The AI-native shift: What SMG's Ignite® means for the future of experience management

Full episode transcript

[00:01:00]

Khalipha: Please welcome to the podcast Jamie Ferguson. Jamie, welcome to our podcast.

Jamie: Thank you, Khalipha. I'm so excited to be here today to talk to you.

Khalipha: Amazing. How are you doing, Jamie, today?

Jamie: I'm doing great. I'm enjoying the spring-like weather we're having. Couldn't be better. How about you?

Khalipha: Good. Good. It is a very warm day here in Cape Town. I wish it was spring, because I'm sure that'd be, like, a little bit cooler, but we're, we're, um, we're having fun. 30 degree Celsius weather down here.

Jamie: Oh, that's warm. That's warm.

Khalipha: Yeah, yeah. When, when it's summer, it's, it's quite beautiful. But yeah, no. Look, Jamie, thank you so much for taking the time to [00:02:00] talk to us today about all things AI, which I don't know about you, but you're seeing it everywhere, right?

Jamie: It is everywhere. That is for sure. It's ... It is ... It, it has just been embedded and ingrained into everything that we are doing.

Khalipha: Exactly. And even though we see AI is everywhere, I feel like, you know, many organizations still struggle to translate it into meaningful operational impact, right? For me, it's about trying to get organizations to see that AI can help reshape brands and the way that they listen to their customers, and how they can empower frontline teams, and how they can design better experiences overall.

So yeah, I'm, I'm really excited to delve into this topic around AI and AI native, which I think it's quickly becoming an industry shorthand, and maybe it's also often misunderstood. So as I kick off with the first question, Jamie, how do you define what it really means, AI native, and [00:03:00] why does it matter right now for brands, especially in retail, restaurants, and consumer packaged goods?

Jamie: Great question. And I think when a lot of people say AI native, what they really mean is AI added onto my legacy infrastructure, and I think that totally misses the point. So when I think about AI native, I mean that the product was designed around AI as the core operating model from day one. So at SMG, we define an AI native platform as being built with a consistent, unified data model with real-time compute and agent layers so AI can reason across the whole experience picture.

So, this would include the ability to consider history, preferences, intent, and then most importantly, act or recommend within that context and not just produce isolated analysis. So in practicality, the outcome is speed and consistency and really [00:04:00] deeper accuracy. AI native systems respond in real time, they surface root causes, and they propose concrete next steps, whereas we'll call it bolt-on AI is often slow, fragmented, and, and really just narrow in scope. It's really important because today's brands are drowning in data, and it's massive, messy, and fragmented.

And AI-native platforms are really the only practical way to turn that continuous stream of signals into timely and trustworthy decisions at scale.

Khalipha: Ooh, you touched on things like speed. And that's one of the things I think of when I think of AI, it's like it's helping me and everybody else move really quickly.

Jamie: That's right.

Khalipha: But jumping onto your point there, yeah, I feel like many brands actually do say that they've, like, added AI into their organization or they're using AI, yet they're not really seeing that meaningful impact. And so,o my next question for you then is, like, [00:05:00] where do organizations usually go wrong when it comes to layering AI into those legacy systems?

And what makes the AI native foundation behind SMG's Ignite so different?

Jamie: This is my favorite topic. Most organizations try to add their AI into their legacy infrastructure, and it fails for three really predictable reasons. The first is they have inconsistent and siloed data, and so AI has no single source of truth.

The second is their systems and pipelines are then not optimized for real-time reasoning, so insights and decisions become really disconnected. And I think the third thing is that human workflows remain manual, so insight to action is slow and error-prone. But I think there's a flip side to that third one in that it, it could instead be that workflows become entirely automated, but it's based on low-quality data and insight, and I think [00:06:00] that might be even worse.

So bolt-on AI can generate really flashy output, but if it does not have that clean, unified data foundation, it's s- it's really noisy and untrustworthy. I think, I know the reason that many companies take this bolt-on approach is because doing an entire rebuild of infrastructure takes a huge amount of work, and I say that because I know from experience.

In 2023, we took a huge step back and we looked at our legacy infrastructure, and even with all the AI that was available at the time, we realized that it was not gonna provide the best results for our clients. And so we made the very hard decision to take the time to build a whole new infrastructural structure that we're calling Ignite.

