# Managing Up and Cross-Functional Alliances
In 2019, a newly appointed CDO at a top-15 European bank walked into her first quarterly business review with a data strategy the board had already approved. Ninety days later she was gone. Not because the strategy was wrong—it was excellent—but because she had treated the CIO as a service provider, the CFO as a checkbook, and the heads of retail and commercial banking as end users who would eventually "come around." By the time she understood that these three people controlled every resource, dataset, and headcount she needed, they had quietly reclassified her initiative as "IT's data cleanup project" and starved it of funding.
The uncomfortable truth: a CDO has almost no direct authority over the assets that determine success. You borrow engineers from the CIO, budget from the CFO, and adoption from the business. Your mandate is real, but your leverage is entirely relational. This lesson is about the mechanics of building that leverage—turning the three people most likely to see you as a threat into the sponsors who defend your agenda when you're not in the room.
Before you can convert a rival into a sponsor, you have to understand *why* the friction exists. It is rarely personal. It is structural—it comes from overlapping mandates and scarce resources. Diagnose it precisely.
The CIO tension is about turf and definition. The CIO has historically owned data as a byproduct of running systems. Your arrival implies that data is now a first-class asset with its own owner—which reads, to a defensive CIO, as a carve-out of their empire. The tension intensifies when reporting lines are ambiguous. If you report *into* the CIO, you risk being subordinated to infrastructure priorities. If you report *around* them to the CEO or COO, you risk being seen as a rival who bypassed them.
The CFO tension is about proof and horizon. CFOs fund things that show returns on a timeline they can defend to a board. Most data initiatives return value on an 18-to-36-month curve, while CFOs are held to quarterly and annual cycles. Left unmanaged, this mismatch makes your program the first line item cut in a downturn.
The business-leader tension is about control and attention. A retail head or a supply-chain leader owns a P&L. They interpret "enterprise data governanceenterprise data governanceData governance is the set of policies, roles, and processes that ensure data is accurate, secure, well-defined, and used responsibly across an organization.Voir la définition complète →" as a tax on their speed and an intrusion on data they consider theirs. Their default is polite non-cooperation.
The mistake the failed CDO made was treating all three as one audience with one message. Each rivalry has a different root cause, and therefore a different conversion strategy.
For each of your top stakeholders, fill in four cells before you ever schedule a meeting:
The last cell is the one novices skip and the one that determines everything. Influence is not persuasion; it is exchange. If you cannot name what you give a stakeholder, you have no alliance—you have a request.
Managing up is not about pleasing your boss. It is about equipping the executives above you—your CEO, your board sponsor, your reporting line—to make the case for your agenda in the meetings you don't attend. Sponsorship is what survives budget season; approval is what dies in it.
A CDO's instinct is to report on data-native metrics: data qualitydata qualityThe degree to which data is fit for purpose: accurate, complete, consistent, timely, valid and unique. Poor quality data undermines analytics, reporting and AI.Voir la définition complète → scores, catalog coverage, model accuracy, pipelinepipelineAll active sales opportunities across the stages of the sales process, together with their combined potential value and probability of closing.Voir la définition complète → uptime. These mean nothing to a board. The discipline of managing up is *translation*—converting every data outcome into a financial or strategic outcome the sponsor already cares about.
Compare two versions of the same update:
> Version A: "We improved customer data completeness from 71% to 94% and deployed a new entity-resolution service."
> Version B: "By fixing customer identity resolution, we cut duplicate marketing spend by €4.2M and enabled the retail team to launch cross-sell campaigns six weeks faster. Completeness went from 71% to 94%, which is the mechanism—but the €4.2M is the point."
Version B gives your sponsor a sentence they can repeat verbatim to the CFO. That is the actual deliverable of managing up: a *transferable narrative*.
Senior data leaders lose credibility not when a project slips, but when a surprise slips. The single highest-leverage managing-up habit is the pre-emptive risk brief. Once a quarter, bring your sponsor the two or three things most likely to go wrong—an AI governance gap, a regulatory exposure, a critical dependency on an under-resourced team—*before* they become incidents.
This does two things. It positions you as the adult in the room who sees around corners, and it converts risks into shared ownership. A risk your sponsor knew about is a risk you manage together; a risk that surprises them is a failure you own alone.
Now the core work. Each of the three structural rivals requires a distinct conversion play.
The CIO alliance is won by making the boundary explicit and making it favorable to *them*. Ambiguity is your enemy; a clean division of labor is your friend.
