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

 Salesforce’s launch of the Agentforce Contact Center has reignited a familiar debate in customer experience: should enterprises consolidate around one major platform, or keep the flexibility to pick best fit tools.

In this CX Today interview, Rob Wilkinson sits down with Baker Johnson, Chief Business Officer at UJET, to unpack what he calls the integration trap. They explore why many organizations feel “stuck” in CRM ecosystems, how switching costs and interdependent integrations slow down change, and why agents often become the human integration layer across four to 10 disconnected systems.

Baker also challenges common assumptions about “context” in AI driven customer service, arguing that real value comes from real time journey context, not just historical CRM records. The conversation closes with practical advice for CX leaders building AI strategies: focus on the customer experience first, and treat AI as a forcing function to simplify operational complexity, not a cure all. 

Transcript

SPEAKER_01 0:08

Hello and welcome. I'm Rob Wilkinson, and today we're reflecting on the relatively recent news of Salesforce's launch of their Agent Force Contact Center. And we're going to have a look at what this push for platform consolidation might mean for enterprise flexibility. The jury's still out as to whether this is positive or negative, I think. We've done some recent conversations with analysts that shows that I think the um the kind of the industry is divided. But if you're an operations uh uh manager and you are challenged with managing the operational gravity of your uh CRM ecosystem, or if you're exploring how you want to truly operationalize your data for AI, of course, then stay with us because we're gonna get practical takeaway today that you can take away and use in your organization this quarter. Uh, because today I'm joined by Baker Johnson, he's the chief business officer at UJet. He's definitely an expert in this space. Um, he sees um an awful lot of companies looking at these decisions every single day. Uh, and he's got uh he's got an alternate or a contrarian perspective uh on what actually this launch could mean for CX leaders. So it's really really really great to actually get an alternative viewpoint. And uh yeah, welcome Baker. Thank you so much for joining me.

SPEAKER_00 1:40

Hey, thanks, Rob. Great to meet you and great to be here. Excited to be back.

SPEAKER_01 1:46

So before we dive into what's quite a meaty topic, um, let's sort of set the scene a little bit if we can. Can you give us a bit of a background as to kind of this debate that's happening right now around vendor consolidation and why is it so important?

SPEAKER_00 2:01

Yeah, sure. Look, I think first off, in terms of like Salesforce and their their announcement, I'll say we we do fundamentally sort of agree on the problem in the market, right? That there's this um this fragmentation between systems, between the system of record and the system of engagement, um, and that we do need to figure out some way to bring all that together. I think we have a, you've mentioned contrarian, right? We have a fundamentally different view of what that might look like. But you know, the question around vendor consolidation itself and and sort of what's driving it, right? I think I'll say enterprises, but mid-market, S and B. 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? Um, and it's it's not just technical in nature, it's it's this commercial necessity, right? 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. Because we have a new channel for our a new tool for every new workflow. Um our agents are managing, you know, four to ten different systems. We've got this sort of swivel chair effect, IT's involved in all of this. 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. 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. 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?

SPEAKER_01 3:56

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. And and you know, they never quite live up to their promises. 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. 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. 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. 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? How do we make the most of it? 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. 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. Now that we do, that's great. 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? Where does that impact them so much now that we're looking at this through the day, you know, the light of today?

SPEAKER_00 5:28

Yeah, well, look, I I think I think you nailed it, right? We've done a really good job of accumulating technology, not necessarily sort of adopting or optimizing it, right? 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? 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. 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. And it made the switching costs so high that we constantly just said, okay, we'll kind of pay the increase this year, right? And so, you know, in these ecosystems, it what really drove them was deep integration. It was easy to buy and deploy systems that would integrate, and that was going to be what drove simplicity, right? But what it's done is it's it's made the cost of any kind of change prohibitive, right? You get stuck. So every system's interdependent, and if you replace one piece of that, you get like these cascading failures across the entire stack. Um, and so you know, you just sort of keep renewing and going with the old systems. Um But, you know, look, I'll say we're talking about Salesforce today. 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? And that just shows you that like organizations are showing up stuck. 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? 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.

SPEAKER_01 7:51

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? But you're absolutely right. We've actually caused ourselves to kind of land stuck amongst that like a cob website, trying to get out, I guess. Um interesting though. So, what is the consequence of this for you know the real teams who are in there who are trying to stay agile? 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. What what how that where does that play out for those teams? 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.