So it, it, it's treating AI as that core operating model.

Khalipha: Woo, look at that. We're ahead of the game.

Jamie: That's right.

Khalipha: We're ahead, ahead of the game. That's incredible. And I think, like, I, I, I wish I had all the [00:07:00] technical jargon to go, "Oh my gosh, yes, it, Ignite does this and that," and, and, and like I understood like the deeper mechanics.

I'm just a marketer. Talk to me about the creative side of things. But like you said, I do see how Ignite was built to be native or AI native from the ground up, which I think is really something powerful. With that power, I guess I'm gonna ask you here if you can maybe talk to us a little bit about Ignite.

Like, what is it, and how is it enabling those real-time decisions that operators can actually trust in today's age?

Jamie: Sure. Let me build Ignite for you. So it's a, it's an AI native unified experience management engine, and it starts with being cloud native, and it's built on that unified experience data model, and then we put the reusable AI agents on top of that.

So on the, on the base foundation of it is the unified data model, which means every signal, structured scores, free text comments, video, community feedback, research, all of that is [00:08:00] normalized into a consistent representation, and then AI can re-reason over that data. So, if we think about the agents then that, that are on top or embedded in that data model, on the front end, we have kind of feedback solicitation agents with our moderation and video agents.

So these agents are really designed to enhance data as it comes into Ignite. Then we have our agents, like Knowledge Agent, that are meant for conversational exploration of the feedback that we're collecting. We have an alerting agent that can detect anomalies in the data, and then we have a whole corpus of action and task-oriented agents that create and track the work.

So because these agents are referencing the same clean data in real time, they produce contextual and auditable recommendations, and they can even initiate, we'll call it low-risk actions that keep the humans in the loop. So this is the combination that really lets operators trust Ignite [00:09:00] to recommend and automate frontline decisions.

I think another aspect, sorry, another aspect of Ignite is that it really breaks down those silos that historically have, have separated brand, customer, and employee experience initiatives. So the unified experience management data model approach normalizes all of those scores in open texts and research outputs, transaction and operational signals into a single kind of queryable representation, so AI agents can then reason across domains.

So, once the data model is... the data is modeled consistently, those same agents that I talked about, Knowledge Agent, Action Agent, Alerting Agent, they can surface then correlated signals that generate cross-functional recommendations. So, for example, if there was a product change that led to rising complaint counts and increased frontline handling time, the models can actually propose a coordinated fix [00:10:00] to the product, plus store level coaching and, and potentially process changes.

So that single source of clean data removes the, "I don't have the right data," excuse and lets all those teams act together.

Khalipha: Oh, that sounds really robust, very well thought through. Love the fact that, again, Ignite's been built from the ground up to be this AI native platform, but it also allows people to have access to the data in real time, and you're still not losing the humans in all of this, too, right?

Jamie: Right. Yep.

Khalipha: Exactly. So,o, then I guess, Jamie, could you then share a scenario where Ignite's AI native approach really transformed operations? So maybe we're talking guest discovery, maybe we're talking store performance or frontline decision-making, but any time where AI native approach really transformed operations, and what changed for the teams who are on the ground?

Jamie: Sure. I love that, and I, I actually have an example that kind of combines guest recovery, store [00:11:00] performance, and frontline decision-making, and it's really ... It, it's a very repeatable example that we think of as unified guest recovery. And one of the most important things in, in, in client experience is making sure that we're recovering dissatisfied guests, and we all know that those guests make the most loyal customers, right?

So with one of our clients, we actually did a pilot with a few of their locations where we combined our Comment Explorer and Knowledge Agent tools with alerting and action agents, and we were automatically able to identify dissatisfied guests from their open feedback, which we then were able to correlate with location and transaction details.

So from that, we were able to route suggested recovery action directly to the manager and give that manager some example verbatims from the feedback that was left and a proposed outreach script to those consumers. So the result of that was that the team really moved away from discovering [00:12:00] issues in weekly reports to more receiving the right priority and contextual recovery tasks in near real time.

So this, what this will do for the location then is cause their resolution times to fall, and the local teams in these locations actually reported that they were able to have clearer and faster coaching conversations with their staff. So the, the same pattern applies across store performance, frontline decisions.

