Fullständig transkription av Salesforce Agentforce Heats Up the Consolidation Fight, UJET Says It Could Backfire

enterprises, but mid-market, S and B. 00:02:41.520 --> 00:02:50.879 I think everyone is looking for simplification, um, you know, said a little less diplomatically, like one throat to choke on the vendor relationship, right? 00:02:51.039 --> 00:02:56.719 Um, and it's it's not just technical in nature, it's it's this commercial necessity, right? 00:02:56.879 --> 00:03:09.120 Because there's so much um administrative and operational burden that has really come to define uh the the last decade or so of customer experience management. 00:03:09.199 --> 00:03:14.240 Because we have a new channel for our a new tool for every new workflow. 00:03:14.400 --> 00:03:18.000 Um our agents are managing, you know, four to ten different systems. 00:03:18.159 --> 00:03:22.400 We've got this sort of swivel chair effect, IT's involved in all of this. 00:03:22.719 --> 00:03:35.360 Um, and so like I said, you know, I I think we agree that um, you know, AI is sort of the mandate to finally bridge some of these silos and start to drive some form of consolidation for us. 00:03:35.439 --> 00:03:42.000 Um but I think what you and I are going to talk about today is is you know what's driving it is the goal is simplification. 00:03:42.319 --> 00:03:54.159 Um but I think you know that comes down to actually addressing the underlying architecture as opposed to simply trying to get you know a unified bill and and one number to call for support, right? 00:03:56.240 --> 00:04:10.879 Okay, I I like that framing a lot, and it helps me actually because we as an industry we've had lots and lots of decades of new technology that's come out that was gonna fundamentally transform everything that we do. 00:04:11.039 --> 00:04:13.680 And and you know, they never quite live up to their promises. 00:04:13.840 --> 00:04:27.680 I think AI might be a little different because it's proving to kind of definitely wash its own face, but when we do this, we've uh we we don't seem to have learnt our lessons of just let's go and scramble to get the new shiny piece of technology. 00:04:27.839 --> 00:04:35.839 Um, let's just kind of like rush out and and get these things, make these things happen without the right sort of planning and everything beforehand. 00:04:35.920 --> 00:04:44.560 So we whilst we have got as an industry uh a history of adapting and adopting technology, we don't always do it very well. 00:04:44.720 --> 00:04:51.600 So it's quite interesting that now that we're in the place where actually we've got all this technology, what do we do with it? 00:04:51.759 --> 00:04:53.199 How do we make the most of it? 00:04:53.279 --> 00:05:01.360 And and it does feel like AI really is that kind of it's the missing piece of that jigsaw that we've had for so long. 00:05:01.600 --> 00:05:06.079 It's the gap that we've, you know, the reason we've been unable to do it is because we didn't have this kind of tool. 00:05:06.319 --> 00:05:08.319 Now that we do, that's great. 00:05:08.639 --> 00:05:21.519 But I guess from your perspective, what are the organizations feeling now when there's pressure to actually rely on all of these uh single massive ecosystems that they've got? 00:05:22.079 --> 00:05:28.079 Where does that impact them so much now that we're looking at this through the day, you know, the light of today? 00:05:28.399 --> 00:05:31.279 Yeah, well, look, I I think I think you nailed it, right? 00:05:31.360 --> 00:05:39.839 We've done a really good job of accumulating technology, not necessarily sort of adopting or optimizing it, right? 00:05:40.000 --> 00:05:54.959 And so, you know, what we see, and we have so many of these conversations, um, you know, what organizations are are the way that you see this pressure around the ecosystem is is they're they're showing up like in a hostage dynamic, right? 00:05:55.120 --> 00:06:05.439 They've got commercial lock-in, they've got vendor lock-in, and these huge switching costs that really sort of dictate their overall operational roadmap. 00:06:05.680 --> 00:06:17.759 Um we tried to migrate our our CRM for three years, um, but we had something like 30 different tools and operational processes across departments that were all sort of integrated in there. 00:06:17.839 --> 00:06:24.800 And it made the switching costs so high that we constantly just said, okay, we'll kind of pay the increase this year, right? 00:06:24.959 --> 00:06:31.600 And so, you know, in these ecosystems, it what really drove them was deep integration. 