Latest / Sidecar Sync / AI-Enhanced Member Services: What You Need to Know | 73

Transcript

AI-Enhanced Member Services: What You Need to Know | 73 · Sidecar Sync

• 0:00
  Amith: It's interesting because associations tend to have this pattern of infrequent but deep engagement.
• 0:06
  Amith: Let me put it another way People come to your event, but then they don't think about you the rest of the year.
• 0:11
  Amith: So think not just about what you can do with AI that you currently do but things that you could not do were it not for AI to begin with.
• 0:21
  Amith: Welcome to Sidecar Sync, your weekly dose of innovation.
• 0:27
  Amith: If you're looking for the latest news, insights and developments in the association world, especially those driven by artificial intelligence, you're in the right place.
• 0:36
  Amith: We cut through the noise to bring you the most relevant updates, with a keen focus on how AI and other emerging technologies are shaping the future.
• 0:44
  Amith: No fluff, just facts and informed discussions.
• 0:47
  Amith: I'm Amit Nagarajan, chairman of Blue Cypress, and I'm your host.
• 0:51
  Amith: Greetings everybody and welcome to the Sidecar Sync, your home for association content focused on the emerging, exciting and crazy world of artificial intelligence.
• 1:04
  Amith: My name is Amit Nagarajan.
• 1:07
  Mallory: And my name is Mallory Mejias.
• 1:09
  Amith: And we're your hosts, and today we are going to be kicking off part one of a special two-part episode all about AI-enhanced member services.
• 1:19
  Amith: This is content we think you'll find useful now and probably well into the future, because the concepts really are about how you can bring member services into the present with AI and build for the future, and we're breaking it down into two different parts.
• 1:35
  Amith: Before we get into part one of this series, let's take a moment to hear a quick word from our sponsor.
• 1:42
  Mallory: If you're listening to this podcast right now, you're already thinking differently about AI than many of your peers, don't you wish there was a way to showcase your commitment to innovation and learning?
• 1:52
  Mallory: The Association AI Professional, or AAIP, certification is exactly that.
• 1:58
  Mallory: The AAIP certification is awarded to those who have achieved outstanding theoretical and practical AI knowledge.
• 2:05
  Mallory: As it pertains to associations, earning your AAIP certification proves that you're at the forefront of AI in your organization and in the greater association space, giving you a competitive edge in an increasingly AI-driven job market.
• 2:20
  Mallory: Join the growing group of professionals who've earned their AAIP certification and secure your professional future by heading to learnsidecarai.
• 2:28
  Mallory: Amit, how are you doing today?
• 2:34
  Amith: I'm having a great day so far.
• 2:35
  Amith: It's beautiful here in New Orleans today.
• 2:37
  Amith: I'm back Just got back yesterday from a week-long ski trip, and that always puts me in a good mood.
• 2:42
  Amith: Of course, I've got the usual catch-up game to play with being out for a little bit, but I'm doing great.
• 2:48
  Amith: How about yourself?
• 2:50
  Mallory: I am doing pretty well myself.
• 2:52
  Mallory: Unfortunately not beautiful weather here in Atlanta, but hopefully the rest of the week we'll get some sunlight out and less of this rainy cold.
• 2:59
  Mallory: But I've been pretty excited to do this AI Enenhanced member services episode because I feel like it's been a minute since we've done a more evergreen type episode.
• 3:09
  Amith: Indeed it has, and this topic, I believe, is going to be of value for people now and for years into the future.
• 3:16
  Amith: Certainly, the AI techniques, the AI tools that you'll use will vary and they'll certainly improve over time, but the idea of improving the fundamental customer service function of an association is so key.
• 3:28
  Amith: Many association folks in my experience get excited thinking about how AI can save them time and improve their internal work, and that's, of course, critically important.
• 3:37
  Amith: But if you can also fundamentally improve the way you serve your members, the way you engage with that audience, that's incredibly interesting, right, because it potentially levels up the quality of the experience, that can help improve member retention, that can attract new members and ultimately just creates more value all around.
• 3:56
  Amith: So that's what these two episodes are, and to me it's a very exciting, evergreen topic.
• 4:03
  Mallory: Yep Excited to kick that off.
• 4:04
  Mallory: I know you mentioned you just got back from some travel, Amit, but I think you have some more travel coming up right, Because we've got those DC and Chicago innovation hubs coming up.
• 4:14
  Amith: I have far too much travel going on this spring.
• 4:16
  Amith: So, yes, I'm bouncing around a lot.
• 4:18
  Amith: I'll be in DC, I'll be in Vegas, I'll be in a couple of other places over the course of the next, I think, four or five weeks.
• 4:25
  Amith: So it's going to be busy.
• 4:27
  Amith: So it's fun.
• 4:28
  Amith: I enjoy getting out there and meeting with people too, but it's a lot right now.
• 4:33
  Mallory: And remind me the DC one is.
• 4:35
  Mallory: It's this month, it's March 25th, 23rd, that's correct.
• 4:38
  Amith: Yeah, so Blue Cypress and Sidecar have an innovation hub in Washington DC on March 25th.
• 4:45
  Amith: It's a full day event and that event is going to be fantastic.
• 4:49
  Amith: We're going to cover topics really all around innovation and at the moment, a lot of that innovation talk, of course, is AI related, and after we wrap up in DC, two weeks after exactly, we will be in Chicago on April 8th doing the same thing.
• 5:05
  Amith: So we have DC Innovation Hub on March 25th and the Chicago Innovation Hub on April 8th, and you're not going to want to miss these events.
• 5:12
  Amith: They're in person, but just the day, and it's an informal gathering of people who are really interested in driving their associations forward with all kinds of innovation.
• 5:21
  Amith: Again, ai is the center of all topics.
• 5:23
  Amith: It seems to be across the whole market right now, but it's going to be really exciting to just talk about a variety of different practitioners sharing their practices and what they've been able to innovate.
• 5:33
  Mallory: Yep.
• 5:33
  Mallory: So if you all are looking for some fantastic content to consume in March and April, I highly encourage that you check those out, and maybe you'll even get to meet Amith while you're there, so it's a win-win situation, all right.
• 5:46
  Mallory: As Amith mentioned, we are talking about AI-enhanced member services today.
• 5:51
  Mallory: In this part one episode, we're going to kind of set the stage for why we believe this is critical right now.
• 5:55
  Mallory: We're going to break down the responsibilities within the member services function and discuss ways that AI can augment each of those member services function and discuss ways that AI can augment each of those.
• 6:08
  Mallory: In our part two episode, we'll focus more on implementation, on actually getting started, how to turn those ideas into action.
• 6:13
  Mallory: We'll also be talking about the change management piece and kind of having that uncomfortable conversation around AI and job displacement that we tend to avoid.
• 6:21
  Mallory: So, first and foremost, we know we're exploring what we believe is a critical transformation opportunity for associations, which is modernizing your member services with artificial intelligence.
• 6:33
  Mallory: To prepare for this episode, we used Google's deep research to analyze hundreds of member services' job descriptions across associations.
• 6:41
  Mallory: So we're going to break down primary responsibilities and duties from those job descriptions, task by task, and discuss ways that AI can augment or transform them entirely.
• 6:52
  Mallory: I think we're all in agreement that member services has transformed from handling basic admin tasks into a strategic function that directly impacts the way your members engage with you.
• 7:03
  Mallory: And retention that directly impacts the way your members engage with you.
• 7:06
  Mallory: And retention this is your organization's frontline, your direct connection to members and primary brand touchpoint.
• 7:11
  Mallory: The challenge associations face today is this growing experience gap.
• 7:15
  Mallory: So members aren't comparing your service to another association's.
• 7:19
  Mallory: They're comparing it to big brands they're interacting with daily that offer things like 24-7 availability without waiting, immediate, accurate responses, proactive rather than reactive service, hyper-personalized experiences and seamless self-service options.
• 7:35
  Mallory: The gap, yes, is frustrating for your members and it directly impacts retention, satisfaction and perceived membership value.
• 7:43
  Mallory: We believe AI is offering a unique opportunity to not only close this gap but potentially leapfrog traditional consumer experiences while reducing costs.
• 7:53
  Mallory: So, amit, you have been talking about friction since I met you, so I know it's at least three years.