Ignite is turning those noisy comments into ranked problems, and then showing those frontline managers real examples and proposing concrete fixes to them that, that then a human can approve and go execute. So it's a practical kind of 60-second insight to action, and, and really that, that line right there is the, is core to our product roadmap and how we're developing Ignite.

Ignite ad: Brand experience, customer experience, and employee experience aren't isolated, they're [00:13:00] interconnected. But too many organizations treat experience management like separate programs, which leads to scattered data, disconnected systems, and missed opportunities. That's why a unified approach matters, and that's why we built Ignite.

Ignite is SMG's purpose-built AI native platform, built to bring brand, customer, and employee experience together in one place. No silos, no lag, just complete real-time visibility across every touch point that matters. Ignite pulls in real-time feedback, transactions, third-party data, and behavioral signals, then processes it with advanced analytics and AI tools.

From sentiment shifts to location level performance, Ignite's AI agents analyze billions of data points, uncovering the why behind the what and telling you what to do next. And it doesn't stop at insights. [00:14:00] Ignite connects intelligence to action, automating tasks, tracking progress, and enabling every team to improve experiences in the moment.

When your experiences are unified, your impact multiplies. Happier customers, empowered employees, stronger business results. This is Ignite. Unified experience management, powered by AI, built by SMG.

Khalipha: Ah, amazing. Yeah, I'm guilty as charged of sometimes being one of those guests who if I don't get like a response, if I feel like my feedback is not being seen, I am out. Goodbye, cheerio. But it's also really cool to see that Ignite is a platform that organizations can use to move quicker, and kind of have more people like me who would typically ordinarily go away and be like, "Wait, hold on.

Actually, there's a solution here. There's something to be solved." That's really great. [00:15:00] Like guilty as charged, I'm somebody who as a guest, if I do not have a great experience, I will go away. I, I won't come back. But it's really great to hear that there are platforms like Ignite that can actually help those organizations bring people like me and so many others who would've ordinarily felt like their feedback, their comments aren't being heard.

Jamie: Mm-hmm.

And you can kind of close that loop in real time, quick time even. So yeah, I, I love that example that you gave, and I think hopefully our audience listen to what you say and make notes. If not, I heavily advise that you start doing that now. But Jamie, let's then talk about like the growing concern around AI, 'cause I love the fact that we have AI built into our platform.

It's helping us move, accelerate quicker. But naturally, there is this concern, there are these question marks around AI, and I think there's a bit of a concern around how AI potentially removes the human touch from customer experience. So how do you think AI native tech [00:16:00] allows teams to actually deliver more humanity and not less?

Jamie: Sure. That's also a great question. We, we spend a lot of time talking about that, and I would say AI native doesn't remove human judgment, it actually amplifies it. So Ignite is explicitly built around this idea that experience intelligence is added to human insight. So, the AI surfaces the patterns, it surfaces the root causes and suggested actions, but it's the humans that need to validate and adapt it, and then go execute it, right?

We often talk in product development about this concept of cognitive load, and what that means to us is the total mental effort for a person to gather and clean up kind of all of the disparate signals coming into from the customer experience standpoint, then look for correlations, interpret noisy open text feedback, and then from all of that, decide [00:17:00] what's important and figure out what to do next.

For most teams, that could mean hours or even days spent stitching together reports and hunting for evidence, and that time could be better spent on coaching and fixing problems or having real conversations with consumers and staff members. So I, I, I like to frame the shift as AI should remove all the time required for the, the workflow tasks, and then people should review the work and not do the work.

It's important because high cognitive load really slows decision velocity, and it also increases errors and inconsistency in how teams respond. And so this can often lead teams to default to ad hoc or really defensive behavior. So, for example, just reacting to a report instead of fixing root cause issues.

And so lowering that cognitive load is the difference between a, a weekly manual find the problem routine and more of a continuous [00:18:00] design and deliver practice where teams can act quickly and confidently. And the other thing is that our approach at SMG really keeps the AI decisions transparent, so people understand why a decision was made, and that builds trust and lets the team add the human judgment, empathy, and context before acting.

So again, AI provides speed, scale, humans provide insights, values, and care. Let me tie this off by saying reducing the cognitive load allows the frontline teams to spend less time on report wrangling and more time on customer conversations, coaching, and design, and those are all inherently human. 

