CRM & marketing automation: real-world application

An MRI system costs a hospital somewhere in the low millions, runs for the better part of a decade, and gets replaced after a decision involving radiology, biomedical engineering, IT, finance, and a procurement office that has never opened one of your emails. That is the shape of the deal Philips's imaging business lives with: long replacement cycles, an installed base worth more than the new-name market, and a buying group where the person reading your content signs nothing. This lesson stays inside one lifecycle programme built for that shape, from the data model up to the pipelinepipelineAll active sales opportunities across the stages of the sales process, together with their combined potential value and probability of closing.View full definition → number the CFO sees. Salesforce, HubSpot and Marketo (now Adobe Marketo Engage) all sell the platforms described here. The mechanics below are the standard enterprise pattern; the arithmetic at the end is a worked model, not a reported Philips figure.

The programme on the table

One job: get a named buying group at a hospital site into a replacement conversation nine to twelve months before the equipment reaches end of life. The record layer is the one the foundations lesson describes, plus two objects most B2B marketers never touch. An asset record for every installed machine, carrying serial number, install date and service contract expiry. And an account hierarchy that keeps a 40-site hospital group distinct from the sites inside it.

The hierarchy decides whether any of this works. If two sites of the same group land as two flat accounts, the group procurement director is counted twice, the dashboard reports two opportunities where there is one tender, and a rep calls a contact whose budget was centralised eighteen months ago. Fixing hierarchy is unglamorous, takes a quarter of data work with the service organisation, and has to happen before anyone writes an email.

Scoring the buying group, not the contact

The scoring model the foundations lesson sets up assumes the unit is a person. Here it is a buying group at a site, which changes the construction. Points still accrue per contact, but the threshold requires breadth: engagement from at least three role types (clinical, technical, financial or procurement) plus an asset trigger, typically a service contract inside its final twelve months. One enthusiastic radiologist reading up for a conference paper never crosses it. Three roles moving in the same month usually means a capital request is being drafted.

Two failure modes worth building against. First, internal traffic: Philips service engineers and clinical application specialists spend their days on product and support pages from hospital networks, and their activity will inflate account scores unless you exclude those contact types and known service IP ranges. Second, staleness. A tender that goes quiet for sixty days has usually been decided by someone else, so points decay rather than accumulate forever. Add negative scoring too: a site that installed a scanner eighteen months ago is out of the market for years, and continuing to mail it replacement content is how you teach a hospital group that your emails are noise.

The campaign, keyed off assets rather than personas

Three tracks, all triggered by asset data rather than a persona guess. Track one runs twelve to nine months out and is entirely about total cost of ownership: uptime, throughput per shift, energy draw, the cost of an unplanned outage in a department running 30 scans a day. Track two, nine to four months, answers the objections that actually kill capital requests: disruption during installation, retraining time for technologists, integration with the existing PACS and IT estate, trade-in and refurbished options, financing structures for a group that cannot spend capexcapexCapital Expenditure (CapEx) is money spent to acquire, upgrade, or extend long-lived assets like equipment, property, or software that deliver value over multiple years.View full definition → this year. Track three is procurement-facing and mostly documents: tender-ready specification sheets, service level terms, regulatory files.

A second-order effect that catches teams out. A replacement conversation opened by marketing can cut across a service contract renewal the field service manager is negotiating in the same building. If the automation does not notify the service owner on the account at the same moment it notifies the SDR, two Philips people end up negotiating against each other in front of the same procurement director. Put the service account owner on the alert, not just sales.

Consent, and the measurement it quietly breaks

European hospital contacts sit under GDPRGDPREU regulation governing how organizations collect, store and use personal data, with fines tied to global revenue for breaches.View full definition →, and in several markets the legitimate-interest route to a marketing newsletter is thin enough that a health technology vendor will not use it. So consent state, source and timestamp live on the contact record and gate every track; a site can be scored on web and asset behaviour while remaining unmailable, and the programme has to route those accounts to field sales instead of email.

