# Synergy estimation: why 70% of deals destroy value (and how to beat those odds)
When Kraft Heinz wrote down $15.4 billion of goodwill in February 2019, much of it tied to brands acquired during the 2015 mega-merger orchestrated by 3G Capital and Berkshire Hathaway, it became the most public confession of a quiet industry truth: the synergy model that justified the deal had been a work of fiction. 3G's celebrated zero-based budgeting was supposed to extract $1.7 billion in annual savings. It did, and in the process, gutted the brand investment that kept Oscar Mayer, Kraft, and Velveeta relevant. The cost synergies were real. The reverse synergies, the value that walked out the door, were larger.
This is the M&A paradox of the post-COVID era. KPMG's longitudinal research, along with studies from Bain, BCG, and McKinsey, consistently shows that 70-90% of acquisitions fail to create value for the acquirer's shareholders. And yet global M&A activity, which rebounded to roughly $3.4 trillion in 2024 and is tracking toward $3.8 trillion in 2026 as rate cuts unlock dry powder, shows no sign of slowing. The single most cited cause of failure, across every major study, is the same: synergies were overestimated, mistimed, or never existed in the first place.
For a CFO, synergy estimation is not a modeling exercise. It is the discipline that separates the deals that earn you a board seat from the ones that end your career.
Every synergy model has four buckets, and each lies in its own characteristic way.
Cost synergies are the synergies bankers and CFOs love because they appear concrete: consolidate two headquarters, eliminate duplicate ERP systems, rationalize procurement spend, reduce headcount in overlapping functions. They typically constitute 60-80% of the announced synergy number in horizontal deals.
The lie is in three places:
1. Timing. Most models assume synergies are realized within 24 months. Bain's data shows the median is closer to 36 months, and approximately 30% of announced cost synergies are never fully realized. In a DCFDCFDiscounted Cash Flow (DCF) is a valuation method that estimates an asset's value by projecting future cash flows and discounting them to present value using a required rate of return.View full definition →, pushing realization from year 2 to year 4, at a 10% WACC, destroys roughly 17% of the synergy NPVNPVNet Present Value is the sum of an investment's future cash flows discounted to today, minus the initial outlay. A positive NPV signals value creation.View full definition →.
2. Cost-to-achieve. A reasonable benchmark is that for every $1 of run-rate cost synergy, you will spend $1.00, $1.50 in one-time integration costs (severance, IT migration, real estate exit, consulting). Models that show $400M of synergies and $200M of cost-to-achieve are almost always understating the second number.
3. Dis-synergies. When you eliminate the duplicate sales team, you lose the customer relationships those salespeople carried. When you consolidate plants, you lose the institutional knowledge of the workers who don't relocate.
If cost synergies lie around the edges, revenue synergies lie in the foundation. McKinsey research on 200+ deals found that roughly 70% of revenue synergy targets are missed, and the average shortfall is around 40%.
Revenue synergies, cross-selling, geographic expansion, bundled offerings, depend on customer behavior, salesforce incentive alignment, and channel conflict resolution. They require integration discipline that most acquirers don't possess. The discipline a CFO should impose: never let revenue synergies justify more than 25% of the deal premium. If they do, you are betting your premium on the most unreliable number in the model.
Capital synergies, working capitalworking capitalWorking capital is the difference between a company's current assets and current liabilities, measuring short-term liquidity and the funds available to run daily operations.View full definition → efficiencies, combined 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 → rationalization, shared R&D, are systematically underestimated in Excel models because they require cross-functional input that financial models rarely capture.
Tax synergies, post-OECD Pillar Two implementation in 2024, have become both more constrained and more strategic. The 15% global minimum tax has eliminated most of the aggressive structuring that used to add 100-300 bps to deal IRRs through inversion or IP migration. CFOs running 2026 deal models who still have aggressive tax assumptions from a 2018 playbook are working with broken inputs. Conversely, deals that bring net operating losses, R&D tax credits, or genuine operational tax efficiencies (not synthetic ones) remain valuable, but the bar for what survives Pillar Two scrutiny is materially higher.
Here is the question almost no synergy model asks: What value does the target lose by being acquired?
This is the concept of reverse synergies, sometimes called "negative synergies" or "value leakage", and it is the silent killer of most failed deals.
Reverse synergies take several forms:
The practical discipline: for every synergy line in your model, require a corresponding "reverse synergy" or "dis-synergy" line. If your team can't articulate what value might be lost, they haven't done the diligence.