• 8:00
  Mallory: I think we can agree.
• 8:01
  Mallory: It's always been relevant, but why do you feel kind of this sense of urgency now?
• 8:08
  Amith: One of the ways to think about business is that often people come for the product but they stay for the service.
• 8:15
  Amith: So they wouldn't necessarily come to you if they didn't need your product.
• 8:20
  Amith: That doesn't make any sense, right?
• 8:21
  Amith: But oftentimes retention the why people stick around is because they value the service.
• 8:27
  Amith: That's true in the software business.
• 8:28
  Amith: In my experience across a lot of software companies, you tend to have really high customer retention when you provide a great, empathetic, high quality, responsive and really value additive service.
• 8:40
  Amith: Function and the product itself, of course, remains important.
• 8:44
  Amith: If your product is no longer valuable, that's an issue, but you can have the best product in the world.
• 8:49
  Amith: But if your service is terrible or if your service is simply non-responsive, that's a problem.
• 8:54
  Amith: So to me, that is one core principle that I do think associations completely get.
• 9:00
  Amith: But I also think that people have become accustomed to their challenges as an association, their choke points internally, the number of member services.
• 9:09
  Amith: Personnel is the number one thing, right?
• 9:11
  Amith: How many people can you afford to have employed by your association, on a full-time basis or on a part-time basis, who can help your members when they have questions, right?
• 9:21
  Amith: So that's the issue is that people the members coming in looking for help, have a limited number of people who can help them, and so sometimes you know organizations, you hear them talking about a goal of having a 24-hour response time as an example, and that's their goal.
• 9:38
  Amith: Oftentimes that's not reached.
• 9:40
  Amith: So it's typical in many associations to have a multi-day wait time to get feedback for member services.
• 9:48
  Amith: Obviously, there's exceptions, there's some organizations who are able to be far more responsive than that, but that's an issue, right, because having to wait a day, even if that's considered the best case, or two days or three days, that's not really acceptable in the world of 2025.
• 11:54
  Amith: That's a big problem.
• 11:56
  Amith: And then sometimes, other parts of friction are even with the so-called self-service aspects of member services and, by the way, member services, you could replace the word member with customer or vendor or volunteer or event.
• 12:09
  Amith: It's service, and so having great service matters across all dimensions of your brand, persona, all aspects of your business.
• 12:18
  Amith: But let's talk about that self-service concept.
• 12:21
  Amith: Associations often will have a website these days, of course, and those websites will allow people to, let's say, register for a webinar or perhaps renew a membership.
• 12:31
  Amith: But oftentimes the amount of friction there is so incredibly high it's almost comical, you know, to register for a webinar with some association websites might take seven or 12 or 15 steps instead of simply like putting in your email address and hitting register, and so this idea of user experience, customer experience is unfortunately a lower priority in practicality than it should be, and so my point would really simply be put this way People are not going to put up with this anymore.
• 13:03
  Amith: They live in a world where things are instant and really good.
• 13:06
  Amith: So you know why would I put up with that when I have alternatives?
• 13:10
  Amith: In a world where you have a monopoly in your particular domain, maybe you can get away with those kinds of things, but that's not true in a competitive ecosystem.
• 13:20
  Mallory: That makes sense.
• 13:21
  Mallory: Amit, I want to take the counter position here to you and kind of to myself, with what I said when I was introing this part of the episode.
• 13:28
  Mallory: When I'm interacting with big brands, right, I do expect minimal friction.
• 13:33
  Mallory: I do expect near perfection.
• 13:35
  Mallory: As an example, I'm a big fan of Whole Foods delivery through Amazon.
• 13:39
  Mallory: I hate going to the grocery store, so that's something I use all the time.
• 13:42
  Mallory: Just placed an order before this episode.
• 13:44
  Mallory: The service is quick, and the only downside with delivery groceries sometimes you get produce that's not great or spoiled, but their refund process is so efficient.
• 13:54
  Mallory: I just I don't even interact with a person, right.
• 13:56
  Mallory: I say I didn't like this product, they automatically refund me.
• 13:59
  Mallory: On the flip side of that, though, if I were calling up my local DMV or OMV and asking a renewal question about my car, I'm going to expect pain.
• 14:10
  Mallory: I'm going to expect long wait times, multiple transfers, all the terrible things, right, that come with customer service, so I don't hold the DMV to the same standard that I do Amazon, for example.
• 14:20
  Mallory: Do you feel like there's an argument there for members of associations?
• 14:25
  Amith: Well, I mean, it kind of goes back to the question of what are your choices and you know if you live in the state of Georgia.
• 14:31
  Amith: You have one DMV you get to go to.
• 14:33
  Amith: You know if you live in the state of Georgia.
• 14:34
  Mallory: You have one DMV you get to go to.
• 14:35
  Amith: You really can't, unless you just want to not have a license or not register your car, which some people may choose to do that.
• 14:40
  Amith: But the reality is is that you don't have choice.
• 14:43
  Amith: They have a monopoly on that service category.
• 14:45
  Amith: Perhaps another category that is worth discussing, kind of in the same vein though, of what you're describing, is airlines.
• 14:51
  Amith: Very few people would simply say that airline customer service is fantastic.
• 14:57
  Amith: It's pretty rare.
• 14:58
  Amith: I mean, southwest used to be, kind of, you know, the solo player in the market that people did say positive things about customer service, and unfortunately that seems to have declined.
• 15:06
  Amith: I'm not a big Southwest flyer, but I just hear people generally say they've gone downhill in that area over the last 10, 15 years.
• 15:15
  Amith: But my point there would be that there is a semi-competitive space.
• 15:17
  Amith: Right, because there are not a large number of airlines, but there's enough of them and there's enough choice, particularly for you living in Atlanta, where it's a major hub and there's a lot of choice.
• 15:25
  Amith: A little bit less so here in New Orleans, but you kind of do put up with bad service there too, because there is a little bit of friction in switching carriers, particularly, let's say, if you have frequent flyer points with Delta or with United or one of these other ones, you tend to put up with a lot of pain because there's kind of what we'd refer to as a switching cost, as both a power from a strategic lens and also as a pain point from a customer viewpoint.
• 15:52
  Amith: So I think that there's kind of like this inverse correlation, if you will, between the willingness to accept friction and people's long-term desire to stay with the brand innately.
• 16:04
  Amith: Ultimately, people do want to have a low friction, high quality experience if they can get it without having to undergo enormous switching costs.
• 16:12
  Amith: So for you, the switching costs other than being an unlicensed, unregistered driver which I'm sure you're not, mallory, however, would be to move states, right, you'd have to leave the state of Georgia.
• 16:23
  Amith: That's a very high switching cost.
• 16:26
  Amith: Now, in the context of associations, of course, the switching costs are.
• 16:30
  Amith: Historically they have been high.
• 16:31
  Amith: If you're an association in a particular domain or profession.
• 16:35
  Amith: Historically there haven't been that many choices in your exact category and perhaps your category plus your geography combined.
• 16:45
  Amith: So a state association in nursing or a international association in another country in architecture, right, there tends to be like one of those things in each of those kind of verticals plus geographies, but these days geographic borders have kind of melted away.
• 17:03
  Amith: Content is ubiquitous and valuable across borders, and it's not just associations that are in the content game.
• 17:10
  Amith: Tons of other players are out there who may not have content that is 100% as good as your peer-reviewed fantastic journal, right, but it might be good enough.
• 17:21
  Amith: And so this friction problem is a real problem because there are alternatives.
• 17:26
  Amith: I think associations first started to feel this pain with Google when search engines became free and ubiquitous.
• 17:33
  Amith: You know, 20 years ago it was a problem for associations initially because people could find stuff, but it wasn't that big of a problem because the stuff that people were finding wasn't really all that great.
• 17:45
  Amith: But with AI, the answers people are getting are truly outstanding.
• 17:49
  Amith: And we were just talking about Cloud 3.7 right before this call right, and you could probably go ask Cloud 3.7 quite a number of very complex, highly specialized questions in a lot of domains and get a good answer.
• 18:01
  Amith: And that's true, of course, of deep seek, it's true of open AI's latest stuff.
• 18:05
  Amith: It's true for a lot of these AIs that have been so broadly trained on so much outstanding content and, by the way, synthetic data, which is a whole other conversation that actually these general purpose AI tools are competitive and they're displacing for associations core services.