The most durable model separates the pipes from the meaning. The CIO owns infrastructure, platforms, security, and delivery reliability. The CDO owns the semantics, quality, governance, and value—what the data *means* and whether it can be trusted for decisions. Codify this in a shared RACI that you draft *together*, not one you impose.
Here is a compact way to make the split legible in a single artifact both organizations can point to:
data_domain: customer_360
platform_ownership: CIO # infra, uptime, security, cost
pipeline_delivery: CIO # ingestion, orchestration SLAs
data_product_definition: CDO # schema semantics, quality rules
quality_sla: CDO # accuracy, freshness thresholds
access_governance: shared # CDO sets policy, CIO enforces
business_value_owner: CDO + LOB # tied to a named P&L metricThe strategic move: publicly credit the CIO for delivery wins. When your customer-360 product drives revenue, the CIO's platform made it possible—say so, loudly, in front of the CEO. A CIO who gets reflected glory from your success becomes your strongest infrastructure advocate. Take the credit alone once, and you'll spend a year fighting for compute.
The CFO becomes a sponsor when you stop asking them to *fund faith* and start giving them a *portfolio to manage*. Reframe your initiatives the way a CFO already thinks about capital allocation: a mix of horizons and risk profiles.
Present your data agenda as three tranches:
This structure lets the CFO fund the whole program by pointing to the efficiency tranche's near-term returns, while you protect the long-horizon work under the same umbrella. You've turned an unfundable 30-month promise into a self-financing portfolio.
The tactical detail that seals it: co-author the business case *with* the CFO's team. A number the finance organization helped calculate is a number the CFO will defend. A number you calculated alone is a number they'll challenge.
Vérification des acquis
1. According to the lesson, why is a CDO's leverage described as 'entirely relational'?
2. The lesson argues that friction between the CDO and other executives is 'rarely personal.' What is the primary source it identifies instead?
3. In the opening case, what was the fundamental cause of the CDO's failure within 90 days?
4. Select ALL correct answers. According to the lesson, why does the arrival of a CDO create tension with a defensive CIO?
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers. Which strategic principles about the CDO role are supported by the lesson?
Sélectionnez toutes les réponses correctes.
Business unit heads are the hardest conversion because their cooperation is voluntary and their default is inertia. You cannot govern your way to their buy-in. You must make one of them a *visible winner* first.
The proven sequence is the lighthouse pattern: find the business leader with the most acute, quantifiable pain and the most cooperative temperament, and pour disproportionate effort into a single win for them. Do not spread thin across all units to seem fair. Concentrate. When that leader stands up in the operating committee and says the data team helped them hit their number, you have done something no governance mandate can achieve—you've created peer pressure among your rivals.
Then let scarcity work for you. Once you have one lighthouse and a queue of interested units, you allocate your team's capacity based on which unit commits real skin in the game: a named product owner, adoption targets, a shared success metric. You've flipped the dynamic. Instead of begging business leaders to comply with governance, you have them competing for access to your capacity.
One caution: never let a business leader treat you as their private analytics team. The alliance holds only if you deliver *reusable* assets—a data productdata productA data asset managed like a product, with an owner, defined users, guaranteed quality, and measurable business value.Voir la définition complète → one unit funds that three others inherit. That reusability is what makes you an enterprise asset rather than a favor bank, and it's the story that justifies your budget to the CFO.
Knowing the three conversions isn't enough; sequence matters. The failed CDO from our opening had all three relationships to build and tried to build them simultaneously and identically. Order them.
Start with the lighthouse business win, because a proof point makes every other conversation easier. Then use that win to convert the CFO, because now you have a real number, not a projection. Run the CIO alliance in parallel from day one, because you cannot deliver the lighthouse without their infrastructure—so the joint operating model is a prerequisite, not a follow-on.
Sustaining the coalition requires one under-appreciated habit: the *pre-meeting*. The decisions that matter are never made in the room where they're announced. Before any steering committee, walk the room privately. Know the CFO's objection before they raise it and pre-address it. Give the CIO their credit line before the meeting so they arrive as an ally. Confirm your lighthouse leader will speak up. When the meeting happens, it should feel like a formality—because you've already built the consensus one conversation at a time.
Finally, audit your coalition quarterly with a brutal question: *Who would defend my agenda if I were on vacation during budget cuts?* If the honest answer is "no one," you have stakeholders, not sponsors—and you are one downturn away from the exit our opening CDO took.