Knowledge check
1. In February 2019, Kraft Heinz recorded a goodwill writedown of approximately what magnitude, exposing the failure of the synergy model behind the 2015 mega-merger?
2. According to KPMG, Bain, BCG, and McKinsey research cited in the lesson, what percentage of acquisitions fail to create value for the acquirer's shareholders?
3. In a typical horizontal M&A deal, what share of the announced synergy number do cost synergies generally represent?
4. Select ALL correct answers about the post-COVID M&A environment described in the lesson.
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers about what happened in the Kraft Heinz / 3G Capital case.
Sélectionnez toutes les réponses correctes.
Two companies stand as the empirical refutation of the "70% of deals destroy value" statistic: Danaher and Constellation Software. Danaher has delivered roughly 21% annualized returns over 30+ years, with M&A as its primary growth engine. Constellation Software, under founder Mark Leonard, has compounded at over 30% annually since its 2006 IPO, executing more than 500 acquisitions of vertical market software companies.
Their playbooks share four characteristics that most acquirers ignore.
Danaher operates the Danaher Business System (DBS), a Toyota-derived operational toolkit applied to every acquired business within months of close. DBS is not a synergy plan; it is a value-creation engine that runs independently of synergy assumptions. Danaher CFO Matt McGrew has emphasized in recent earnings calls that the company underwrites deals on standalone improvement potential first, with synergies as a secondary contributor.
Constellation Software's approach is the inverse but equally systematic: it explicitly does *not* integrate. It buys small vertical software businesses, leaves management in place, applies capital allocation discipline, and demands hurdle rate IRRs. Mark Leonard's famous "hurdle rate" memo set the bar at high-teens IRRIRRThe Internal Rate of Return is the discount rate that makes a project's net present value equal zero. It expresses an investment's expected annualized return.View full definition → on every deal, with no exceptions for strategic narrative.
Danaher and Constellation both refuse to pay for synergies in the purchase price. This is the single most important discipline. The acquirer captures synergies only if the purchase price does not embed them. If the target was worth $1B standalone and you paid $1.4B, you've already paid the seller for $400M of "your" synergies. Anything below $400M of realized synergy is value destruction.
The Monday-morning test for any CFO: calculate the standalone DCF value of the target. If your offer exceeds it, the gap is the synergies you must deliver just to break even on the deal, not to create value.
The first acquisition is always hard. The fiftieth, for Constellation, runs on a playbook so refined that integration plans are templated. Roper Technologies, another serial acquirer with ~15% annualized returns, similarly runs a decentralized portfolio approach with central capital allocation.
Most corporates do 1-2 deals per decade and treat each as a unique snowflake. The lesson is not that everyone should become a serial acquirer, it's that *if you are not building repeatable integration capability, you are an amateur competing against professionals* in the same auction.
Constellation Software publishes a remarkable statistic: it evaluates roughly 100 deals for every one it closes. Danaher's hit rate is similar. The discipline of saying no, particularly when the banker has a "competitive process" countdown and the CEO has emotional ownership of the deal, is the single greatest source of value creation in M&A.
Theory is cheap. Here is what to actually do.
Build a synergy "validation memo" before signing. Every synergy line must have: (1) a named executive accountable, (2) a quantified baseline, (3) a realization timeline with quarterly milestones, (4) a cost-to-achieve estimate, (5) a stated reverse synergy or dis-synergy offset. If any line is missing one of these five, it doesn't go in the model presented to the board.
Apply probability weighting. McKinsey and Bain both recommend risk-adjusting synergy estimates. A reasonable framework: 80% probability for cost synergies in overlapping functions, 50% for cost synergies requiring system integration, 30% for revenue synergies. The board sees both the gross and the risk-adjusted number.
Separate the "deal model" from the "tracking model." The synergy model used to justify the deal must become the synergy tracker used by the integration team. Most companies build a beautiful deal model that is never opened again after close. The integration team then builds its own tracker with different baselines, and accountability evaporates. The CFO's job is to ensure these are the same document.
Mandate post-mortems. Two years after every deal, the finance team writes a memo comparing actual synergies to model synergies, with explanations for variance. Circulate it to the board. This single discipline changes deal-team behavior more than any pre-deal process.
1. Cap revenue synergies at 25% of the deal premium. If your model requires revenue synergies to justify more than a quarter of what you're paying above standalone value, the deal is built on the least reliable input in M&A. Restructure the bid or walk away.
2. Model reverse synergies explicitly. For every synergy bucket, require your team to quantify the value that could leak, customer attrition, key talent loss, channel disruption, strategic optionality forgone. A model without dis-synergies is incomplete, not