Most CRM programmes start in the same place. Someone opens a whiteboard and draws the sales funnel: lead, opportunity, quote, closed won. Workshops follow on stages, fields and dashboards. Six months later the pipeline looks great, and the business still can't answer a simple question: what did we actually sell this customer, and is it working?
I've spent more than 14 years on ServiceNow, eight of them delivering Customer Service Management. For the last couple of years I've also been taking telcos and enterprise service providers across EMEA through Sales and Order Management and CPQ. Putting those two worlds together has left me with a firm view. CRM is won or lost at the order, not the opportunity.
The Opportunity Is the Easy Part
Opportunity management is a solved problem. Every serious CRM does it well enough, and nobody has ever lost a customer because their pipeline stages were slightly wrong.
The hard part starts after “closed won”. The order has to be decomposed, fulfilled, provisioned, billed and supported. In telco, one commercial product can become a dozen technical services across several delivery teams and systems. That is where the cost sits, where the customer experience is made or broken, and where most CRM platforms quietly hand the problem to something else.
For most of my time delivering CSM, that was true of ServiceNow too. The platform picked the customer up after the sale, but the sale itself happened somewhere else, and we spent a lot of effort stitching the two back together. Sales and Order Management is what closed that gap. But it only pays off if you design for it from day one.
Start With the Product Model, Not the Screens
The hardest conversation in every SOM engagement I've led has been the product catalogue. Not the UI. Not the integrations. The catalogue.
Product offerings, product specifications, characteristics, bundles, pricing rules, and how each commercial product maps to the technical services that deliver it. ServiceNow's model follows TM Forum patterns, which gives you a real head start, but it doesn't make the decisions for you.
Get this wrong and everything downstream inherits the problem. Quotes that can't be fulfilled. Orders re-keyed by hand. Install base records that don't match what's actually running at the customer site. I've seen more programme time lost to vague data modelling than to any technical issue.
My advice is simple. Before a single form is configured, take your five most common products and walk each one end to end on paper, from quote to fulfilment to the first support case. If you can't do that cleanly, you're not ready to build.
The Sold Product Is Where CRM Meets Operations
This is the part most CRM evaluations miss. In ServiceNow, a completed order doesn't just close a deal. Through the Common Service Data Model it creates sold product and install base records that link the customer to the services and infrastructure behind them.
That is the moment CRM stops being a sales tool and becomes an operating model. A case raised against an install base item already knows what the customer has, what it depends on and what else is affected. A field work order knows what the engineer is walking into. An outage in the network can be traced to every customer it touches.
That's why I push clients to treat CSDM as a CRM concern, not a CMDB hygiene exercise. The people who own the CMDB and the people designing the sales process rarely sit in the same room. They need to.
Decide What ServiceNow Owns, and Be Honest About It
No enterprise runs on one system. There's an ERP, usually SAP. There are billing platforms, network inventory and provisioning systems. ServiceNow doesn't need to own all of it, and trying to usually ends badly.
The worst integrations I've seen come from ambiguity about the system of record. Pricing held in two places. Customer master in three. Order status that depends on who you ask.
So be explicit, early. Where does price live? Who owns the customer master? Where does the commercial order end and the technical order begin? Draw the boundary, document it and defend it. A clean handoff to SAP beats a clever bidirectional sync every time.
CPQ Has Caught Up
For a long time the fair criticism of ServiceNow CRM was that configure-price-quote was the weak link. Complex configurations ended up in spreadsheets or a bolt-on tool, and the quote lived outside the platform that had to deliver it.
ServiceNow's acquisition of Logik.ai changed that conversation. Constraint-based configuration is built for products where the quote is itself an engineering problem, which describes most of what telcos and technology providers sell. It's still settling into the wider order flow, so go in with your eyes open, but the direction is clear. If you're approaching renewal on a separate CPQ, it's worth a proper look now.
AI Agents Need Clean Orders
Everyone wants agentic AI in their customer workflows, and I'm fully behind it. We deploy it. But an agent can only act on what the platform knows. If the product model is loose and the install base is wrong, the agent will confidently do the wrong thing, faster.
The organisations getting real value from AI in CRM are the ones who did the unglamorous work first: a tight catalogue, accurate install base and clear ownership of data. The AI is the reward for that discipline, not a replacement for it.
If You're Evaluating ServiceNow CRM
Four questions will tell you more than any feature comparison:
- Can you walk your top products from quote to support case without a spreadsheet?
- Who owns your product catalogue, and is that person in the CRM programme?
- Is the boundary between ServiceNow and your ERP written down?
- Would your install base data survive an AI agent acting on it?
If the answers are uncomfortable, that's fine. It means you've found where the real work is, before it finds you.
This is the work our CRM practice does every day, alongside our ITSM, SPM and SecOps practices, because the order doesn't stop at the edge of the CRM. If you're planning a ServiceNow CRM programme, or rescuing one, get in touch.