00:06:31.759 --> 00:06:38.319 It was easy to buy and deploy systems that would integrate, and that was going to be what drove simplicity, right? 00:06:38.399 --> 00:06:42.079 But what it's done is it's it's made the cost of any kind of change prohibitive, right? 00:06:42.160 --> 00:06:43.040 You get stuck. 00:06:43.279 --> 00:06:51.920 So every system's interdependent, and if you replace one piece of that, you get like these cascading failures across the entire stack. 00:06:52.160 --> 00:06:58.000 Um, and so you know, you just sort of keep renewing and going with the old systems. 00:06:58.399 --> 00:07:02.879 Um But, you know, look, I'll say we're talking about Salesforce today. 00:07:03.120 --> 00:07:14.959 I think 70%, something like 70% of their new revenue growth last year came from price increases alone rather than like net new sales, right? 00:07:15.120 --> 00:07:19.040 And that just shows you that like organizations are showing up stuck. 00:07:19.199 --> 00:07:35.920 And so what they're looking at is um how do we get beyond our main platforms, existing commercial structures, or by the way, like other partnerships that they have that might influence how they direct what we buy and implement, right? 00:07:36.480 --> 00:07:47.759 Um to really figure out how you get to not the preferred solution that's being recommended, but the one that's the best architectural fit for your use cases. 00:07:51.680 --> 00:08:14.800 So I think you're yeah, you've you've really nailed the point

there with the what we could call the integrations trap, which you know the promises of integration for so long was you know, uh everyone had pages on their websites to chat about how many different integrations they had, and it was a big selling point for everyone to say, Oh, look, we can all plug into each other, isn't that phenomenal? 00:08:14.879 --> 00:08:15.680 But you're absolutely right. 00:08:15.759 --> 00:08:21.680 We've actually caused ourselves to kind of land stuck amongst that like a cob website, trying to get out, I guess. 00:08:21.920 --> 00:08:23.839 Um interesting though. 00:08:24.079 --> 00:08:31.920 So, what is the consequence of this for you know the real teams who are in there who are trying to stay agile? 00:08:32.080 --> 00:08:40.240 Um, but the operation is kind of often tied up to probably one of the bigger vendors roadmap and all those other things integrated into it. 00:08:40.320 --> 00:08:43.120 What what how that where does that play out for those teams? 00:08:43.200 --> 00:08:49.360 Because they're the guys looking after our customers, so we need to make sure that that is uh something that we're aware of. 00:08:49.759 --> 00:08:54.960 Yeah, look, I I think you'll know this firsthand from your history, kind of operating contact centers, right? 00:08:55.039 --> 00:09:00.480 And I I sort of touched on this, but you get a you get a massive sort of IT burden, right? 00:09:00.639 --> 00:09:04.399 You've got administrative debt with kind of procurement and everything else. 00:09:04.559 --> 00:09:11.759 And then the the real burden ends up being on our our our agents and the customers they're serving, right? 00:09:11.840 --> 00:09:19.600 Because the human contact center agent is is effectively become the manual integration layer between all these disconnected systems, right? 00:09:20.000 --> 00:09:27.679 And so, you know, we we got here because we're gonna get you know a full suite, you'll get the convenience of a full suite, right? 00:09:27.759 --> 00:09:34.480 But it's a little bit of a marketing illusion that that falls apart as soon as you see an agent's desktop, right? 00:09:34.639 --> 00:09:36.559 Four to ten systems open. 00:09:36.720 --> 00:09:40.240 Um, and and you know, give me just a minute, I'll pull that up for you, right? 00:09:40.399 --> 00:09:49.679 All of that stuff that we that we experience, even just as consumers in our daily lives, um, because we're not able to actually do anything about the architecture itself, right? 00:09:49.759 --> 00:09:59.840 So I think we have to challenge this strategy of treating the contact center as a secondary feature to the CRM, right? 00:10:00.000 --> 00:10:02.240 Um I don't think you can stay agile and innovate. 00:10:02.320 --> 00:10:08.559 And by the way, like you know, the AI, like we said, it's moving so quickly, you've got to maintain some agility. 00:10:08.799 --> 00:10:30.960 But if that is tethered to a major vendor's you know, twice annual release cycle or the specific technical debt or acquisitions they've made on the back end that informs their data schema, you know, all of those things are gonna really, really hamper uh your ability to be uh agile and to innovate and to adopt new tools. 