Author transcript
Rob Wilkinson: Hello and welcome. I’m Rob Wilkinson at CX Today, and today we’re taking a closer look at why CRM is shifting from customer record to customer resolution. If you’re challenged with disconnected customer journeys, or you’re exploring how to connect sales, service and operations to deliver better outcomes, then stay with us, because today I’m joined by Daniel Wilks, Vice President of CRM, EMEA at ServiceNow.
Daniel works closely with organizations looking at exactly how CRM, workflows and AI can support more connected customer experiences, and his perspective is especially timely as enterprises start to rethink what effective customer service actually needs to deliver. Welcome, Daniel. Thanks for joining me.
Daniel Wilks: Morning, Rob. Thanks for inviting me. Great to be here.
Rob Wilkinson: Really excited to get stuck into this one today. Just for a little context for our audience, Daniel: when you speak to organizations about CRM these days, what feels different about the conversation compared to even a couple of years ago? Because things have changed quite a lot.
Daniel Wilks: When I talk to customers, I tend to see the same themes coming up consistently. A lot of that is around the pace of the environment we’re in at the moment and the challenges that generates. The business processes used to manage the customer lifecycle are fundamentally broken. There are too many different applications and systems. Teams spend too much time chasing down information and not enough time actually working with customers.
We know the C-suite are really being pushed to deploy AI as a way to start doing more with less, and that really applies to the customer lens and customer service. We know AI blows the lid off a lot of use cases in customer service, while at the same time helping to reduce costs. Most organizations I speak to are experimenting with AI in some way, but maybe aren’t yet seeing the true return on investment. So how do they start showing results quickly?
And finally, there’s growing frustration with legacy CRM systems, but often little appetite to replace them, because customers don’t believe there’s a viable alternative and they’re looking at the perceived complexity and lift involved. So those are some of the themes I see.
Rob Wilkinson: It’s a really valid point. This year, people have been talking about “Frankenstacks” everywhere – massive amounts of legacy tech bundled on top of each other. That’s the starting point. Then you factor in the reluctance of some vendors to enable companies to use their own data, and the way they tie them into all sorts of things that stop them from moving around and doing what they technically could.
In addition to that, where do the leaders you’re talking to every week feel the pressure most? Because it’s got to be where customer data, service teams and operational workflows aren’t connected properly – so you’re getting that fragmentation as well as legacy tech debt.
Daniel Wilks: Absolutely. Over the last 20 years or so, companies have poured billions of pounds into customer relationship management and CRM systems, but the results have been frustrating. The promise was there: build an omnichannel intake and a 360-degree view of your customer, and you’ll drive better business and customer outcomes. But going back to your point about fragmentation, those systems are typically siloed. They’re overly customized and too manual to operate in a way that really delivers an engaging customer experience and operational efficiency.
If we look at some examples, sales representatives have basically become data clerks. They’re spending more time entering data into systems and navigating complex configure, price, quote tools than they are interacting with customers. Customer service reps are operating as human middleware. They’re toggling and swivel-chairing between three to five different applications to resolve a customer query.
And that leaves customers frustrated. We’ve all felt that as customers ourselves. We expect a fast, connected interaction, especially in the age of AI, but many of us end up bouncing between teams and channels just to get what should be a relatively straightforward answer.
The problem is that those legacy CRM systems were built to collect and store customer data. They were basically built as walled gardens, and the reality is those systems might know everything about your customer but can’t actually do anything about it. In that scenario, our teams are serving the system rather than serving the customer, and we need to find a way of delivering in the age of AI.
Rob Wilkinson: It’s like a self-fulfilling thing with those walled gardens, isn’t it? You can see the benefit, but if you want to improve on it, it’s really tricky and difficult to do. And I think you’ve been very kind to say people are only using three or four systems. [laughs] I think that’s a very underestimated figure, so fair play.
But moving away from what that’s like for the agent, what does that fragmentation look like for customers themselves? They’re going to be feeling that pressure. What does that look like in reality, and how is it exacerbating things for the people trying to resolve issues inside the business?
