# Leading finance transformation: change management for cfos
When Ronald Edmonds took over as Dow Chemical's Controller in 2014, the company's finance function employed roughly 2,500 people across 40 countries, ran on 130+ ERP instances, and consumed nearly 1.4% of revenue, well above the chemical industry benchmark of 0.8%. Eight years later, Dow's finance operations cost the company $400 million less annually, ran on a single SAP S/4HANA backbone, and routinely closed the books in five business days. But here's the part most finance leaders miss: Dow's leadership team estimates that less than 30% of the value came from the technology. The rest came from how they led the people through the change.
That ratio matters because finance transformations are failing at an alarming rate. Gartner's 2025 Finance Transformation Pulse Survey found that 70% of finance transformations fail to deliver their original business case, the same failure rate John Kotter identified for general corporate change initiatives back in 1995. Three decades, billions in consulting fees, and a generative AI revolution later, the number hasn't moved. The variable that separates the winners from the wreckage isn't the tech stack. It's CFO leadership.
The post-COVID environment has put finance functions under pressure unlike anything in living memory. CSRD reporting obligations went live for large EU undertakings in FY2024 disclosures, OECD Pillar Two's 15% global minimum tax is now in force across 55+ jurisdictions, and IFRS 16 lease accounting has fully bedded in for most balance sheets. Layer on generative AI, real-time treasury demands, and boards asking why month-end still takes 10 days, and you have a perfect storm. McKinsey's 2025 Global CFO Survey found that 83% of CFOs have a finance transformation program currently underway. Most of them will fail.
The failures cluster around four predictable patterns:
The single biggest failure mode is treating transformation as a technology implementation. The CFO sponsors an S/4HANA migration, a Workday Adaptive rollout, or an Anaplan deployment, sets a go-live date, and assumes the operating model will sort itself out. It never does.
Consider GE's well-documented finance transformation under former CFO Jeff Bornstein. The technology layer, consolidating onto Oracle Cloud and standardizing on Hyperion, was executed competently. But the transformation failed to deliver because GE didn't redesign the work itself. Controllers continued doing reconciliations manually because no one explicitly killed the legacy processes. The system was new; the behavior wasn't.
Finance professionals derive identity from technical mastery, IFRS expertise, tax structuring, complex reconciliations. When a CFO announces that automation will eliminate 40% of transactional work, what the controller hears is: "Your craft is obsolete." Without addressing this identity shift directly, you get passive resistance, the most lethal form, because it's invisible until the project misses milestones.
Senior leadership endorses the vision. Staff are enthusiastic about new tools. But directors and senior managers, the "frozen middle", quietly sabotage execution because the transformation threatens their span of control, their headcount budget, or their political capital. Prosci's change management research consistently shows that middle-manager resistance accounts for over 50% of transformation delays.
Go-live is the start of transformation, not the end. Yet most CFOs reassign the transformation team within 90 days of cutover, declare success in the next earnings call, and move on. Sustained behavior change requires 18-24 months of reinforcement after the technical implementation completes.
Dow's transformation, which ran from 2017 through 2023 under CFO Howard Ungerleider (and later Jeff Tate), is the most instructive case study available because the company published enough detail at investor days for outsiders to reverse-engineer what they actually did. The headline numbers, $400M in annualized finance cost reduction, finance cost as a percentage of revenue down from 1.4% to 0.7%, close cycle compressed from 10 days to 5, are impressive. The methodology is more interesting.
Dow didn't sell the transformation as "we need to modernize finance." Ungerleider framed it externally as part of the company's $6 billion EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition → improvement commitment to investors following the DuPont merger spin-off. Internally, finance transformation was positioned as enabling commercial agility, specifically, giving business unit leaders real-time margin visibility to support dynamic pricingdynamic pricingAutomatically adjusting prices in real time based on demand, competition or user behaviour to optimise revenue, margin or conversion.View full definition → in volatile petrochemical markets.
This matters because it changed who owned the transformation. It wasn't "Ron's project in Controllership." It was a Dow enterprise commitment with finance as the delivery vehicle. Commercial leaders had skin in the game.
CFO lesson: Never sell finance transformation as a finance initiative. Anchor it to a board-level commitment, revenue growth, margin expansion, capital efficiency, and force the rest of the C-suite to co-sponsor.
Dow spent the first 14 months, before signing the SAP contract, redesigning the finance operating model. They moved to a "Global Business Services + Centers of Excellence + Embedded Business Partners" structure, consolidating 40 country finance teams into four regional GBS hubs (Midland, São Paulo, Prague, Mumbai).
Critically, they documented exactly which activities moved where, which roles were eliminated, and which new roles were created, *before* any system selection. By the time vendors were pitching ERP solutions, Dow knew exactly what processes the technology had to support.