• 18:23
  Amith: So that's an issue.
• 18:24
  Amith: And if you can get an answer from a generic AI model on your phone or on a website in seconds and it's either free or extremely cheap and to get the same answer from your association requires several forms of backflips, you're not going to go to that association for very long.
• 18:42
  Mallory: Well, with that, I want to move into this task by task breakdown.
• 18:46
  Mallory: As I mentioned at the top of the episode, we are using a Google Deep Research report and we're going to mention, for each of these tasks, the current state, what we believe is the AI opportunity, the potential impact and then a more practical example for you to wrap your head around For each task as well.
• 19:04
  Mallory: Amit, I'm going to ask you some clarifying questions and I also want to get your take on the difficulty to implement it.
• 19:11
  Mallory: So low, medium high.
• 19:12
  Mallory: And then also the perceived impact low, medium high.
• 19:15
  Mallory: First and foremost, man we can probably all relate to this in some capacity.
• 19:20
  Mallory: But responding to inquiries and, as Amit said, we're focusing on member inquiries, but this could be vendors, various other groups, volunteers as well the current state you have staff manually handling emails and calls during business hours with varying response times.
• 19:36
  Mallory: Potential AI opportunity here is something like a 24-7 intelligent chatbot and automated email responses that can understand complex questions.
• 19:45
  Mallory: The impact of this, I think we can all imagine, would be huge.
• 19:49
  Mallory: So significantly reduced response times, consistent quality and the ability for AI to detect member sentiment, which could be important for various kind of downstream things.
• 20:00
  Mallory: And then, finally, a practical example maybe.
• 20:03
  Mallory: This is an AI-powered member service agent that can handle inquiries across all channels, including questions about certification requirements, upcoming events, membership benefits, at any time of day.
• 20:15
  Mallory: So that is our first task responding to member inquiries.
• 20:19
  Mallory: In theory, amit, this sounds great, sounds easy, sounds perfect.
• 20:22
  Mallory: But in actuality, I certainly understand that email inquiries can be more difficult to resolve than you think.
• 20:30
  Mallory: From the Sidecar perspective, we do have people reaching out to Sidecar on occasion, asking kind of straightforward questions like tell me about the AIP certification or tell me about the AI learning hub for members offering you have.
• 20:41
  Mallory: But oftentimes, when customers reach out to us, it can be a little complicated, like can you issue me this certificate for this virtual event that happened two years ago?
• 20:52
  Mallory: Right, it takes multiple steps.
• 20:53
  Mallory: It's not a quick one and, done so, do you see this task as something that can really be feasible for associations in terms of AI enabling it?
• 21:05
  Amith: Well, I think so, because what you're trying to do is break down the parts that AI can do well and the parts that AI maybe not, is maybe not best suited for, at least at the moment.
• 21:14
  Amith: And so, if you kind of break it down, I like this idea of you know, build a pie chart in your mind of the distribution of the tasks that are complex versus not complex.
• 21:25
  Amith: Or perhaps another way to break it down is AI suitable versus not suitable for AI?
• 21:30
  Amith: And what I think you'd find is that certainly FAQ kind of inquiries will be not only handled well by AI probably handled better than by a human, because the AI doesn't get tired of responding to the same question over and over again, whereas humans do tend to get tired of that kind of thing, and maybe they're copying and pasting the response, but it tends to get difficult.
• 21:50
  Amith: So that kind of thing is certainly easy for AI to handle.
• 21:54
  Amith: I think that quite a few of the transactional things like can you swap out my guest registration from person A to person B, right Very common kinds of things that you get, particularly leading up to large events.
• 22:05
  Amith: It tends to be an overwhelming volume of both FAQ style inquiries but also minor kind of incremental transactions that people are looking to do as you get close to a major event.
• 22:16
  Amith: Those are things AI can handle beautifully already, and the thing to remember we say this a lot, mallory, on this pod is that the AI you're using today is the worst.
• 22:24
  Amith: AI, you're ever going to get right.
• 22:27
  Amith: It is the worst, even though it's amazing.
• 22:29
  Amith: It's cloud.
• 22:29
  Amith: 3.7 is both amazing and it's the worst you're ever going to get.
• 22:32
  Amith: So that's the thing to think about.
• 22:35
  Amith: You don't want to build your plans based around today's AI, because you know we all want associations to move faster.
• 22:41
  Amith: But in reality, if you were going to implement an automated member services function like this, I don't think you'd probably be doing it tomorrow or even next week, or possibly not even next month.
• 22:51
  Amith: It might take you six months to plan and implement something like this.
• 22:54
  Amith: It doesn't need to, by the way, from a tech perspective, but from just an organizational perspective, a cultural perspective.
• 23:00
  Amith: Budgets, all that.
• 23:01
  Amith: It takes time to do stuff, and every six months, you get a doubling in AI right in terms of AI capacity, ai power, and that is equivalent to also halving the underlying costs of the models.
• 23:12
  Amith: So AI is going to be able to do a very large percentage of these things.
• 23:17
  Amith: I'll give you a couple of examples, too, that AI can go beyond just doing what we do.
• 23:23
  Amith: So we tend to think of like the universe of possibility is that which we as humans can do.
• 23:28
  Amith: Okay, so that's cool.
• 23:30
  Amith: We definitely want to solve for that, because that's the expectation is, people want you to be able to do at least what you do with people in order to respond to these inquiries.
• 23:38
  Amith: But what about what people can't do?
• 23:40
  Amith: Your member services representatives, as wonderful as they may be, are not experts in your domain typically.
• 23:47
  Amith: Maybe some of them have some experience as surgeons, as accountants, as lawyers but typically not.
• 23:53
  Amith: Typically the people doing member services are focused on member services.
• 23:58
  Amith: They're not people with domain expertise.
• 24:00
  Amith: But AI can be a domain expert and not only deal with kind of rote FAQ style inquiries but also domain questions.
• 24:08
  Amith: So people can start emailing you, texting you, calling you about things that are actually helpful to them in the lane of their professional work.
• 24:17
  Amith: Now why is that interesting?
• 24:19
  Amith: It's interesting because associations tend to have this pattern of infrequent but deep engagement.
• 24:26
  Amith: Let me put it another way People come to your event but then they don't think about you the rest of the year.
• 24:31
  Amith: That's a problem in terms of a business model, because it's easy to get forgotten and easy for people to choose to not use you if they don't use you frequently.
• 24:40
  Amith: So if you can provide in-domain professional knowledge and response to inquiries about that professional domain of knowledge, that's leveling up right, because you don't offer that currently.
• 24:50
  Amith: So think not just about what you can do with AI that you currently do, but things that you could not do were it not for AI to begin with.
• 25:00
  Amith: So to me, that's a big opportunity.
• 25:02
  Amith: Around member services and member inquiries.
• 25:04
  Amith: It is, in fact, a process very similar to what you described, where you have people responding to emails, responding to text messages in some cases, and certainly some phone calls, and many of those things can be handled by AI.
• 25:16
  Amith: One key thing, though, I do want to point out I said it earlier and I want to drill into it just for a second you do need to account for, in these system designs, things that AI can't handle.
• 25:25
  Amith: So if you try to get the AI to answer something that it's not suited for, it's probably going to piss somebody off.
• 25:31
  Amith: So a good example of that is someone who is not happy with you.
• 25:35
  Amith: So let's say I write an email Mallory to Sidecar and I go hey, mallory, your customer service really stinks.
• 25:40
  Amith: And let's say it's the AI version of Mallory that's responding, and that's the thing that I'm complaining about.
• 25:46
  Amith: Do I really want an AI response to my complaint about the AI?
• 25:49
  Amith: Right, maybe not?
• 25:52
  Amith: Right, maybe not?
• 25:53
  Amith: And so I probably want to speak to a representative.
• 25:56
  Amith: Maybe I want to have that forwarded to someone else.
• 25:58
  Amith: So the first thing you want to do and this is something AI can indeed do is you want an AI to decide whether an AI should handle it or if it should be handled by a human.
• 26:07
  Amith: So, in that case and you actually probably want to be fairly thoughtful about that, so that you narrow the use cases where the AI attempts to respond and leaves the rest of the messages for people to actually respond to- I like that distinction between AI suitable and not AI suitable.