Measurement takes a second hit. Since Apple's Mail Privacy Protection shipped with iOS 15 in September 2021, opens are pre-fetched and inflated, which matters more here than in consumer marketing because clinicians read on phones. Any scoring model still weighting opens is scoring Apple's proxy server. Weight clicks, page depth, document downloads and repeat visits from the same account instead. If profile stitching across service portals, events and web sits in a customer data platformcustomer data platformSoftware that unifies customer data from every source into one persistent profile that marketing, sales and service teams can act on.View full definition → of the kind the CDP foundations lesson covers, the consent flags have to travel with the profile, not be re-derived downstream.

The handoff, which is where the programme usually dies

When a buying group crosses threshold, the alert has to reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → the one rep who owns that node of the hierarchy, with the asset record, the three engaged roles and what each of them read attached. A generic "MQLMQLA Marketing Qualified Lead (MQL) is a prospect whose engagement and fit signals indicate they are more likely to become a customer, justifying handoff toward sales.View full definition → assigned" task gets ignored, and rightly so, because a rep carrying eight hospital groups cannot act on a name with no context.

Then the discipline: a disposition code within 48 hours, four values only (contacted, qualified, disqualified, recycled). Disqualified accounts return to a marketing-owned track automatically with a suppression window. Recycled accounts keep their engagement history, which is why they convert better than anything cold. The SLASLAA formal commitment defining the service level a provider guarantees to a customer, with measurable targets and consequences if they are missed.View full definition → report goes to the sales VPVPA clear statement of the benefits your product delivers, the problems it solves and why customers should choose you over alternatives.View full definition → and the CMO on the same Monday email, because an SLA nobody reads is a suggestion. One more thing to arbitrate: if SDRs are paid on meetings booked, they will cherry-pick the easy sites and let complex group tenders rot at 47 hours. Compensation has to recognise buying-group progression, or the routing logic gets gamed within a quarter.

Following it through to pipeline

Illustrative arithmetic on a book of this shape. Say the programme covers 1,100 installed imaging assets across 380 sites in one region. Service data flags roughly 90 assets entering their final twelve months each quarter. The scoring model, requiring three roles plus the asset trigger, clears about 35 buying groups. Sales accepts 22, disqualifies 9, recycles 4. Nine reach a formal tender, and three close inside the following four quarters at a system-plus-service value in the low millions each.

The number that moves is not volume, it is timing. Reaching a group before the specification is drafted is what gets your uptime and throughput language into the tender document. Arrive after the spec is written by a competitor's application consultant and you are bidding on someone else's criteria at a discount. That is the pipeline argument to make to a CFO: the programme does not manufacture demand for scanners, it changes who defines the requirement.

CMO action items

• Ask for the account hierarchy report before anything else: how many of your top 50 customer groups are represented by more than one unlinked parent account? If the answer is above five, your pipeline count is wrong and no amount of automation fixes it.
• Get service or product data into the marketing platform as a trigger, even one field. Contract expiry date beats every persona attribute you own.
• Rebuild scoring so the threshold requires multiple roles from one account, then re-run last year's closed-won deals through it. If fewer than half would have fired, the weights are describing readers, not buyers.
• Put a 48-hour disposition SLA on named-account alerts, publish the breach list weekly to sales leadership and yourself, and check whether SDR incentives reward the hard accounts.

Common mistakes that kill results

Buying the platform before owning the data. A six-figure annual automation licence sits idle for a year while someone reconciles duplicate hospital accounts and works out which serial numbers are still installed. The platform automates what the data supports, nothing more.

Set-and-forget sequences. A track written in 2022 is speaking to a procurement climate that has changed, and its measured open rate is now partly Apple's proxy. Review quarterly, retire what has stopped earning clicks, and rewrite objection content when the competitive set moves.

Automating through the decision. In a multi-million euro tender with a formal RFP, automation earns the meeting and keeps the account warm between rounds. It does not close. Teams that keep nurturing while a rival's clinical specialist is on site running a workflow assessment lose on relationship, then blame the email.

Claiming the whole deal. If marketing reports the full contract value on every influenced tender while the field service manager reports the same account, leadership stops trusting both numbers. Agree the attribution rule with sales before the first quarterly review, not after.