Daniel Wilks: I might contextualize that a little before giving a practical example. If we think about the age of generative AI and agentic AI that so many of our customers are looking to utilize, you can’t just bolt an AI model onto a legacy CRM system, or a set of systems on a broken foundation, and expect it to magically change everything.
The customer’s request will often start in the front office, at the interaction point. But completing the request requires access to many different systems across the enterprise. As you touched on a moment ago, the average enterprise has over 360 different applications. So many of the organizations I’m talking to are finding that AI alone struggles to operate on top of a CRM system that might be built on brittle custom code and fragmented systems.
If you have an AI copilot sitting in one of your applications, that’s great, but it doesn’t help your customer service agent resolve a task that might require spanning many different systems and applications to get the job done. Conversational AI can chat with a customer and answer questions – that’s the probabilistic intelligence that sits within those models. But very few organizations so far have been able to deploy AI in a way that actually drives resolutions or progresses situations for customers.
Take a customer who asks to change an order, for example. Completing that request isn’t a simple task of interacting with a chatbot. It requires a mechanism to locate the original order record in a system, run a configure, price, quote process to validate the configuration and determine pricing, check inventory levels and shipping timelines and cost that up, and then push order modifications through to downstream systems like the ERP. So what looks like a simple request actually has many dependencies across different systems. If you’re the customer and you ask the question and are given some advice without a resolution, you’re left frustrated.
And the same applies beyond the customer lens, to the people who actively work with customers to help them – customer service agents and sales teams. On average, they spend less than half their day actually helping customers, because they have to toggle between these different systems and act as a manual bridge. Where you’ve got those manual handoffs, that’s where processes break and customers are left frustrated.
So the opportunity in this domain is to enable customer service teams and sales teams to partner with AI. If we do that, we allow them to focus on true customer service, decrease the time spent on routine tasks, and resolve issues faster. That’s about automation and technology amplifying the customer experience.
Rob Wilkinson: “Amplifying” is the key word, and that’s what we get confused about. I think we’ve done this before with omnichannel – we basically thought it was going to make everything amazing without getting the foundations right, to your point. So we need to learn the lessons of the past and make sure we’ve got our ducks in a row: get those foundations in place, not just data, but also sort out the fragmentation of systems so you can actually improve. AI is not a magic bullet, and people need to realize that if you just put it on top of weak foundations, it’ll make things ten times worse. So it’s a really good shout.
When you’re dealing with organizations, is there a moment that makes them go, “We’ve got to do something now”? When do they realize that CRM isn’t just a place to store information anymore, and that it can become part of how work gets delivered and outcomes get achieved?
Daniel Wilks: I’m not sure there’s a single trigger. Most organizations are measuring customer satisfaction in some way, using CSAT scores to understand how customers feel about their products, their services and their interactions. So there are data points in there.
We also look at things like preventable customer losses. We know it’s much easier to sell to an existing customer than to acquire new ones. So if organizations can spot that they have customer retention issues or high churn, that’s another leading indicator. That can be at the macro level, or it can relate to one specific customer. If a customer churns or nearly churns, and the review reveals that teams didn’t know there was an issue – there were support tickets open, or the operations team didn’t know it was a high-value customer – and by the time anyone found out, the customer had already gone, I think that’s the moment organizations realize they could and should have saved the customer, but didn’t have the information or the processes to see and address the problem. And that changes everything. Suddenly CRM isn’t just a sales tool, it’s a retention tool, and it has to connect to operations across the organization.
The other thing is that there’s a ceiling for manual workarounds. I talked earlier about customer service teams acting as human middleware, swivel-chairing between different systems. That might work when you have a small number of customers, but once you reach any scale, that fragmentation and individuals performing heroics just doesn’t work anymore. You go beyond the point where operations teams can be emailing updates and support teams can be manually forwarding tickets on to success teams.