• 26:25
  Mallory: Amit, I know we have a good bit of influx of new listeners and viewers on the Sidecar Sync podcast.
• 26:30
  Mallory: I realize we mentioned Blue Cypress without maybe clarifying what that was, but for our new listeners and audience, Blue Cypress is the parent company of Sidecar and Blue Cypress has been working on a product that does exactly this right Responds to, can respond to member inquiries using AI.
• 26:48
  Mallory: Can you talk a little bit about that?
• 26:50
  Amith: Sure Happy to so.
• 26:51
  Amith: Within Blue Cypress, we're an incubator for new technology ideas, and we provide a variety of different professional services for associations as well, and one of our newest products that we're excited to be launching right now, as a matter of fact, is an AI agent called Izzy I-Z-Z-Y and Izzy is named after Isidore Sharp, who was the founder of the Four Seasons, and we were looking for people to honor with this particular AI tool, and very few organizations have a service standard quite at the level of a Four Seasons.
• 27:22
  Amith: We also thought about organizations like Nordstrom's and others that have extraordinary customer service, and we asked ourselves well, what if you could have AI deliver a Four Seasons or Nordstrom's or plug in a different brand that you think is aspirational from a quality of service perspective, every single time to your members across all associations?
• 27:42
  Amith: It would be remarkable.
• 27:42
  Amith: So that's what Izzy is, that's how Izzy got named, and Izzy is an AI agent that can plug into any number of your channels, includes email, includes SMS If you're international and you have people on WhatsApp or on Facebook Messenger, you can plug into that as well as well as secure communication modalities, izzy can process those messages.
• 28:04
  Amith: Izzy is an asynchronous agent right now, meaning that Izzy is not suitable for real-time voice communication today.
• 28:10
  Amith: That will be something Izzy will be able to do, probably by the end of the year, but right now AI is not quite fast enough to be excellent at that.
• 28:18
  Amith: It could just be like okay, so we're not tackling that yet.
• 28:21
  Amith: We're providing only asynchronous so email, sms, et cetera.
• 28:25
  Amith: And Izzy also is very, very close friends with another one of our AI agents named Betty.
• 28:30
  Amith: Betty is the domain expert, right, so Betty's a knowledge agent.
• 28:33
  Amith: Betty surfaces herself on the association's website or in apps.
• 28:38
  Amith: What Izzy does is Izzy's able to talk to Betty whenever an inquiry comes in.
• 28:42
  Amith: That's about domain knowledge.
• 28:43
  Amith: And then Izzy also is given tools, just like you would give a human agent tools to work with your AMS or your LMS.
• 28:50
  Amith: You train Izzy to work with the various software tools you already have in place, like your current AMS or LMS, and Izzy is able to use those tools.
• 28:59
  Amith: Again, you have to provide permission and the specific capabilities to do things like process member renewals, register people for events, cancel registrations.
• 29:08
  Amith: So you essentially give Izzy a vocabulary, that is, a set of tools that Izzy can use to provide service to your incoming inquiries, and Izzy takes care of it.
• 29:18
  Amith: All this is done with logging, so you have complete tracking of everything, so it's a really cool product.
• 29:23
  Amith: We're excited about Izzy.
• 29:24
  Amith: The goal of Izzy and I know we're going to touch on this later in this conversation the goal is not to displace member services personnel, but rather to increase the level of quality, both through responsiveness and better answers that your members get, and to enable your member services team to do the high-touch stuff that they're best suited for, which right now you can't do.
• 29:46
  Amith: Simple example of that is proactive outreach.
• 29:49
  Amith: What about actually calling your members or emailing them to talk about their needs before they actually ask you for something because they're upset about something?
• 29:59
  Mallory: Aziz sounds fun.
• 30:00
  Mallory: I like the names.
• 30:00
  Mallory: Amit and I are always big fans of infusing some fun when we can, so I'm glad it's not called service bot or service agent.
• 30:07
  Mallory: In your opinion, Amit, responding to member inquiries, having an AI augmented process for that, what would you say is difficulty to implement?
• 30:14
  Mallory: Low, medium, high, and then the potential impact of that.
• 30:17
  Amith: I would rate the difficulty as medium, because you do have to integrate whether it's Izzy or something else.
• 30:23
  Amith: You could build it yourself.
• 30:24
  Amith: There's other products that do this.
• 30:26
  Amith: You have to integrate this service capability with systems you have in place and with knowledge, so it's not low.
• 30:33
  Amith: I don't think it's high either, though, because the level of integration is not super, super deep Impact.
• 30:39
  Amith: If there was a category above high, I'd probably pick that.
• 30:41
  Amith: It's enormous.
• 30:44
  Mallory: All right, moving to the next task processing applications and renewals.
• 30:49
  Mallory: So the current state is manual verification, payment processing delays and limited follow-up capacity.
• 30:55
  Mallory: The AI opportunity here is automated application review, intelligent document processing and anomaly detection, the potential impact, streamlined processing workflows and improved conversion experiences for your members.
• 31:09
  Mallory: Streamlined processing workflows and improved conversion experiences for your members, and an example here would be a system that can automatically verify credentials, process payments and send personalized renewal reminders based on member behavior.
• 31:19
  Mallory: Amith, my question here.
• 31:21
  Mallory: We probably have a lot of people in our audience who are saying we have some pretty complex membership categories and eligibility requirements.
• 31:29
  Mallory: Can AI really understand that kind of nuance when processing applications?
• 31:33
  Mallory: Yes, full stop, no explanation, just yes.
• 31:40
  Amith: Yeah, ai is capable of understanding those kinds of nuances as well as the average human can.
• 31:46
  Amith: At this point, quite frankly and probably that's an understatement what I would tell you is this that I think that the nuance is actually where AI tends to shine really well.
• 31:54
  Amith: So things that are like this is a bit of a fallacy in all of our thinking, including my own, that when you think about computerizing or automating something you tend to think of like the rote, repetitive stuff that's the same every time, because that's historically where computers have played well.
• 32:15
  Amith: Right, if you can give it a set of very rigid rules, then the AI sorry, not the AI, but the computer pre-AI would just do those things over and over again, right?
• 32:19
  Amith: So it's very easy to process the same transaction over and over again if all the parameters are exactly the same.
• 32:23
  Amith: But the minute something changes or requires some judgment call, then traditional classical computing tends to just stop.
• 32:30
  Amith: And that's when you need a human to do a review of eligibility requirements.
• 32:33
  Amith: Or perhaps let's say it's an application not for membership, but someone's submitting a proposal to speak at an event.
• 32:40
  Amith: That requires a high degree of subjectivity.
• 32:43
  Amith: You have to review text and read what they've submitted and determine whether or not you already have a speaker on that topic and whether that topic is aligned with what you want for that event and whether that person has credentials that you think are suitable to be able to address your audience on that topic or in general, and on and on and on Right, and actually that's where I really that's, that's the missing ingredient from traditional automation.
• 33:06
  Amith: Traditional automation requires a highly rigid set of constructs or rules.
• 33:11
  Amith: Ai can operate in the gaps.
• 33:14
  Amith: Ai can operate where the nuance lives.
• 33:17
  Amith: So this is also an exciting piece, both because I don't know really anybody that I've ever met that enjoys processing applications.
• 33:26
  Amith: So if that's part of what you do or part of what your team does at your association, help is eventually going to be on the way and it's available really soon if you want it, and so you can take some of those steps out from a manual process.
• 33:38
  Amith: Now, does this mean that you don't review anything ever again?
• 33:42
  Amith: No, it is like a trust but verify kind of mindset.
• 33:46
  Amith: So when the AI does the review, does that mean, for example, that you have the AI completely program your upcoming annual conference and choose all of the speakers, align all the sessions and nobody does anything by hand, of course, not Just like if you hired a really smart recent college graduate to do that job for you.
• 34:03
  Amith: You would absolutely review their work before you published it on your website or before you considered it done.
• 34:09
  Amith: Similarly, with AI, you can have anywhere from like a random sampling, where you just check some of the work, to an actual review process where humans actually look at the work of the AI and approve it or reject it.
• 34:21
  Amith: So let's talk about one specific concept within this to maybe put a little bit more concrete structure around know structure around what we're talking about.