Khalipha: Incredible. And I love that you ended that. It's still human at the end of the day. It's not about replacing the people, it's just literally just enhancing what we do and giving us...

Jamie: That's right.

Khalipha: ...humans time to focus on what really matters, while letting the AI kind of do the heavy lifting for us.

Jamie: Yep.

Khalipha: So Jamie, okay, sticking still to AI, right, and how as AI kind of becomes [00:19:00] embedded in community feedback, market research, frontline operations, what does this actually mean for unified experience management, and what does the next era of UXM look like?

Jamie: Oh, I like that. As I think about ... If we think about the era we're in today, I would say AI is more about better reports. The era we're moving to in unified experience management is more about orchestrated outcomes. So a unified data model, AI agents, letting the researchers, the frontline operators, and the product really speak the same language, so insights don't sit in a report, they trigger actions that actually move the business forward.

And so in experience management, when we think about it across brand, customer, and employee experience, it's shifting from a set of disconnected reporting functions to single outcome oriented systems that embed our experience design and experience [00:20:00] delivery ongoing feedback loop into our client's processes.

So, this really means, I think, three big things. The first is AI driven predictions that anticipate changes before they're escalated. The second thing is integrated validation, so testing product, marketing, process, training decisions, all before they're launched. And then the third thing is really faster, more impactful actions through embedded agents that keep the humans in the loop.

But the end result changes what unified experience management measures and how teams work. So instead of isolated KPIs and monthly slide decks, organizations are gaining orchestrated outcomes, so faster learning cycles, higher decision velocity, and measurable lifts in NPS conversion rates, retention rates, et cetera.

And this is all driven by coordinated actions that originate in our experience design, and then they're executed in our experience delivery. And, and again, I don't think it can be [00:21:00] said enough, it's still human centered, right? The agents accelerate and orchestrate, but humans are the ones validating, prioritizing, and applying values and empathy at the point of action.

Khalipha: Amazing. Incredible. Oh, Jamie, thank you. Th- and honestly, that, that really is music to my ears, because I think people get caught up in feeling like, "Oh my gosh, AI is taking over. Oh my gosh, it's going to replace this, and oh my gosh, are we be able to do anything because there's AI?" But I think you've shown how, here at SMG, for example, we've built something from the ground up that is AI native, but again, it's still human-centric.

It still needs people to validate. And so, like, I, I can't thank you enough for the insights you've been sharing with not just me, but our audience on this topic. And as we kind of close off the episode, we typically ask our guests to share their pearls of wisdom, and you've been sharing pearls. You've been sharing not just gems and pearls, you've been sharing diamonds with us.

Not just me, but our audience members. And so [00:22:00] with everything that we've discussed in this episode, AI, Ignite, experience management, and really the future of experiences, what would be your three takeaways that you'd like to leave our audience with?

Jamie: Amazing. I'm very passionate about this. So, um, three things I would have everyone take away from this episode.

Data is key. AI can only be optimal when it has a clean, unified source of truth. Um, the second thing is trustworthy AI reduces cognitive load and increases speed to action. So AI can surface the patterns and propose actions, and that gives people time back to exercise judgment and empathy. And lastly, grow your experience management strategy responsibly.

I don't think anybody out there is purposely being irresponsible, but I think it's incumbent upon us all to make sure that the tools are doing what you want them to do, [00:23:00] and the insights that they're surfacing are accurate and that humans are embedded in the loop. So I would just kind of tie it off with smarter AI plus cleaner data, plus humans in the loop really equals faster, more humane decisions that are actually gonna move businesses forward.

Khalipha: Amazing. Jamie, thank you so much. You've been such a lovely guest. I could totally go on and on and on to try get people to just continue to understand that AI is incredibly helpful. It's not something that's replacing human insight or the human work that still needs to be put in. It really is, it's something that enhances, and we can see that with Ignite and how it's been, it's been really, really valuable to clients, to the people who use it already.

So yeah, Jamie, thank you so much, and I'm sure we'll be back again. You're a fabulous guest.

Jamie: I hope so. I would love to come back. Thank you for letting me come in today and talk to you about the things I'm very passionate about.