when you have great knowledge, you still need to

16:52
make sure your AI agent  is  interpreting the question right, pulling the right information,  leading the conversation in the right way. So we have a tool around that to tune your AI agents that would be on top of your knowledge, et cetera, et cetera. So just lots and lots of  best practices to make sure your knowledge is trustworthy, your AI answers are trustworthy, and you're in that virtuous circle of cost reduction and efficiency. Perfect.

17:22
So  you went over quite a bit there, which is great. you your organization at E-Gain, you guys have been building for a long time now and probably well versed around AI and kind of, you know, launching go to market  products. What do you think over the last 12 months that you've seen or folks at E-Gain have seen that a lot of enterprises are actually missing? You  know, I'm just curious on your part, because I feel like

17:51
you know, especially what you guys are doing is trying to make sure folks are compliant in that example and up to date. But I do see, and I know Scott sees this as well, a lot of organizations that are still missing the boat on a variety of topics and governance. So I'm curious what your take is, if you're seeing that at all. Yeah. So  one I already touched on, which I won't belabor, is the difficulty of scaling from pilot  to full expansion. Let me give another one. um And by the way,

18:21
I just wrote an article along with a Deloitte consultant on this topic. It is, you guys look to be roughly my age, maybe a little bit younger, but it's something that's near and dear to my heart. It's around baby boomers like myself retiring and all the knowledge that walks out the door when they retire. Think of a field service workforce where you're out in the field fixing equipment.

18:45
You talk to any company in that space and they'll say like, yeah, there's six guys that everyone has on speed dial because they know how to fix the most gnarly, difficult pieces of equipment. uh I actually saw an executive from a field service company at uh a conference and he said, that's the number one thing that keeps them up at night is 50 % of his workforce is retirement eligible. Huge problem, uh cuts across all industries  and

19:14
It's particularly critical for baby boomers because a baby boomer on average has 8.7 years of tenure at their given company. So they have not only domain expertise from their industry, they have company expertise. And that's not going to be replicated, right? We all know the younger generations due to a bunch of factors just don't have the same loyalty to companies. So we have built into our tool capabilities to automatically and easily capture that knowledge. You can have the

19:44
the uh experts sit down and put in an outline and our AI will sort of take that rough outline of how to fix a piece of equipment or how to resolve an issue and turn it into a policy or procedure  that aligns with best practices. We  can conduct an interview with them, verbal, you know, a recorded interview. And again, our AI tools will turn that into knowledge and the policies and procedures. So that's another area that we're seeing.  When I give this talk at conferences, uh

20:14
people line up, super relevant issue. And the article that we published along with Deloitte kind of goes through a broad framework of how do you address this and scales the problem for companies.  And I can share a link with you guys if you want to put that in the show notes.  I'm very proud of the article. Deloitte  publishing it as a sort of a  nice  acknowledgement of the thought leadership there. Absolutely. Yeah, we'll definitely get a  link in the show notes. If you could send that over afterwards, it'd be great.

20:43
Yeah, I think why don't we kind of shift gears into ROI is a big question that comes up, especially when we're talking to clients and they're thinking about something new, taking a big step in the world of AI. so, you know, what do you, when your customers roll out an E-gain solution, maybe you could tell us a little bit about, you know, how do you explain the ROI and what KPI improvements that they're going to start seeing, you know, right out of the gate, maybe six months, nine months, or how would you explain that? So I'm going to

21:13
I'm going to, you know, as a podcast, can be a little attention grabbing. I'm going uh to boldly claim that with  a knowledge led transformation, many organizations can realize in the customer service realm up to a 75 % cost reduction. Now it depends on their baseline and where they're starting from. And let me break that down.