This is where most CFOs lose their organizations. Dow eliminated roughly 800 finance roles over the program. Ungerleider and Edmonds announced the headcount impact in the first town hall, gave affected employees 18 months of advance notice, offered generous severance, funded reskilling for those who wanted to transition to data analytics or business partnering roles, and reported progress quarterly.
The transparency cost short-term morale but bought long-term credibility. Compare this to companies that obfuscate headcount impacts, where the resulting rumor mill destroys trust and drives the best people out first.
Dow identified 25 senior finance business partners across the operating segmentssegmentsDividing a market into distinct groups of customers who share similar needs, characteristics or behaviours, so each group can be served with a tailored approach.View full definition → and made them co-architects of the new operating model. They had veto power over service-level agreements with GBS, sat on the governance committee, and were measured on the transformation's success. This converted the most powerful potential resisters into owners.
Knowledge check
1. According to Dow Chemical's leadership team, what percentage of the value created by their finance transformation came from the technology itself?
2. What failure rate did Gartner's 2025 Finance Transformation Pulse Survey identify for finance transformations, and how does it compare to Kotter's 1995 finding on general corporate change?
3. OECD Pillar Two's global minimum tax, cited as one of the post-COVID pressures on finance functions, establishes what minimum effective tax rate?
4. Select ALL correct answers about the regulatory and operational pressures facing finance functions in the post-COVID environment as described in the lesson.
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers about the state of finance transformation programs based on the lesson's data.
Sélectionnez toutes les réponses correctes.
You're a CFO. You have a board commitment, a $30 million budget, a 36-month timeline, and 1,200 finance staff who are nervous. What do you actually do?
Generic "we need to transform" speeches die on contact. Replace them with a specific, quantified case for change:
Procter & Gamble's CFO Andre Schulten used exactly this approach in 2023 when launching P&G's finance transformation, publicly committing to reducing finance cost by 25% over four years and tying executive compensation to the milestones. When the case for change is mathematically undeniable, resistance becomes harder to sustain.
Identify the 15-25 people whose endorsement will determine success, these are almost never the highest-titled people. They're the long-tenured controller everyone respects, the regional CFO whose business partners trust her, the IT director who knows where the data bodies are buried.
MapMapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.View full definition → each one on two axes: their influence on the broader organization, and their current stance (advocate / supporter / neutral / skeptic / blocker). For anyone in the high-influence/skeptic quadrant, you personally, not your transformation lead, need to spend 1:1 time understanding their concerns and converting them or neutralizing them.
Transformations die in the gap between monthly steering committees. Establish a bibiTechnologies and processes that turn raw data into actionable insights via reporting, dashboards and analysis, so teams can decide based on facts rather than intuition.View full definition →-weekly 90-minute transformation operating review chaired by the CFO. Standing agenda: milestone status (red/amber/green with specific decisions required), people impact metrics (voluntary attrition in affected populations, engagement survey results, training completion rates), benefits realization tracker, and top three risks requiring CFO intervention.
Microsoft's CFO Amy Hood is well-known internally for running finance operating reviews with this cadence, and for accepting "red" status reports without shooting messengers. The cultural willingness to surface bad news quickly is what enables course-correction.
Every major transformation has a productivity trough roughly 4-8 months after go-live. Close cycles get longer, not shorter. Errors increase. Staff complain that "the old system was better." This is normal. Brief your CEO, audit committee, and board *in advance* that this will happen. If they're expecting linear improvement, they will lose nerve at exactly the wrong moment and pull resources.
Unilever's CFO Graeme Pitkethly explicitly briefed investors during the company's 2020-2023 finance transformation that productivity would dip before improving. When it did dip in Q2 2022, there was no panic, because expectations had been set.
The most underused lever in finance transformation is compensation alignment. If your top 50 finance leaders' annual bonuses are 80% tied to traditional finance metrics (close time, audit findings, budget accuracy) and only 20% to transformation milestones, they will rationally deprioritize transformation when the two conflict. Flip the ratio for the duration of the program.
1. Anchor to a board-level outcome, not a finance metric. Reframe the transformation as enabling a commitment the CEO has already made to investors. This converts your peers from spectators into co-owners.
2. Redesign the operating model before selecting technology. Spend the first 6-12 months on process and organizational design. The ERP, EPM, and AI tools should be selected to fit the future operating model, not the other way around. Companies that reverse this sequence pay a 30-40% premium in rework costs.
3. Be brutally transparent about headcount impact within the first 90 days. Announce the number, the timeline, the severance terms, and the reskilling investment. Ambiguity destroys trust faster than bad news. Treat the people decisions as the central transformation decision, not an HR afterthought.
4. Build a bi-weekly operating rhythm you personally chair. Monthly cadence is too slow; quarterly is fatal. Track three metrics every cycle: milestone