Overcoming resistance & data politics

# Overcoming resistance and data politics

In 2019, a newly appointed CDO at a global insurer walked into her first quarter with a mandate, a budget, and a board sponsor. Eighteen months later she was gone, not because her architecture failed, but because the head of claims quietly withheld the loss-adjustment data her models needed, the CFO's team "couldn't find bandwidth" to reconcile definitions, and three regional heads each ran a shadow analytics team that reported numbers contradicting hers in front of the CEO. Her technology worked. Her politics didn't.

This is the failure mode that kills more CDOs than bad pipelines: the assumption that data problems are solved by data solutions. Resistance to a data agenda is almost never about the data. It's about power, fear, and the reallocation of decision rights. Your job in this lesson is to learn to read those forces and act on them deliberately.

Diagnose the resistance before you respond to it

The most common CDO mistake is treating all resistance as the same problem, usually as an education problem, to be solved with another deck explaining why data matters. It isn't. Resistance comes in distinct flavors, and each requires a different move. Misdiagnose the type and your intervention backfires: you evangelize to someone who's afraid, or you reassure someone who's simply protecting turf.

Four archetypes cover most of what you'll encounter:

• The Hoarder controls a dataset that is a source of their organizational value. The regional 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 → who "owns" customer profitability data knows that once it's in the shared warehouse, their monopoly on the narrative disappears. Their resistance is rational self-interest.
• The Skeptic-by-Scar has lived through two failed data programs and reads your initiative as the third. Their resistance is pattern-matching, and it's often the most useful signal you'll get, they know where the bodies are buried.
• The Transparency-Averse fears what the data will *reveal* about their function's performance. The operations head whose true defect rate has never been measured does not want it measured. This is fear disguised as methodological objection ("your 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.View full definition → isn't good enough to draw conclusions").
• The Sovereign doesn't oppose data, they oppose *your* control of it. They'll happily build analytics, as long as it reports to them. This is a decision-rights fight, not a data fight.

Before any major initiative, run a simple stakeholder mapmapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.View full definition →. Plot each key player on two axes: influence over your outcome (low to high) and current stance (blocker to champion). Then annotate each blocker with their archetype and, critically, their *underlying interest*, the thing they're actually protecting.

Stakeholder      Influence   Stance      Archetype        Real interest
---------------  ----------  ----------  ---------------  --------------------------
VP Claims        HIGH        Blocker     Hoarder          Data = leverage in budget cycle
CFO              HIGH        Neutral     Skeptic-by-Scar  Burned by prior MDM overrun
Head of Ops      MEDIUM      Blocker     Transparency-    Fears defect rate exposure
                                         Averse
Regional MD EMEA HIGH        Blocker     Sovereign        Wants analytics under P&L control

This is not a document you share. It's your operating mapmapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.View full definition →. The discipline of naming the *real interest*, not the stated objection, is what separates CDOs who move the organization from those who write strategy papers that die in committee.

Convert the interest, not the argument

Once you know what someone is actually protecting, you stop arguing with their stated objection and start negotiating with their real interest. This is the central skill.

With the Hoarder, trade, don't confiscate. The claims VPVPA clear statement of the benefits your product delivers, the problems it solves and why customers should choose you over alternatives.View full definition → who fears losing narrative control gets something in return for contributing data: co-authorship of the metrics, first access to the enriched dataset, and public credit as the source. The most durable move is to make the person who owns the data the person who benefits most visibly from its sharing. You are not taking their asset; you are increasing its return. In practice this means designing your data products so that contributors get preferential, earlier, richer access than consumers. Reciprocity beats mandate every time a mandate can be quietly ignored, and it usually can.

With the Skeptic-by-Scar, concede the past and shrink the bet. Do not defend prior programs; you didn't run them and you gain nothing by owning their failures. Instead, name the failure explicitly ("the last MDMMDMMaster Data Management (MDM) is the discipline of creating and maintaining a single, consistent, trusted version of an organization's core business entities like customers, products, and suppliers.View full definition → effort tried to boil the ocean and burned $4M with nothing shipped") and structure your ask as the opposite: a narrow, reversible, time-boxed pilot with a kill criterion you define *with* them. Skeptics become your best allies because their credibility is high precisely because they've said no before. When a known skeptic endorses your pilot, the room believes it.

With the Transparency-Averse, control the exposure sequence. This is delicate. Forcing sudden visibility onto a fearful stakeholder creates an enemy who will spend political capital to discredit your 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.View full definition → forever. Instead, give them a private preview and a runway. "You'll see these numbers first, for 60 days, before anyone else does, and we'll build the improvement story together before we publish." You are converting a threat of exposure into an offer of a head start. The operations head who would have sabotaged your dashboard becomes the sponsor of the improvement initiative it triggered.

With the Sovereign, separate control of the asset from control of the capability. The regional MD who wants analytics under their P&L can often be satisfied with a federated model: shared definitions, standards, and platform (yours), with embedded analysts who have a dotted line to them (theirs). You give up local control you never needed and keep the enterprise coherence you can't compromise. Fights over org-chart lines are usually proxies for fights over autonomy, give the autonomy where it's cheap, hold the line where it's structural.

The unifying principle: never negotiate against the objection on the table. The objection is theater. The interest is the script. Address the interest and the objection dissolves, often without the person ever admitting it was never really about 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.View full definition →.