21:38
By the way, the McKinsey Institute, maybe 18 months ago, two years ago was  saying in customer service, 40 % is very realistic, but we are seeing organizations 60, 70 % cost reduction. So what are the elements? First of all, call deflection. When you've got really good AI agents connected to your CRM system or your bank or credit union, a core, and you can sort of integrate with the relationship and transactions, you can power

22:07
really great call deflection,  up to 45, 50, 60%. That's a huge cost save ah because  our knowledge capability also comes with agent assist capabilities. uh Your time to competency in your call center,  which means you're performing and agents performing like a wizened veteran, you know, shorter. The amount of time you need to do for training is lessened up to 50 % because you don't have to train the folks on

22:36
you know, how certain policies work, they're trained in the moment. We step them through exactly what to do or say. As the result of that same sort of say this, do this step-by-step guidance, we see, you know, 15, 20 % improvement in handle time. Even as issues get harder, because the AI agents and the digital channel are taking all the easy issues, we'll see a 30, 35 % improvement in first contact resolution, so you don't get those callbacks which drive your costs.

23:07
You'll see some impact and reduced turnover.  Gartner published a study a while ago that talked about, know, contact center agents. One of the reasons they hate their job is they feel like they don't have great tools. You know, average contact center agent has like seven windows open. Well, in this case, we're typically a window within whatever tool they're in most of its telephony system or Salesforce, et cetera. So their number of windows they need to have open goes down. They like their jobs more.

23:34
And by the way, you'll get all of these savings, handle time, oh FCR, call deflection, while also getting  Net Promoter Score improvement. People want their issue resolved once. They wanna resolve it in the channel of their preference. It's  debatable, but some people think younger generations prefer non-human channels, by enhancing your AI agent in digital channel capabilities, you'll give people the flexibility to go to the channel they  best want.

24:03
Those are some of the key areas that we see cost reduction and when you add it all up over a couple of years along with things like change management, process redesign, right? It's all not just the knowledge management. You need some ancillary business change around it. You can get up to 75 % cost reduction.

24:22
Yeah. Amazing. Yeah, that is amazing. Especially in this  market of trying to improve NPS. Like you said, right? There's a, there's been a huge push and customer experience over the last several years with SAS companies. And I think for a long time, it was just throwing bodies at it. Now, if you have some of that automation, you reduce, you know, an agent, like you said, that amount of windows probably has less processing power on their laptops. So probably, you know, seven windows.

24:51
pulling information,  it's probably slow, so that their experience is slow that goes back to the customer. So that's really impactful.  And I think where  the whole industry is headed,  I think outside of just the verticals we're talking about, everyone's starting to look at that from an agentic standpoint. Like how do I get out of  the mind frame of,

25:14
just push one for directions to our store, push two to speak to a customer representative. And you're yelling in the phone, it still doesn't recognize you. So I think the companies that are grasping exactly what you're talking about today, they obviously leapfrog everybody else.  And  the agents who are there have a much higher job satisfaction level because they've got that specialist knowledge intelligence layer that just kind of propels them up and makes them feel.

25:43
like they have ownership of that role. Yeah. And I can give you a data point, on that. Rogue Credit Union, who's kind of a great account for us. They really use us enterprise wide. After eGain was implemented in the contact center, they saw a 22 point lift in agent satisfaction for that very reason, right? Hard data, they're on stage and in webinars talking about that. So that's why I can share that. Yeah. Makes sense. Nobody wants to do the easy stuff all day. Yeah, exactly.

26:13
Um, so, so thinking about this, mean, obviously it kind of take a step back. mean, do you think enterprises will standardize on one foundation model or will, you know, kind of the future involve orchestrating multiple models. You know, depending upon the workload, obviously, but that could have an impact exactly what we're talking about here. If, they're not standardizing on a particular foundation model, um, or kind of, you know,

26:41
spreading  across the ether,  that could make any technology difficult. So I'm not sure what you're seeing out there. I'd be curious if you had any take on that. Yeah, so I have a take. I'm sort of going back to business principles and business  foundational thinking here. ah This is a new industry,  the large language model industry. It's a new industry.  I predict, and I'm not the only one predicting this, by the way, but I really think what's going to happen is, of course, you're going to get differentiation.