When organizations reach scale and those manual workarounds become excessive, that’s where the CFO gets involved, because there’s an opportunity to drive far greater efficiency. The two things go hand in hand: you can deliver an exceptional customer experience, but do so in a way that drives efficiency and takes cost out of the equation. That’s typically a tipping point for organizations.
Rob Wilkinson: I can see that playing out. And your point about context being king is really important for people to remember. CRM is where the context is, but it needs to be freed up – visible, transparent and able to be acted on by everyone in the business, so you can drive the efficiencies you’re talking about. Otherwise you just replicate the struggle.
So we’ve talked about the business side. Customers want resolution, but teams are working across very disconnected systems and handoffs. Let’s look now at the technology and operating models. Because the difficult part isn’t just knowing the customer – it’s connecting that context and knowledge to the action you want to deliver. How do we do that?
Daniel Wilks: Most organizations tend to start with data integration, and we talk about data integration a lot ourselves at ServiceNow. For an AI model or a system to deliver, it needs access to the data. But starting with the data is probably the wrong starting point. If you connect all the data and put it in one place to create a unified customer view, it doesn’t change anything operationally. It just becomes a data store.
I think the correct sequence is to start with the decision-making that needs to happen. That means defining the process: what decisions need to happen, and when, in the context of your customer processes. If a customer is a churn risk, who needs to know, what do they need to do about it, and when do they need to do it?
Once you’ve got the decision-making clear, you connect the relevant teams in the organization, and make sure they can align around it – not from an org-chart perspective, but as a shared accountability model. Who’s responsible for what? How should it work? This is mainly about governance and culture, not technology, but it’s essential to have it in place before you start automating anything.
Then you start thinking about how to enable those workflows to flow throughout the enterprise. Now that the decision-making is clear and the shared accountability model is in place, you can build the workflows around it as deterministic processes across the enterprise: if A happens, then you do B. Those workflows only work once you’ve defined the decisions that need to be made and agreed who acts and when, because without that, a workflow is just the automation of a broken process.
And finally, data integration – which I started with – really comes last. Once the workflows are defined, data integration becomes relatively straightforward: here’s the information the workflow needs, here are the systems it needs to access, and here are the connectors. Data integration becomes the enabler rather than the starting point. So it’s that flow: decision-making, connect the teams, define the workflows, and give access to the data.
Rob Wilkinson: That shared accountability is a really good point, and it really helps clarify the process, so thanks for that. ServiceNow often talks about connecting work across the enterprise. What does that mean in practice for CRM, especially across sales, service, operations and fulfillment?
Daniel Wilks: ServiceNow has long talked about being the connective tissue across the enterprise and its landscape of many different applications, systems and processes. Connecting work across the enterprise in CRM builds on that, because it’s about one unified customer lifecycle and workflow instead of many disconnected tools, experiences and inefficiencies.
Think about a scenario where a deal closes with the sales team. We allow that context to flow directly into the service teams, with stakeholder maps, contract terms and service level agreements all visible. Service cases can trigger operational workflows like fulfillment, finance and field service. We can surface implementation delays in real time on the right dashboards, and the organization can see what was promised and what’s actually happening, instead of siloed tickets.
So the real pain point ServiceNow addresses is the wall that exists between a customer query and case resolution. ServiceNow treats the whole customer journey as one workflow, with cross-functional visibility and automation, and that’s where the value compounds. As I said before, delivering an exceptional experience and delivering operational efficiency are two sides of the same coin. That’s particularly the case in complex scenarios, where customer engagement needs to span multiple teams and systems.
If we look at an organization like the Bundesliga in Germany – one many of us will know well if you’re into football – until recently, a good deal of the work they did serving their customers, the football fans, handling partner requests and running day-to-day club operations still ran through a variety of individual legacy systems that were built to log activity, not to complete it. They’re changing that by moving onto a single platform with ServiceNow rather than the several disconnected systems that were in place before. It’s a really strong example of a unified customer lifecycle in practice.