• 34:30
  Amith: So let's just take that speaker submission concept right and that's probably more part of the events department.
• 34:36
  Amith: If you're a larger association than it is membership.
• 34:39
  Amith: But membership often has like volunteer applications that they are dealing with and sometimes applications for membership itself.
• 34:46
  Amith: In some associations it's literally anyone who wants to join can join as long as they pay.
• 34:50
  Amith: But in some organizations, of course, there's a whole series of steps you have to go through.
• 34:54
  Amith: But let's take the speaker proposal submission process for a moment, because it's the same basic workflow.
• 35:01
  Amith: So one of the first things you want to do is determine whether or not the person's content that they submitted is relevant to the event.
• 35:10
  Amith: So if I have an event all about AI for associations and someone submits a session on agriculture or Is that relevant to my event?
• 35:22
  Amith: Probably not.
• 35:23
  Amith: Does that require a human to make that determination?
• 35:27
  Amith: Really it doesn't right.
• 35:29
  Amith: You can take that even to some very low level, low capability AI models and say here's an abstract from this individual who submitted it and here is an event definition of what we're looking for.
• 35:41
  Amith: It's essentially like a rubric that says these are the topics and these are the criteria for evaluating a submission.
• 35:47
  Amith: And I was extremely good at giving you that high level rough cut of is it a good fit or a bad fit?
• 35:52
  Amith: Right, and Mallory, you could go do that right now without programming.
• 35:56
  Amith: You could go do that one by one If I gave you a whole bunch of documents that were submissions and if I gave you a rubric, you could go and personally do that inside Plot or ChatGPT, right, and those tools would give you a pretty good answer saying yay or nay?
• 36:14
  Amith: And then probably actually give you, if you asked for it.
• 36:16
  Amith: In the prompt it would also give you an explanation of why it's a good or a bad fit.
• 36:18
  Amith: So that first round of review.
• 36:19
  Amith: My example is kind of silly, right.
• 36:21
  Amith: So typically people aren't that far off.
• 36:24
  Amith: There's a little more nuance to it, but you know that first phase often can cut out 20 to 50 percent of the submissions.
• 36:32
  Amith: The next thing that people often want to do is they want to compare that speaker's topic to other topics to determine if they have too much of a particular thing, right, right.
• 36:41
  Amith: So how many people are going to be speaking, let's say, on AI for, let's say, audio generation, at the upcoming Digital Now, which, by the way, is November 2nd through 5th in Chicago?
• 36:51
  Amith: So Digital Now, we want to have a fairly diverse array of content around AI for associations as the overall theme, but we don't want to have the same content over and over, right?
• 37:07
  Amith: So if we have six speakers that all submit a proposal to speak on using 11 labs for podcast generation or whatever, probably wouldn't be great.
• 37:13
  Amith: So that AI now has to like compare that new proposal against proposals that have already been accepted, or maybe just group them together so you can more easily analyze this, and the list goes on and on Each of these topics.
• 37:26
  Amith: Think about what they're doing.
• 37:27
  Amith: They are essentially taking unstructured data, which is the submission, and they're trying to gain some structure out of it.
• 37:34
  Amith: They're trying to basically pull some structure and, in some cases, make some, effectively, judgments around that unstructured content and, as we've talked about on this pod and many other venues, that's something that AI is incredibly strong at.
• 37:48
  Amith: Already.
• 37:48
  Amith: Here you're just dropping it into your workflow as an association.
• 37:51
  Amith: So the manual process that I suggested Mallory could do without doing any programming, of course, is an option and that'll probably save you some time.
• 38:00
  Amith: But what will save you a lot more time is if you build AI into the workflow.
• 38:04
  Amith: So if you have a website that takes in a submission to build a little AI, step directly into that to do that first cut of review and even give real-time feedback to the submitter saying, hey, this isn't quite right for us, which is really valuable to the person that's taking the time to submit something to you to begin with.
• 38:22
  Mallory: With this task, Amit, so processing applications and renewals, or even extrapolating and maybe using it for the event speaking proposal process.
• 38:28
  Mallory: What would you say is difficulty to implement and then impact?
• 38:45
  Amith: current process.
• 38:45
  Amith: So some people just have a website that dumps data into like a Google sheet or into an Excel document with all the submissions.
• 38:49
  Amith: You could take that Excel file or that Google sheet and directly drop it into any of the modern AI tools and ask it to compare each of those submissions relative to a rubric or a criteria, essentially, and give you an answer, and so you could do that manually and that would be quite useful.
• 39:06
  Amith: I don't know that it would really save you an enormous amount of time ultimately.
• 39:10
  Amith: I mean you would save time and that you wouldn't have to manually read every one of the responses or the submissions, but you'd still be doing some work by hand.
• 39:18
  Amith: But that version is low effort because you could do that really today with any of these consumer grade tools and the impact would be probably low to medium.
• 39:25
  Amith: I think it'd be useful.
• 39:26
  Amith: Low effort because you could do that really today with any of these consumer grade tools and the impact would be probably low to medium.
• 39:31
  Amith: I think it'd be useful, but it wouldn't really like cut out a ton of time probably.
• 39:32
  Amith: I don't know.
• 39:33
  Amith: Do you agree with that or?
• 39:34
  Mallory: Yeah, I agree with that.
• 39:35
  Mallory: I mean, I think it would maybe have a high impact on your staff who, like, don't necessarily want to do that, but maybe not necessarily on the members, not quite as much as handling those inquiries and having, like, a knowledge assistant.
• 39:48
  Mallory: I think that would be pretty impactful.
• 39:50
  Amith: No, I think that's true and, at the same time, the next level up, the version of this that would likely have a very high level of impact, is if you put some automation in place right where you don't do it manually, where you say hey, you know, we, through the manual process, we've proven to ourselves that in fact, I can do an extraordinary job of evaluation of content or whatever it is in this application process.
• 40:14
  Amith: Now let's weave it into our actual business process.
• 40:15
  Amith: Right, so on our website when you submit a proposal, when you submit a membership application to have AI directly woven in in an agentic way, as we've talked about.
• 40:24
  Amith: Now.
• 40:25
  Amith: You're talking about a medium to high implementation effort but also a much higher level of impact, because you both reduce staff time.
• 40:32
  Amith: But also, in this case, it's that dual win or the win-win, like we talked about with the responses to inquiries.
• 40:39
  Amith: Here you now have people getting reactions from you as a system much faster, right, possibly real time or within minutes, whereas it might have taken them days, weeks or longer or maybe never, to get a response to their proposal previously.
• 40:53
  Amith: And actually just think about that for a moment.
• 40:55
  Amith: On the association side of the equation, you have this seemingly endless stream of stuff you have to review, either in-house or, a lot of times actually through a volunteer committee which you don't directly control.
• 41:06
  Amith: These are again volunteers, and so that's frustrating on a lot of levels.
• 41:10
  Amith: But think about it from the other perspective for a minute.
• 41:12
  Amith: So it's the famous Amazon empty chair in the conference room which represents the customer, and so in this case, if you think about the person submitting the content as maybe not the customer, but they're a stakeholder in your process they're not getting a great experience, either because they're waiting a really long time.
• 41:30
  Amith: A lot of times, the feedback they get is either yeah, you're in, great Congratulations, or sorry, you're not in.
• 41:37
  Amith: And they might have to wait three months to get a response like that, which really stinks, because they might not know if they're going to come to your event until very close to the event.
• 41:45
  Amith: They might not ever hear back if they're getting a no, that happens in some cases, but it's really really bad.
• 41:52
  Amith: And well, you want the best content, don't you?
• 41:55
  Amith: You want your association to be the premier place for people to want to come speak or to submit content to your publications, and so making it as low friction as possible is really valuable, because you have people that are more likely to come to you with their ideas if that's the case.
• 42:11
  Amith: So I think it's a really value-added process from that viewpoint.
• 42:15
  Amith: The other thing that you could do is, let's just say, someone submits a proposal.
• 42:20
  Amith: So Mallory submits a proposal to speak at Digital Now, and we're like, oh, mallory has a great background, she's, she'd be a great speaker for us.
• 42:27
  Amith: She even has, let's say, prior speaking with us.
• 42:31
  Amith: That was well received by the audience, or whatever, but she submitted a topic that we've already accepted.