00:10:31.120 --> 00:10:39.519 Um and so I mean, I think one of the questions you'd have to ask right now is like, why are the CRM vendors pivoting so aggressively into CX, right? 00:10:39.600 --> 00:10:51.919 Is it about leveraging the data, or is it some form of a you know defensive pivot away from sort of the the existing system of record strategy that they've they've built over the last couple of decades? 00:10:53.360 --> 00:10:56.240 I hope that it's maybe a bit a combination of both. 00:10:56.399 --> 00:11:00.159 I'm gonna I'm gonna sit firmly on that fence right now. 00:11:00.320 --> 00:11:20.879 Um because I do see you know the benefits of the context that CRM brings, especially from an AI perspective, if you want to build agency systems that can you know carry out automated workflows, that that context is super important, and um, and I know you can move it around, but I think I think that there's certainly value in that. 00:11:21.039 --> 00:11:22.320 But you're absolutely right. 00:11:22.480 --> 00:11:32.480 There is a almost a trend at the moment of this, um, and and it's only a reflection of the kind of very fast-moving space at the moment, isn't it? 00:11:32.639 --> 00:11:40.399 But um, but look, I'm gonna say this you you use the word context, and context is an expression of data. 00:11:40.720 --> 00:11:46.480 Typically, what we see in the CRMs is a historical repository, these are static systems of record. 00:11:46.720 --> 00:12:00.720 So when you call in because you wanna, you know, you've got I'm I'm a major brand and I sell electronics, and and you're trying to deal with a support issue on one product line. 00:12:00.960 --> 00:12:06.639 I don't actually need to know as an agent what your five-year purchase history is. 00:12:06.799 --> 00:12:08.320 I don't need all this information overload. 00:12:08.559 --> 00:12:09.279 That's data. 00:12:09.600 --> 00:12:19.519 Context is that you were on the website 30 seconds ago trying to solve the problem with a virtual agent that couldn't resolve it. 00:12:19.600 --> 00:12:22.320 And so you gave up on it and you called into me. 00:12:22.399 --> 00:12:27.519 And I, and and oh, by the way, the agent assist tool that I'm using isn't starting over from scratch. 00:12:27.679 --> 00:12:32.159 It has the context of that conversation and is informing the human agent. 00:12:32.399 --> 00:12:37.039 Um, I just I think context is really, really pivotal here. 00:12:37.200 --> 00:12:45.759 And there are technical limitations to how quickly you can surface real-time information within those existing systems of record. 00:12:46.000 --> 00:12:51.360 I I think you're gonna see more of a shift in the data strategy overall from CRM to like a CDP. 00:12:52.559 --> 00:12:57.519

That's really interesting because that, I mean, as we know, data's the kind of fundamental foundation of all of this. 00:12:57.679 --> 00:13:02.159 These things, you know, without the right data is your foundation, none of this stuff works well. 00:13:02.240 --> 00:13:04.639 So you've hit a really important point there. 00:13:04.799 --> 00:13:15.440 And it's interesting because I hadn't thought about it in terms of the recent nature of the kind of the context and how important that is for that particular interaction. 00:13:15.679 --> 00:13:19.039 Um, so yeah, so I you you might have swayed me there. 00:13:19.120 --> 00:13:31.840 So I I want to get back to this uh the set the Salesforce launch a little bit more because technically um we should look at this because we've looked at the human element, we've looked at kind of how this impacts our people. 00:13:32.000 --> 00:13:40.639 Um let's look at it from a technical, from an architectural standpoint, say, what is like the harder problem to solve today? 00:13:40.879 --> 00:13:51.360 Um, and that I say that by saying that I guess beyond just adding voice to a CRM, which might not necessarily like to your argument, might not be the best thing to do. 00:13:51.919 --> 00:14:04.559 Yeah, um I I could go 30 minutes on on this topic, but I'll start by saying you can't fix a structural or what we're talking about an architectural problem with a feature. 00:14:04.799 --> 00:14:12.480 Um, you know, adding voice to that historical repository doesn't fix the foundation or or the friction points. 00:14:12.559 --> 00:14:15.519 It really just changes kind of the line item on the bill, right? 00:14:15.679 --> 00:14:21.200 Now, I will say the the telephony is a complex technical problem in its own right. 00:14:21.360 --> 00:14:28.080 You know, you've got real-time resiliency between carriers, you've got latency, you've got you know, global presence. 