Rob Wilkinson: I suppose the more AI becomes involved in that way, the more customer service leaders need to think differently. What is that change? How should they think about the difference between automating an interaction and actually resolving the customer’s need? What’s key there?
Daniel Wilks: You’ve got two very distinct elements there. I talked a moment ago about the wall between a customer query and actually resolving the case. Automating an interaction is about making contact faster, smoother or cheaper for the customer, and giving them a whole range of ways to access services. But without resolving the customer’s need, that’s basically just an expensive way of providing advice.
Resolving the need is about actually fixing the customer’s problem, so they don’t go away frustrated and don’t come back with the same issue again. The two are different because automating an interaction looks like a chatbot that answers FAQs instantly, deflects to self-service or reduces handling time. That’s really valuable – it can cost an organization less to operate that way. But if the customer’s real issue is “my order is stuck, how do I resolve this?” and the chatbot just gives some troubleshooting steps and a link to an FAQ, then the automation happened, but the issue wasn’t resolved. That’s where the customer goes away frustrated or has to come back again. We’ve all been on the receiving end of that, and it’s where you start impacting customer satisfaction.
So I think leaders should ask themselves: is AI handling this interaction perfectly but leaving the need unresolved? Because if it is, we haven’t created the value through AI that we could have – we’ve just made the problem disappear faster. The real win comes from connecting automation, speed and efficiency to resolution capacity, which means using context, judgment and action to resolve customer cases.
That’s where we believe ServiceNow as a platform becomes relevant, because we’re not just enabling faster cases, but cases that stay resolved, because the AI can see and coordinate across service, operations and fulfillment simultaneously on a single platform.
And as I mentioned before, this is about coexistence. The organizations that are winning here aren’t replacing service teams with AI. They’re enabling their teams to use AI to handle volume and triage faster, so humans can focus on the 15% of interactions that actually require more judgment and authority, spend more time on complex cases, and do that with enough context to resolve them first time.
Rob Wilkinson: There are so many angles this could go down – it’s been so interesting. Try to pinpoint something for me: if someone wants to take their first step in this direction and do something this quarter, what’s the most sensible way to find and fix some of the gaps in customer resolution they might have?
Daniel Wilks: I’d start by getting an understanding of what’s really going on now, and thinking about where the opportunities are to improve at a process and experience level. As I mentioned before, applying technology on top of broken foundations is going to expose the problem and make it worse rather than resolve it.
An organization could spend time tracing a sample of recent cases – say 10 to 15 – from open to close, looking at how each process evolved and mapping out the system handoffs. How many systems did information have to move between? Where were the manual handoffs? Where did the case stall? What happened to the customer? Was resolution actually achieved, or was it just visibility? It’s almost like doing a mini audit, and the patterns arising from it will start to reveal where the process flow that allows resolution for your customers actually breaks.
There are some common gaps – for example, cases closed off in the ticketing system when the customer never really confirmed the issue was resolved, or fulfillment delays that are invisible to service teams until the customer complains.
Once an organization understands those underlying processes and what’s really going on, it can evaluate whether this is an automation problem or a workflow problem, and decide where to apply technology to resolve it. Because as I said, if we just apply AI on those foundations, it might not change anything – it might just compound the issue. We have a habit in this industry of not getting things right at the start, so it’s time to learn those lessons and get it right this time.
Rob Wilkinson: Thank you so much, Daniel, for very eloquently framing this shift from record to resolution. Before we wrap up, for anyone watching who wants to learn more about ServiceNow’s approach to CRM and customer resolution, or the work you’re doing with organizations, what’s the best way to do that?
Daniel Wilks: Ping me any time – you’ll find me on LinkedIn. We’d love to have the conversation.
Rob Wilkinson: Perfect. And of course, you can discover more articles, resources and videos just like this one at cxtoday.com. But for now, that’s all we’ve got time for. Thanks again to Daniel Wilks at ServiceNow for sharing his thoughts. I’m Rob Wilkinson – thanks very much for watching.