SPEAKER_00 8:49

Yeah, look, I I think you'll know this firsthand from your history, kind of operating contact centers, right? And I I sort of touched on this, but you get a you get a massive sort of IT burden, right? You've got administrative debt with kind of procurement and everything else. And then the the real burden ends up being on our our our agents and the customers they're serving, right? Because the human contact center agent is is effectively become the manual integration layer between all these disconnected systems, right? 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? But it's a little bit of a marketing illusion that that falls apart as soon as you see an agent's desktop, right? Four to ten systems open. Um, and and you know, give me just a minute, I'll pull that up for you, right? 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? So I think we have to challenge this strategy of treating the contact center as a secondary feature to the CRM, right? Um I don't think you can stay agile and innovate. And by the way, like you know, the AI, like we said, it's moving so quickly, you've got to maintain some agility. 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. 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? 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?

SPEAKER_01 10:53

I hope that it's maybe a bit a combination of both. I'm gonna I'm gonna sit firmly on that fence right now. 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. But you're absolutely right. 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?

SPEAKER_00 11:32

But um, but look, I'm gonna say this you you use the word context, and context is an expression of data. Typically, what we see in the CRMs is a historical repository, these are static systems of record. 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. I don't actually need to know as an agent what your five-year purchase history is. I don't need all this information overload. That's data. 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. And so you gave up on it and you called into me. And I, and and oh, by the way, the agent assist tool that I'm using isn't starting over from scratch. It has the context of that conversation and is informing the human agent. Um, I just I think context is really, really pivotal here. And there are technical limitations to how quickly you can surface real-time information within those existing systems of record. I I think you're gonna see more of a shift in the data strategy overall from CRM to like a CDP.

SPEAKER_01 12:52

That's really interesting because that, I mean, as we know, data's the kind of fundamental foundation of all of this. These things, you know, without the right data is your foundation, none of this stuff works well. So you've hit a really important point there. 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. Um, so yeah, so I you you might have swayed me there. 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. Um let's look at it from a technical, from an architectural standpoint, say, what is like the harder problem to solve today? 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.

SPEAKER_00 13:51

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. Um, you know, adding voice to that historical repository doesn't fix the foundation or or the friction points. It really just changes kind of the line item on the bill, right? Now, I will say the the telephony is a complex technical problem in its own right. You know, you've got real-time resiliency between carriers, you've got latency, you've got you know, global presence. 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? Digital channels are going to make voice obsolete. 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? 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? Um, I think the the real transformation is not about adding voice or even having a seamless handoff between digital channels and voice. Um and by the way, voice could be AI as well, or it could be human, or it could be some combination. But I think the real transformational challenge that enterprises have to look at is an orchestration layer, right? 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.

SPEAKER_01 15:50

It's it it's certainly an exciting period of time. You you're absolutely right. I I uh the whole um voice is dead argument. 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. It's just an evolution of everything, right? 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. 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. 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?

SPEAKER_00 16:54

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. 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. 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? 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? 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? 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? Um and and and I'll go back to sort of the vendor lock-in challenges, right? 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? 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.

SPEAKER_01 18:59

Renewal tax. I yeah, I love that phrase. That's so right. 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. 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. 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?

SPEAKER_00 19:48

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. 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? And and when you have a hammer, everything looks like a nail. But you know, AI is not going to be this panacea or cure-all. 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? 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. 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? They weren't iterating on legacy institutional concepts of marketing and sales and support and IT, right? 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? 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? 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? 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.

SPEAKER_01 21:39

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. It's about uh ignore all that, that blank sheet of paper, think it can be such a powerful thing. Uh and yeah, you're absolutely right. 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. So no, I think that's a really, really, really great point. Unfortunately, though, that is all we've got time for. Um so Baker, thank you so much for joining me and answering all these questions. 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. What's what's the best way for them to do that?

SPEAKER_00 22:33

Well, you can find us pretty easily at ujet.cx.

SPEAKER_01 22:38

Simple as that. Uh, and also uh you can find our wealth of related resources, stories, and other videos just like this one at cxtoday.com. So that wraps up things for today. I've been Rob Wilkinson. This is CX Today. Thanks everyone for joining us.