• 42:35
  Amith: It's 95 percent similar to something else we already have programmed.
• 42:38
  Amith: So what would normally happen?
• 42:39
  Amith: Well, first of all, it would take a long time for Mallory to hear back, and then then she'd eventually get a no right, not a hey.
• 42:45
  Amith: Mallory, thanks so much for your interest in speaking at Digital Now.
• 42:49
  Amith: We'd love to have you involved, but that topic is not quite right for us.
• 42:52
  Amith: What about one of these three topics right To give you feedback and suggest an alternative, and then let you resubmit with the alternative topic?
• 42:59
  Amith: These are all things that AI can do beautifully and very, very rapidly.
• 43:04
  Amith: So there's a lot of opportunity here to level up in terms of the quality of your content and the way you engage these people who are really trying to submit valuable insights into your organization.
• 43:18
  Mallory: The next task we want to dive into is database maintenance.
• 43:20
  Mallory: So the current state of that is manual updates, the duplicates, inconsistent data.
• 43:22
  Mallory: That is, manual updates, the duplicates, inconsistent data.
• 43:25
  Mallory: The AI opportunity is automated record deduplication, data enrichment and quality monitoring.
• 43:32
  Mallory: The impact higher data accuracy Don't we all want that?
• 43:38
  Mallory: With minimal staff time investment, and an example here would be having an AI that continuously scans and cleans member records, identifies outdated information and even suggests updates.
• 43:46
  Mallory: So, amit, something we hear over and over again with associations is data quality is a barrier to AI implementation.
• 43:54
  Mallory: This one, for me, is a little bit chicken in the egg, because I know your answer is AI can absolutely do this right now and it's not that difficult to implement.
• 44:03
  Mallory: That's what I think you're going to say.
• 44:04
  Mallory: But if you have to kind of get your data in a good spot to implement AI, but then AI can clean up your data, it's kind of like which comes first.
• 44:12
  Mallory: So I'm hoping you can talk a little bit about that.
• 44:16
  Amith: Yeah, well, I think there's elements of this that are definitely low hanging fruit, right, and there's pieces of it that AI can help you with fairly quickly but, at the same time, to fully and thoroughly implement.
• 44:26
  Amith: This is definitely not an easy lift, because people have typically dated infrastructure, to be kind in the way you describe it.
• 44:34
  Amith: You know it's kind of funny.
• 44:35
  Amith: I remember my prior life.
• 44:37
  Amith: I'm no longer involved in this, but for 20 plus years I started and then ran an AMS company, and this software company provided very large associations with their core database system, and I remember one of the things we used to do when we demoed the software is we'd go and ask people in the room hey, do any of you have problems with duplicate data?
• 44:57
  Amith: As kind of leading into a really cool feature we had that would help you merge duplicate data.
• 45:01
  Amith: And, of course, everyone pretty much said, oh man, yeah, no, that's such a pain point.
• 45:06
  Amith: So, and except one time, I remember there was this guy who was adamant that his organization had no duplicate records, zero and I said do you have no data?
• 45:18
  Amith: Because that's the only way I can imagine having no duplicates.
• 45:20
  Amith: What he was referring to was a technical aspect, that the primary key of the table was in fact, distinct and unique and guaranteed by the database to be unique.
• 45:28
  Amith: Of course that doesn't mean anything, because you can have multiple records that do have different ID values, but of course they're the same person and that's fundamentally the issue.
• 45:36
  Amith: So, in any event, I thought that was quite interesting.
• 45:39
  Amith: That happened once over 20 something years, but it was a problem back then.
• 45:44
  Amith: It's been a problem since the beginning of time.
• 45:46
  Amith: It's even a problem with paper filing preceding digital right, where you have the same file and someone fills out you know, I don't know, maybe you've gone to a doctor's office and they've asked you to fill out the same damn patient intake form again, even though you've been going to that practice like for 10 years.
• 46:00
  Amith: We've all experienced that right.
• 46:02
  Amith: So that a version of of the analog version of duplicate data.
• 46:06
  Amith: So what is the capability we're talking about here?
• 46:09
  Amith: The first thing is is that we want to be able to detect, detect duplicates, right.
• 46:14
  Amith: So deduplicating or merging the data, that's a complicated thing because you probably have multiple systems and even within a given system, a lot of systems do not support the idea of eliminating duplicates or merging records, at least they don't do it very well.
• 46:29
  Amith: So we're not going to really talk about how to solve for that, because your underlying database systems like AMS, lms, et cetera.
• 46:36
  Amith: It's really, really tough, and so modernizing that technology, of course, is a good idea, but sometimes that's a tall order, at least in near term.
• 46:46
  Amith: So what we're going to talk about right now for a minute is how do you detect duplicates, because, let's say, for example, from the customer perspective, the annoying thing is getting three emails from you that all say the same thing, because you have them in your database three times.
• 46:59
  Amith: Well, maybe you have them in your database three times, but you want to figure that out and send them one email.
• 47:06
  Amith: Right, and that might sound simple, because you say, oh well, I have a distinct list of emails, but what if the emails are slightly different?
• 47:12
  Amith: Right, a lot of people have email aliases.
• 47:14
  Amith: You might be Mallory at Sidecarai, you might be Mallory M at Sidecarai.
• 47:19
  Amith: You might have several emails.
• 47:20
  Amith: So it's a lot more complex than it may appear.
• 47:23
  Amith: So AI is capable of doing this.
• 47:25
  Amith: So there's some different kind of layers that I described First.
• 47:28
  Amith: Number one thing here I'm about to say is make sure you're working with a system that you trust, so don't go and be a free user of ChatGPT.
• 47:36
  Amith: Don't send your data to deepseekai, because that's not housed here.
• 47:40
  Amith: But if you want to just take a spreadsheet of emails that you're about to send out like first names, last names, emails, organizations and then throw them into a cloud again a paid account with cloud or one of the other tools that you consider trustworthy you'll be able to get a list back that's deduplicated with a very high level of accuracy from a consumer tool.
• 47:59
  Amith: Now, the more columns you put into that file, the more likely the AI will be able to actually help you dedupe it.
• 48:05
  Amith: Now, that's just done with a really simple language model.
• 48:08
  Amith: Right, it's not nearly as good as what you might have if you use a more sophisticated approach, but there are tools that you can use as just a regular business user not a developer, not an IT person to help you a little bit at least to minimize the impact on your customer or member.
• 48:23
  Amith: So that's a good thing.
• 48:24
  Amith: To help you a little bit at least to minimize the impact on your customer or member, so that's a good thing.
• 48:26
  Amith: The part of it I'd say like the next level up from there is to implement an AI data platform where you bring in your data from your AMS, your LMS, your CRM, your marketing automation system and so forth, and then unify that data and then get a more sophisticated AI technique, which is using something called vectorization, to be able to compare all of the data across the board and find duplicates with a much, much higher level of precision.
• 48:50
  Amith: That's probably a medium level of effort.
• 48:53
  Amith: It's not enormous, but it's non-trivial.
• 48:55
  Amith: You still have to set up an AI data platform, which is a somewhat technical topic, and then you have to bring the data in, then you have to run this process for deduplication, but it's well within the capability of most associations to be able to do this.
• 49:09
  Amith: It's just a project you have to choose to undertake.
• 49:12
  Amith: And then the last thing I want to quickly say is that duplicates, of course, are often the beginning of the pain, but they're not necessarily the end.
• 49:19
  Amith: Another problem is decaying data, right?
• 49:23
  Amith: So a lot of times you have a scenario where you have a member and that person worked at company A and then they decided to leave and, shockingly, you were not the first call they made when they changed jobs.
• 49:35
  Amith: You, the association, were not the very first person, very first organization they called to inform, call to inform, but they did update their LinkedIn or maybe they updated something else, right?
• 49:45
  Amith: So what if we could have AI automatically match people up based on their publicly available data and then update our database very seamlessly so that we knew where people went right, tracking them down and updating our database, and that's also something you can do.
• 50:00
  Amith: That's actually fairly low effort level.
• 50:02
  Amith: There's a variety of data services you can buy.
• 50:04
  Amith: There's ways of doing matching fairly easily, and then you can, depending on your system.
• 50:08
  Amith: Of course, you can update your system with that new data.