27:10
when there's this much capital out there  creating new companies, people are gonna try to uh differentiate and get some market share. So you already see that there's the leaders with quality and sophistication and range of functions performed. In the China model like Deepsea, you got low cost and you're already seeing companies kind of float between those two, some of the big boys in the US  and the Chinese models. But then you've also got models that are differentiating around modality.

27:39
So is it text, image and video, their ability to handle those. And then  also ones that are fine tune, uh detailing and differentiating around fine tuning and customization. So I think this is gonna continue  and you're already seeing companies sort of interchanging among those for different use cases or even across the company. It's a way to improve service, reduce costs.  And uh if you sort of look at other industries over the...

28:08
over time, probably what will happen is some of these differentiators that take hold will get gobbled up by the big guys, and then it just becomes sort of a subproduct within their umbrella. But the end result for the companies that are consuming these models  is that they'll have the flexibility to sort of manage their costs, flip among the models, and all of this is an argument, by the way, for a composable architecture in your solutions that use the large language models. That's what we are.

28:38
to sort of explain that in layman's terms. You can kind of plug and play different capabilities of E-gain into your tech stack. Large language model is one of those. We're sort of indifferent to which large language model our clients use. They can change once they've implemented us. And that's to reflect this likely changeability in the future of the evolution of large language models. Yeah, interesting. Yeah, and then if you think kind of

29:08
more specifically about the customer experience. A lot of our listeners are trying to kind of get that futuristic view of what's going to happen with customer service, what's going to happen with the actual agents that sit in the seats. And since we talked last year, a lot has changed in just one year. What do you think's going to happen over the next five years in the customer experience or customer support? Right. So I live and work in the Bay Area and

29:37
autonomous vehicles are everywhere. I'm not sure where they are. We are listener cities. if you think about- so much up in New Hampshire, Yeah, no, they're everywhere. And think about the challenge there, right? And every time one crashes is great news, but their safety record is much better than regular car drivers. I think the analogy holds in that I truly believe that the future five years from now, AI agents without

30:06
much human supervision will be resolving a lot of issues. 70, 80, 80%. The contact center folks, the humans will be left with the corner cases, the hardest ones, the escalations.  And I think that's the future we're moving toward. I think it's an exciting future. uh I think it's one where companies can then take their savings there and reinvest it in innovation and other uh enhancements.

30:34
things like the  meeting the needs of the customers with next best product, et cetera. it's,  I'm not one of these doom and gloom that it means job elimination. I think it's an exciting future where companies can invest in better meeting the needs of their customers uh in other ways oh than simply answering questions and resolving issues. Yeah, totally agree with you. If you're a specialist in a sport, for example, we talked about skiing on the last podcast.

31:00
you're really specialized in something and that's what you want to get on the phone or on a video and talk to a customer about. You don't want to be answering all those simple questions. So yeah, I totally agree with you. Yeah, and I would add, um you know, as part of that, because  to your point where this is moving and we'll have positive impacts,  I still see and curious what your take is here as well. uh

31:27
a lot of mistakes that organizations  are making with their AI journey. And a lot of it could be they have, you know, they feel that they have great data,  years of data, but it's poor data.  They have lack of governance, uh really no business objectives, and they're trying to automate everything ah immediately without a real plan, like, you know, kind of building that center of excellence. And I'm curious if, I'm sure you and your colleagues at

31:55
EGain run into this quite often. And I think that that's going to hamper a lot of organizations to really get to what you were just mentioning. Yeah,  I would say I touched on it briefly. I think the make-buy decision is probably one of the biggest mistakes folks are making. They look at it and they say, oh, how hard can it be to sort of sort out our knowledge?  These  large language models and these AI agents are so powerful. uh

32:24
We can do that. I actually worked with  a credit union here locally and  hats off to them. They diligently went through and rewrote every policy and procedure and then they built this little simple tool. The CIO likes to sort of build stuff themselves and like, yeah, it's working. But fast forward three, four years, rate changes, uh new issues. uh