00:14:28.240 --> 00:14:39.440 And, you know, I think that's why CRM vendors have been happy to partner and have avoided it, you know, for the last 20 years, um, while at the same time actually predicting that voice would die, right? 00:14:39.519 --> 00:14:41.519 Digital channels are going to make voice obsolete. 00:14:41.600 --> 00:14:53.519 I mean, I think Salesforce actually championed sort of the death of the phone over the last five years in an attempt to sell the digital channels and AI, but now suddenly they're they're offering native voice, right? 00:14:53.600 --> 00:15:04.480 So is it an attempt to solve customer experience again, or is it trying to shore up um you know, sort of a gap in the in the go-to-market strategy? 00:15:04.639 --> 00:15:15.279 Um, I think the the real transformation is not about adding voice or even having a seamless handoff between digital channels and voice. 00:15:15.440 --> 00:15:21.519 Um and by the way, voice could be AI as well, or it could be human, or it could be some combination. 00:15:21.600 --> 00:15:27.679 But I think the real

transformational challenge that enterprises have to look at is an orchestration layer, right? 00:15:27.759 --> 00:15:48.159 Um what we have to look at is not uh sort of vendor consolidation, but but a new blueprint for uh enterprise systems to to deliver against a value chain in real time, um, as opposed to kind of continuing to add more features into uh the existing stack. 00:15:50.399 --> 00:15:53.360 It's it it's certainly an exciting period of time. 00:15:53.440 --> 00:15:54.399 You you're absolutely right. 00:15:54.480 --> 00:15:58.960 I I uh the whole um voice is dead argument. 00:15:59.120 --> 00:16:07.679 I think that I think that drum was beaten by the same people who've said that agents won't have a job either, um, which I think we all know is is not the case. 00:16:07.759 --> 00:16:09.519 It's just an evolution of everything, right? 00:16:09.679 --> 00:16:20.879 So uh and we've been here before, email was gonna finish contact centres, social media was gonna finish contact centres, and hey, we're still here, so um, yeah, I think uh I think you're I think you're absolutely right there. 00:16:21.039 --> 00:16:34.799 Um I guess then when uh when you're speaking to the companies that you speak to, um when those enterprises are looking at the kind of AI strategy, so let's get into the kind of the kind of crux of that. 00:16:34.960 --> 00:16:53.840 What what evidence um or or early signals can you share that might convince uh uh an operation that's possibly a bit cautious in terms of purchasing anything that you know centralizing everything might actually slow down their innovation opportunities as well? 00:16:54.320 --> 00:17:04.400 Yeah, look, I think the first thing is I don't enter into these conversations with our customers trying to convince them of a specific strategy. 00:17:04.720 --> 00:17:12.160 Um, what I want to do is sort of, you know, we talk about in product, like fall in love with the problem, like understand what their actual challenges are. 00:17:12.319 --> 00:17:22.960 Um but what we do see with sort of the all-in-one or the centralized approach, and and you and I were talking about this a little bit before we turned the mics on, right? 00:17:23.359 --> 00:17:38.160 Is that we we've seen an erosion as you start to put all of these different vertical solutions within a single enterprise vendor, um, you see an erosion of expertise and and reliability, right? 00:17:38.319 --> 00:17:56.160 So you can't have uh a single global outage, by the way, that that you know, with a database or cloud infrastructure that brings an entire install base to a standstill, right? 00:17:56.640 --> 00:18:28.559 Um and you know, we see across across sales and support that if you're in a if you're in a larger sort of enterprise vendor motion, very often what you lose is the operational expertise required to understand your use case and and to help you sort of design and deploy and maintain the right compliant solutions for your industry, your vertical, your region, right? 00:18:28.880 --> 00:18:34.240 Um and and and I'll go back to sort of the vendor lock-in challenges, right? 00:18:34.319 --> 00:18:44.640 Like uh there's a there's a renewal tax that just happens inevitably the more you're anchored in and tethered uh to these sort of single systems, right? 00:18:44.720 --> 00:18:57.839 So I just don't think that you can build really a future-proof strategy um when you're working with with folks who who aren't experts in your specific uh industry and market. 