• 50:10
  Amith: So I think it's a really important concept.
• 50:13
  Amith: That's actually kind of related to the duplicate data, because one of the reasons you end up with duplicates is because of subtle changes.
• 50:19
  Amith: Someone gets married and changes their surname, someone changes jobs, so they have a different email, right?
• 50:26
  Amith: So these keys that we use to try to identify dupes oftentimes are invalid because people have slight differences If they're literally the exact same record, even systems from the 1970s could detect those duplicates to implement could be kind of low to medium range.
• 50:45
  Mallory: I really like, too, how you keep bringing up ways that, like an individual can go and do this right now.
• 50:47
  Mallory: That's really important to remember.
• 50:48
  Mallory: You don't have to like overhaul your whole business.
• 50:50
  Mallory: You can go to chat to BT and Claude as a paid user and do this yourself and then impact.
• 50:55
  Mallory: To me it kind of seems like it depends because, I mean, if you have really good data about who your members are and where they work, right, you could do like some excellent proactive outreach to them.
• 51:05
  Mallory: So would you say this is kind of maybe like medium impact.
• 51:09
  Amith: I think it starts off as medium.
• 51:11
  Amith: I think it could be very high if you implemented it kind of at the organization wide level and made it a consistent and nearly automated process to continually have correct data.
• 51:20
  Amith: Because think about it in your own experience as our listeners, the number of times you have personally experienced pain because of duplicates, just doing a lookup in your database and saying, oh, I'm trying to find Mallory, and you find eight versions of the Mallory record.
• 51:35
  Amith: You got to look through every one of those to find the transaction or the event registration or whatever that you're looking for.
• 51:40
  Amith: It's an enormous amount of friction internally.
• 51:43
  Amith: It also tends to have a pretty big impact on external perception or external experience, because if there's multiple duplicate records they often have a way of kind of making their way into the consumer experience on the website, where Mallory has a harder time logging in or she can't find her whole history.
• 52:01
  Amith: Not all of her transactions are on her web profile, only some of them are.
• 52:04
  Amith: But there's this whole other like phantom Mallory record that's out there with a bunch of other transactions.
• 52:09
  Amith: So there are a lot of issues that come from this.
• 52:12
  Amith: This is one of those.
• 52:14
  Amith: You know, death by a thousand paper cuts thing.
• 52:16
  Amith: No association I know of has ever ceased to exist because they had too many duplicates in their database.
• 52:20
  Amith: But the impact of those duplicate records could, in fact, radically impact the quality of customer experience, member experience, cause them to need way more staff time put towards like rote functions versus higher order things.
• 52:34
  Amith: So I do think it's a pretty big impact area and, to your point earlier, I really like to focus on tangible here and now, how an average person listening to this pod or watching us on YouTube can put these ideas into action.
• 52:49
  Amith: Today and many times we're talking about like scaling up a process, taking it to the systems level.
• 52:55
  Amith: It is actually better if you didn't try to automate it to begin with, right.
• 53:00
  Amith: So think about, like this idea of the automated review process of an application.
• 53:04
  Amith: I don't want to automate that completely for an association right off the bat.
• 53:08
  Amith: I want to experiment with a bunch of different manual steps using AI to do the ratings, the rankings, etc.
• 53:13
  Amith: But to learn from that process to see what's working well and say, oh, this is really good, these are great responses.
• 53:20
  Amith: Now let's scale that up and automate it Right and automate it once we know the process makes sense.
• 53:26
  Mallory: Yep, and that's exactly the process we follow here at Sidecar Next up.
• 53:31
  Mallory: I want to talk about communications in general within member services, but really service communications and updates.
• 53:38
  Mallory: The AI opportunity here is tailored member service notifications based on specific needs and preferences.
• 53:50
  Mallory: Impact would include things like increased response rates and improved member satisfaction overall.
• 53:56
  Mallory: An example here would be an agent that can send personalized renewal reminders with specific benefits relevant to each member's usage patterns, or customized service updates about changes that affect their particular membership type.
• 54:10
  Mallory: Amit.
• 54:11
  Mallory: When we talk about personalization with AI, we're talking about personalization at a one-to-one level.
• 54:17
  Mallory: I know Blue Cypress has started experimenting with this with a few associations with a product called Rex.
• 54:24
  Mallory: Can you share any of like the preliminary results that we've seen from that or insights from personalized communications?
• 54:31
  Amith: Sure, rex.
• 54:31
  Amith: So Rex is our personalization agent and engine and Rex actually is the brains behind the Rasa newsletter.
• 54:39
  Amith: So for those people that are familiar with Rasa's AI newsletter, that's a recommendations tool, right?
• 54:44
  Amith: So it just happens to be in the form of a newsletter.
• 54:46
  Amith: Rex is the brain behind that, and so we've done a number of interesting experiments with Rex in the area of personalized networking recommendations and personalized event session recommendations.
• 54:58
  Amith: So for some organizations over the last six months, we've set up some very simple campaigns that have essentially shared through email, suggested people that they could connect with at an upcoming event and suggested sessions that might be relevant to them, and the response has been incredible.
• 55:16
  Amith: First of all, if you measure success of those types of campaigns by clicks and opens, the data has been off the charts, you know, in excess of 100% open rates on some of these campaigns, which sounds crazy because people have been opening these emails more than once.
• 55:30
  Amith: So the distinct or unique open rates close to 100%, with click rates that are higher than you know, most email campaigns dream of having for an open rate.
• 55:38
  Amith: And why is that?
• 55:39
  Amith: Why is that?
• 55:40
  Amith: It's the same reason why AI powered newsletters are dramatically more effective than generic newsletters, because people get stuff that's actually relevant to them right.
• 55:50
  Amith: At the end of the day, what we want is what we want as individuals, right, what I want, what you want.
• 55:56
  Amith: We all have different needs and those needs are constantly changing, and up until recently, you had to be kind of at the, you know, to be able to punch at the weight class of an Amazon, netflix or Microsoft to be able to do personalization at scale.
• 56:08
  Amith: The good news about AI getting continually smarter and cheaper is that the underlying technology becomes more general purpose, more powerful and less expensive all the time, and so it is possible for associations to do true personalization.
• 56:22
  Amith: I do want to say one other quick thing that you know this idea of segment-based approaches.
• 56:27
  Amith: It's better than nothing sometimes, but sometimes it's actually worse than nothing.
• 56:32
  Amith: And so like, do you send like one generic email to everyone or do you try to like go through and create personas and create segments, which was, for a while, considered like state-of-the-art both in terms of tech but also from a marketing strategy perspective?
• 56:45
  Amith: Right, there's entire consulting firms who do nothing other than like design personas.
• 56:48
  Amith: The problem is, is they're one dimensional, meaning that you can only fit into one of those typically.
• 56:53
  Amith: So you know, you might have a persona for young professionals, a persona for later career professionals, and each of those personas might have, you know, people, even like get pictures of people and say, hey, this is Jane, jane is an early career professional, here's Harold.
• 57:07
  Amith: Harold is a late career professional, right.
• 57:09
  Amith: And then they kind of like personify these people in a way where they try to like say, ok, well, I'm going to put these kinds of people, get this kind of content.
• 57:15
  Amith: It's very, very, very generic and of course it's overly generalized by definition, and so you miss a tremendous amount that way, and sometimes I say it's even worse in some cases than like one size fits all communications, because at least in the latter you're not excluding content from people who may find it useful, right.
• 57:34
  Amith: So, for example, like some associations I've talked to said oh well, retirement planning advice, we're going to focus on early career professionals because that's when they really need to get started.
• 57:41
  Amith: Early career professionals, because that's when they really need to get started.
• 57:43
  Amith: True, early career professionals should absolutely pay attention to retirement planning right away, but a lot of times they don't.
• 57:49
  Amith: And so then the late career professionals are like I've got a problem, I want to retire in five years and I don't have much of a 401k or an IRA put aside.
• 57:59
  Amith: What do I do?
• 58:00
  Amith: And the association often has a role to play in educating those folks too.
• 58:03
  Amith: So that's the problem with overgeneralization, and this type of AI can really help solve for that, which is really exciting.
• 58:11
  Mallory: Difficulty to implement and potential impact.
• 58:14
  Amith: So the difficulty to implement, I would say, is solidly medium.