32:48
It's hard to keep that up to date without the best practices. I went at that ABA conference, one executive from a large regional bank said that  to get the knowledge right for their AI implementation, he used a great term. He said  it was hand-to-hand combat.  And that was just to get it right in a snapshot in time.  But again, you have this issue of new regs, new issues, uh and so it's hard to keep it right. uh

33:17
People will try to make their own tools. They'll try to use uh big enterprise solutions that have knowledge modules that are okay or good enough. But I will tell you, in the age of AI, in the industries where you have to get it right, where you have to get trust, good enough is not sufficient. You need to have someone who's best of breed. This is all they do. Because uh once trust is broken, you guys kind of alluded to that, it's bad.  Customers lose, yeah, customer you lose brand.

33:47
brand equity, team members don't trust new tools, it's bad.

33:52
Yep, totally agree.  think, you know, it's been a year since we had you on already. If you're, if we're thinking about maybe two years from now,  and we try to do some predictions every once in a while. If we, if we talk again in two years, what, what do you think would be one prediction you could make about AI for enterprise that, that you think might happen? So everyone today is talking about cost reduction. I think in two years, it's going to be much more around revenue growth.

34:22
And it's touching on, and it's interesting to me. It's not been an area of a lot of talk and focus, but even with what we're creating, it's a huge untapped area, upsell, cross-sell, next best product, uh needs-based selling. And I gave a banking example, but I'm a big backpacker and skier. um

34:43
when you buy skis, you know, how old it wouldn't it be great if the AI agent asked you how old you are, what your your ski level is, what mountains you like to ski, what type of snow and then recommended products that's needs based selling. So AI can do all that great E and do that now. I think that's going to be the next frontier. And, you know, where CEOs get excited is revenue and growth, cost reduction sort of table stakes, it's revenue and growth that people get excited about. And I think AI will be there at that table.

35:09
Yeah, it'll, it'll drive productivity. We've talked about that a few times on some recent shows, but it's, it's definitely going to be a wave of productivity by focusing on the things that can grow revenue for sure. Right. Yep. Absolutely.  Um, so Evan, this has been a great conversation. Really appreciate you coming back on the, macro AI podcasts. Any closing thoughts or anything else you want to cover, uh, before we let the listeners go.

35:35
Well, first of all, just want to thank you. I love our conversation. You guys ask great questions. And I do want to plug an EGain conference. It's free to attend. So we do our Solve conference in Chicago. Again, I'll send you a link so you can include it in the show notes to register. And it's two days of content like this. It includes our CEO talking about where he sees the industry going. It will include our clients getting on stage and talking about their journeys and their results.

36:04
what they learned from those journeys. And it'll include workshops around things like banking and healthcare where we have new solutions with deeper capabilities. And again, free to attend. It's at the airport in O'Hare. So those that are out of town can jet in and jet out without too much travel. So it's a great event. We, people who attend it rate it very highly as really thought provoking and a treasure trove of best practices. No, that's great to hear. mean, obviously, you're on the forefront of this with your organization, Enterprise AIs.

36:34
clearly moving from hype to practical execution. So I love hearing that events like this have been scheduled and have been a success in the past and will continue to do so. So we will definitely get that out to our listeners, especially if it's free, especially if folks are looking to travel and come in. We'd love to have you guys there. You could maybe even do show from the shop floor. show floor. not a idea. Put a creative idea in your heads. Make a trip. Absolutely.

37:01
So really appreciate everything today. I think our topic ah was very relevant. ah It's especially important in regulated industries like banking and healthcare, ah really where accuracy, compliance and customer experience matter. So I want to thank you again, Evan. Thanks for taking time out of your super busy schedule  to share your experience.  And to our listeners, we will add Evan's LinkedIn profile in the show notes again.

37:28
Feel free to connect with him. If you have any other questions around E-gain, please reach out to him,  myself or Scott.  And thank you for following the Macro AI Podcast. Please share this episode and join us next time.  Thank you so much,  Bye bye.