00:18:59.680 --> 00:19:00.720 Renewal tax. 00:19:01.039 --> 00:19:02.720 I yeah, I love that phrase. 00:19:02.880 --> 00:19:03.839 That's so right. 00:19:03.920 --> 00:19:13.519 Uh it's um I can you know reflect on uh too many uh occasions where you've gone to try and drive some improvements within an operation. 00:19:13.839 --> 00:19:35.680 There's an obvious route, an obvious solution to a problem, uh, but actually there's a there's a there's a there's a there's an issue in terms of either the cost of putting that solution in play or the cost of getting the data, that's always been a mystery to me why it's so difficult for you know a company to share the their own customers' data with them. 00:19:35.839 --> 00:19:48.319 So before we close, because we're running out of time, I could likewise I could talk about this all day, but uh if uh if you could offer just one piece of advice to our audience watching this week, uh what would that one piece of advice be? 00:19:48.799 --> 00:19:55.200 Yeah, look, I'll I'll start with it, it's a it's the first, it's the it's how I'll answer your question. 00:19:55.279 --> 00:20:05.680 It's also a response to everything you just said, which is you know, the the most important aspect of AI and customer experience is the customer experience, you know? 00:20:05.920 --> 00:20:09.359 And and when you have a hammer, everything looks like a nail. 00:20:09.440 --> 00:20:12.799 But you know, AI is not going to be this panacea or cure-all. 00:20:13.200 --> 00:20:22.400 Um, but it's going to it's going to fundamentally alter over time how we operate, not only as organizations, but as individuals, right? 00:20:22.480 --> 00:20:35.119 And and that includes a transformation or a consolidation of existing roles and responsibilities, the departmental structure of our organizations, and really like totally new blueprints for organizational design. 00:20:35.279 --> 00:20:50.880 So if you go look at like the two and three-year-old companies today that have scaled to$50 or$100 million like overnight with 10 or 15 people, they have a totally different approach to value chain delivery, right? 00:20:50.960 --> 00:20:58.559 They weren't iterating on legacy institutional concepts of marketing and sales and support and IT, right? 00:20:58.799 --> 00:21:07.039 They just looked at, you know, what's the pain that we're solving for customers and what's the shortest path to solving it for them, right? 00:21:07.200 --> 00:21:17.200 So, you know, my advice on all of this is don't start with a false paradigm debate of all in one versus best of breed, right? 00:21:17.359 --> 00:21:29.119 Start with a clean sheet of paper and imagine no technical or process debt and design for today's reality and then figure out what the right path is for you to get there, right? 00:21:29.279 --> 00:21:38.559 And and what I bet is that if you actually do that, that you know the conclusion you arrive at is not going to look anything like today's status quo. 00:21:39.839 --> 00:21:53.519 You've hit the nail on the head for me that it really is all about um not having all the legacy uh noise uh clogging up your perspective and and and almost dictating what you do. 00:21:53.680 --> 00:22:00.000 It's about uh ignore all that, that blank sheet of paper, think it can be such a powerful thing. 00:22:00.079 --> 00:22:01.839 Uh and yeah, you're absolutely right. 00:22:01.920 --> 00:22:11.759 There's so much evidence, recent evidence of organizations who who don't have that baggage, who have absolutely thrived and and you know flown. 00:22:11.920 --> 00:22:14.720 So no, I think that's a really, really, really great point. 00:22:14.960 --> 00:22:17.440 Unfortunately, though, that is all we've got time for. 00:22:17.599 --> 00:22:21.759 Um so Baker, thank you so much for joining me and answering all these questions. 00:22:21.839 --> 00:22:30.319 Um before we close, though, just um anyone watching this and who wants to explore this subject uh in more detail or maybe kind of get in contact with you and the team. 00:22:30.400 --> 00:22:32.640 What's what's the best way for them to do that? 00:22:33.359 --> 00:22:36.640 Well, you can find us pretty easily at ujet.cx. 00:22:38.000 --> 00:22:38.960 Simple as that. 00:22:39.039 --> 00:22:46.880 Uh, and also uh you can find our wealth of related resources, stories, and other videos just like this one at cxtoday.com. 00:22:46.960 --> 00:22:48.640 So that wraps up things for today. 00:22:48.720 --> 00:22:50.000 I've been Rob Wilkinson. 00:22:50.079 --> 00:22:51.279 This is CX Today. 00:22:51.359 --> 00:22:53.119 Thanks everyone for joining us.