• 58:18
  Amith: For this, you need to implement an AI data platform of some sort to get your data into a location where a recommendation engine Rex or anything else there's a bunch of them out there, right?
• 58:28
  Amith: You need to have your data in order to do recommendations.
• 58:31
  Amith: Absent the data, there's nothing to recommend.
• 58:33
  Amith: So you have to get your data somewhere and that is not a low lift for most people.
• 58:37
  Amith: And the impact I really think the impact of this one's high Mallory because it's so powerful to get people stuff they want.
• 58:49
  Amith: You know one little tidbit from one of the campaigns I remember hearing about from our team, about Rex and the event stuff we were able to, through this technology, connect professionals together in one association that's in the scientific domain who had similar research that could collaborate but had never met each other even though they were in a very narrow field of science, and that's really powerful because you know those folks can collaborate in new ways.
• 59:10
  Amith: Maybe they, you know, do something truly amazing right in terms of discovery.
• 59:14
  Amith: So if AI can help bring people together in that way, that's like the fundamental thing associations are about right to associate.
• 59:20
  Amith: So I think it's a really really exciting opportunity.
• 59:23
  Amith: So I think it's a really really exciting opportunity.
• 59:24
  Mallory: The last task we want to break down here is upselling and cross-selling.
• 59:29
  Mallory: So the current state of this responsibility within member services is kind of ad hoc, probably a lot of missed opportunities there.
• 59:36
  Mallory: The AI opportunity is predictive product and service recommendations and propensity modeling, potential impact.
• 59:46
  Mallory: It would be an increase in non-dues revenue and enhanced member value, and an example here would be an AI that can suggest relevant courses or events to members based on their career stage and past participation.
• 59:56
  Mallory: So this kind of falls in with the prior one on personalized communications.
• 1:00:00
  Mallory: Amit, my thought here is I was interested to see this included in the deep research report because I don't know that.
• 1:00:07
  Mallory: I personally have had a ton of conversations with association leaders around cross-selling or upselling.
• 1:00:13
  Mallory: So do you feel like this is an important issue for associations, a missed opportunity?
• 1:00:18
  Mallory: What are your thoughts?
• 1:00:19
  Amith: I think it's a massive missed opportunity.
• 1:00:21
  Amith: It's hard to do at scale, certainly it's hard to do even as an individual, because you have to know a lot about the individual that you're trying to cross-sell or upsell in order to be useful, and you also have to know your full catalog of available things, right, whether it's products or if it's events or educational offerings.
• 1:00:36
  Amith: So it's really an AI scale problem.
• 1:00:39
  Amith: But think about it from the viewpoint of let's rewind in time to, let's say, I don't know, it could be decades ago, it could be hundreds of years ago.
• 1:00:47
  Amith: Imagine you're a shoe salesperson and you work out of a little store in a little town and you have a town of a couple thousand people and you're the local shoe salesperson, right, and you get to know all your customers.
• 1:01:01
  Amith: You know them, you know their families and you also probably remember a lot of their purchase history.
• 1:01:06
  Amith: You know when did they buy, let's say, dressier shoes or when did they buy casual shoes, and so, therefore, when that customer comes in, if you have a new product that's relevant to them because let's say it's a, you know, super powerful like break resistant shoelace that has come out that you know my teenager would love to have because he's always going through shoelaces, right.
• 1:01:28
  Amith: So and that comes out and I say, hey, that'd be great.
• 1:01:31
  Amith: But, like, if the customer coming in doesn't use shoes, that use shoelaces, maybe it's not a great thing to cross sell.
• 1:01:39
  Amith: That's probably a silly example, but the point is, is in that old school example you had a salesperson who knows their customer really well, he's known them probably for quite a while and also knows their product catalog of their 50 or a hundred products, whatever they sell, really really well.
• 1:01:55
  Amith: So they are the brain at the intersection of products to offer and the customers they're trying to sell them to.
• 1:02:00
  Amith: They also have timing, you know, built in right?
• 1:02:03
  Amith: Because if the customer customer just spent $100 on a pair of shoes which to me still sounds like a lot of money, but I realize shoes cost a lot more than that now typically but in any event, they're probably not going to buy another $100 pair of shoes tomorrow because they just bought one today, right?
• 1:02:16
  Amith: So all of these various factors go into how to make recommendations work.
• 1:02:22
  Amith: And upselling and cross-selling is one of the most beautiful aspects of it because, yes, you're generating more revenue for your association.
• 1:02:29
  Amith: But let's say that I know Mallory really well and I say, hey, there's this course we offer.
• 1:02:33
  Amith: Mallory, that would be perfect for you.
• 1:02:35
  Amith: You are already really strong in AI and all of these categories, but there's this one area that I know you've told me in the past you were interested in learning more about, and we just introduced a new course in this area.
• 1:02:46
  Amith: It's reasonably priced, it's available online at your convenience.
• 1:02:50
  Amith: What do you think?
• 1:02:50
  Amith: And Mallory was like crap.
• 1:02:51
  Amith: That sounds awesome.
• 1:02:52
  Amith: It's like exactly what I was hoping to learn and I can do it on my own time, and it's only $15 or whatever, right it's like reasonably priced.
• 1:02:59
  Amith: So that's the beauty of cross-selling and upselling is that you're creating more value for everyone involved.
• 1:03:04
  Amith: But once again to your first point, mallory.
• 1:03:06
  Amith: It requires good recommendations.
• 1:03:09
  Mallory: Yeah, impact.
• 1:03:10
  Mallory: I think for this one sounds like it could be high, incredibly high for your association Difficulty to implement.
• 1:03:16
  Mallory: I feel like if you need all of that background, that backstory, like in the shoe salesman example of your customer or your members, that seems like in theory it would be a high difficulty to implement.
• 1:03:26
  Mallory: What do you think?
• 1:03:28
  Amith: I generally agree with that.
• 1:03:29
  Amith: I think there's lower versions of that that are easier, where you can do it through like, let's say, emails, but, if you want to like, integrate this with your website.
• 1:03:35
  Amith: That tends to be an area that's harder for people to implement.
• 1:03:38
  Amith: Prerequisite number one get your data house in order.
• 1:03:41
  Amith: Get all of your data into an AI data platform of your choosing.
• 1:03:44
  Amith: That means wiring your data into one data environment from your AMS, your LMS, your CMS, your FMS, all the MSs right.
• 1:03:53
  Amith: Bring them all in.
• 1:03:54
  Amith: Then you can run a recommendation engine to understand the relationships amongst all these.
• 1:03:58
  Amith: You know disparate entities of data and then you can build the functionality to say now I know what I should be recommending.
• 1:04:05
  Amith: Let me connect people with products or people with people or people with events in the right way.
• 1:04:10
  Amith: But you can't really upsell and cross-sell intelligently until you have the recommendations in place.
• 1:04:16
  Amith: So I think you're right.
• 1:04:17
  Amith: I think it's effectively.
• 1:04:18
  Amith: It's a medium to high level of implementation difficulty, but the impact is extraordinary of implementation difficulty, but the impact is extraordinary.
• 1:04:32
  Mallory: Well, everyone, thank you for tuning in to part one of our AI Enhanced Member Services episode.
• 1:04:33
  Mallory: We are really excited to get to part two, where we'll take all this inspiration that you're feeling and kind of talk about the practical next steps of how you can implement this Reminder.
• 1:04:41
  Mallory: We broke down everything task by task for ease of explanation, but the real transformative power here is when you start thinking about bundling some of these tasks into one kind of overarching way to transform your member services.
• 1:04:55
  Mallory: So with that, we will see you all in part two.
• 1:05:00
  Amith: Thanks for tuning into Sidecar Sync this week.
• 1:05:02
  Amith: Looking to dive deeper?
• 1:05:04
  Amith: Download your free copy of our new book Ascend Unlocking the Power of AI for Associations at ascendbookorg.
• 1:05:12
  Amith: It's packed with insights to power your association's journey with AI.
• 1:05:16
  Amith: And remember, sidecar is here with more resources, from webinars to boot camps, to help you stay ahead in the association world.
• 1:05:24
  Amith: We'll catch you in the next episode.
• 1:05:26
  Amith: Until then, keep learning, keep growing and keep disrupting.