Build the political infrastructure that outlasts you

Individual conversions are tactical. If you rely on them alone, you'll spend your entire tenure re-litigating the same fights every reorg. The senior move is to build structures that change the default behavior of the organization, so that sharing, transparency, and shared definitions become the path of least resistance rather than a heroic act you personally broker each time.

Three mechanisms matter most.

First, decision rights, written down and ratified above you. The single most powerful artifact a CDO can create is a data-domain ownership model where every critical data domain has a named accountable owner, a defined steward, and explicit rules for who can change definitions and who must be consulted. This sounds like governance from the fundamentals, but the political point is different. The value is not the document; it's the *ratification*. When the executive committee formally approves the model, you convert a thousand future turf fights into a single reference: "per the domain model the EC signed in Q2, customer master is owned by Marketing with Finance as consulted party." You've moved the fight from your desk to a settled policy. The Sovereign can no longer relitigate autonomy in every meeting.

**Second, make the metric definition process visible and contested *on purpose*.** Most data politics erupts because two functions compute "active customer" differently and each defends their number as truth. Rather than imposing a definition (which makes you the enemy of whoever loses), run a metric-arbitration forum where competing definitions are surfaced, the business trade-offs are debated openly, and the decision is documented with its rationale. You are not the judge; you are the convener. This does two things: it depersonalizes the disagreement, and it makes the losing party a participant in a fair process rather than a victim of your preference. People accept outcomes they lose when they believe the process was legitimate.

Third, instrument the resistance so it's visible to power. Hoarding thrives in the dark. When the claims VPVPA clear statement of the benefits your product delivers, the problems it solves and why customers should choose you over alternatives.View full definition → "can't prioritize" the data-sharing work for three quarters, that stall is invisible unless you make it visible. A simple data-contribution scorecard, which domains are shared, which are still siloed, which SLAs are met, reported to the executive committee turns private foot-dragging into a public fact. You never accuse anyone. You simply report the state of the world and let the CEO's attention do the work. The most effective political pressure a CDO applies is transparency about the transformation itself.

Sequencing, air cover, and knowing when to fight

Even with the right conversions and structures, timing and positioningpositioningThe mental space you want your brand to occupy in your target customer's mind relative to alternatives.View full definition → determine whether you win. Three judgement calls separate effective CDOs from correct-but-defeated ones.

Sequence for momentum, not logic. The logically optimal first project is often the politically worst. It may be the biggest, most contested domain, exactly where resistance concentrates. Instead, pick an early initiative where a willing stakeholder has a real pain, the data is attainable, and a visible win lands in one to two quarters. That win becomes your reference story. When you approach the Hoarder next, you're not selling a promise; you're pointing at a peer who already benefited. Resistance erodes faster from evidence of a colleague's success than from any argument you can make.

Spend your sponsor's capital deliberately, and rarely. A board or CEO sponsor is a finite resource, not a hammer. If you escalate every stalled meeting, you train the organization to see you as someone who can't operate without running to the parent, and you exhaust the sponsor's willingness to intervene. Reserve escalation for structural blockers, a ratification that won't happen, a Sovereign who's actively contradicting enterprise numbers publicly. Handle the rest yourself. When you do escalate, arrive with the interest-mapped diagnosis and a specific decision you need, not a complaint.

Know which fights to lose on purpose. Not every hill is worth it. If a Sovereign insists on owning a low-stakes local dataset that doesn't threaten enterprise coherence, let them win visibly. A conceded fight that costs you nothing buys goodwill you'll spend later on the fight that matters. The CDOs who flame out are often the ones who are *right about everything* and therefore fight *everything*, accumulating enemies faster than allies. Political capital is a portfolio: you must lose small to win big.

The insurer's CDO from our opening lost because she treated the claims VPVPA clear statement of the benefits your product delivers, the problems it solves and why customers should choose you over alternatives.View full definition →'s data-withholding as an operational obstacle to be escalated, rather than a Hoarder protecting budget-cycle leverage. Had she offered co-ownership of the loss-ratio metrics and preferential access to the enriched model output, the VPVPA clear statement of the benefits your product delivers, the problems it solves and why customers should choose you over alternatives.View full definition →'s interest would have flipped from blocking to sponsoring. The data was never the problem. The interest behind the data was, and she never diagnosed it.

Key Takeaways

• Diagnose the archetype and the real interest before you respond. MapMapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.View full definition → every high-influence blocker as Hoarder, Skeptic-by-Scar, Transparency-Averse, or Sovereign, and name the interest they're actually protecting, not the objection they voice.
• Negotiate against the interest, never the objection. Trade with Hoarders, shrink the bet for Skeptics, sequence exposure for the Transparency-Averse, and give autonomy-where-cheap to Sovereigns.
• Ratify decision rights above your level. A domain-ownership model formally approved by the executive committee converts thousands of future turf fights into a single settled reference.
• Make resistance visible to power without accusing anyone. A contribution scorecard reported to the EC turns private foot-dragging into a public fact that leadership attention resolves for you.
• Manage political capital as a portfolio. Sequence for early wins, spend sponsor capital rarely and deliberately, and concede small fights to bank goodwill for the structural ones you cannot afford to lose.