Zero-based budgeting done right

# Zero-based budgeting done right

In 2013, when 3G Capital and Berkshire Hathaway took control of H.J. Heinz, they did something that made the packaged-food industry flinch: they told every manager that the prior year's budget was worthless. Printer color quotas, the number of pages an employee could print per month, corporate jets, minibars in hotels on business travel, all of it went to zero and had to be re-earned. Within two years, Heinz's 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 → margin climbed from roughly 18% to nearly 28%. When the same playbook was applied to Kraft after the 2015 merger, it produced a very different story: brands starved of marketing, innovation pipelines gutted, and a $15.4 billion writedown in 2019. Same tool. Opposite outcomes.

That divergence is the entire subject of this lesson. Zero-based budgeting (ZBB) is neither the miracle cost machine its consultants sell nor the morale-destroying fad its critics decry. It is a precision instrument. Used on the right cost base with the right governance, it produces savings that compound for years. Used as a blunt annual ritual, it becomes theater, expensive, exhausting theater that trains your best people to game you.

What ZBB actually is (and what it isn't)

You already know the textbook definition: build the budget from zero, justify every dollar rather than incrementing off last year's baseline. That definition is correct and almost useless, because it describes a philosophy, not an operating system. What separates ZBB that works from ZBB that fails is not the "start from zero" mantra, it's the *unit of analysis*.

Traditional budgeting organizes spend by the org chart: this department got $4.2M last year, it needs 5% more this year. ZBB done right ignores the org chart entirely and reorganizes spend around two orthogonal structures:

Cost categories, spend grouped by *what is bought*, not who buys it. Travel, professional services, facilities, IT hardware, marketing agencies, real estate. This cuts horizontally across every department. The insight is that no single manager sees the total company spend on, say, external consultants, so no single manager is accountable for it, and it drifts upward for years.

Cost packages (or decision units), discrete bundles of activity with an owner, a defined output, and a cost. "The monthly close process." "North America field sales support." "The corporate travel program." Each package is examined on its own merits: what does it deliver, what would happen if it received 70% of its funding, 100%, 130%?

The reason this matters: incremental budgeting hides cost inside organizational silos. ZBB's real power is *visibility*, not zeroing. When you assign a single "category owner" to all travel spend across a $2B company, they suddenly discover the company books 340 last-minute fares a week and has 14 separate travel vendors. The savings don't come from managers heroically justifying each dollar. They come from consolidating fragmented spend that no one was watching.

This is the first correction to the popular understanding. ZBB is a transparency and accountability system disguised as a budgeting exercise. If your ZBB program doesn't create new cross-cutting ownership of spend categories, you're just doing painful annual budgeting.

Where ZBB creates durable savings

Not all costs respond to ZBB, and pretending they do is how CFOs destroy value. Segment your cost base before you begin.

The sweet spot: discretionary, fragmented, and low-strategic-differentiation

ZBB delivers its most durable savings on what practitioners call SG&A overhead and indirect spend: travel, facilities, corporate functions, professional services, IT infrastructure, marketing operations, office supplies, contractor labor. These costs share three traits that make them ideal:

• They are fragmented across owners, so they've never been optimized holistically.
• They accrete invisibly, a subscription here, a vendor there, and incremental budgeting locks the accretion in permanently.
• They are weakly correlated with revenue or strategic advantage. Cutting the number of software licenses nobody uses does not weaken your competitive moatmoatA lasting edge over competitors: a resource, capability or position they cannot easily replicate, letting a firm earn above-average returns over time.View full definition →.

Unilever's application is instructive here. Rather than a one-time slash, they built ZBB into a permanent capability with dedicated category owners who continuously benchmarked and re-justified indirect spend. The result was reinvestment: money extracted from waste flowed into brand and R&D. That is the tell of ZBB done right, savings are recycled into growth, not just dropped to the bottom line.

The danger zone: costs coupled to growth and capability

Here is where Kraft went wrong. When you apply the same zero-based scrutiny to marketing, R&D, sales capacity, and customer-facing headcount, you are no longer cutting waste, you are cutting the future. These costs *look* discretionary on a spreadsheet (you can, technically, spend zero on advertising next quarter) but they are strategically load-bearing. The savings are real and immediate; the erosion is real and delayed. That lag is what makes ZBB dangerous: the cut lands in this fiscal year, the brand equitybrand equityThe commercial value your brand adds beyond functional product attributes: the price premium, preference and loyalty it generates.View full definition → collapse lands three years later, long after the deal team has moved on.

The discipline: before a cost category enters the ZBB process, classify it on two axes.

| | Low strategic differentiation | High strategic differentiation |

|---|---|---|

| Fragmented / weakly governed | Attack hard (travel, indirect procurement, facilities) | Optimize carefully (sales ops, some IT) |

| Concentrated / well-governed | Modest gains (already managed) | Protect / grow (R&D, brand, key talent) |

The top-left quadrant is where ZBB earns its reputation. The bottom-right is where ZBB destroys companies. The tragedy is that both look like "cost" in the general ledger.

The practical how-to: running ZBB without breaking the company

Assume you've decided to run ZBB on your indirect and overhead base. Here is the operating sequence a CFO can actually execute.

1. Establish visibility before you establish targets

The single most common failure is announcing a savings target before you understand the spend. Spend eight to twelve weeks doing nothing but *cost transparency*: pull every dollar of indirect spend from the GL and AP data, re-tag it by category (not by cost center), and rebuild it bottom-up. Most organizations discover 15-25% more categories than they knew existed, and vendor counts that are 3-5x what leadership assumed. This diagnostic alone often surfaces "found money", duplicate contracts, auto-renewals, orphaned subscriptions, before any hard decisions are made.

2. Appoint category owners with real teeth

Assign each major cost category a single senior owner who is accountable for the *total company* spend in that category, independent of which department consumes it. This is the structural innovation. The Head of Travel now owns every travel dollar; the CIO owns every software license enterprise-wide. Give them the mandate to set policy, consolidate vendors, and approve exceptions. Without this cross-cutting ownership, spend re-fragments the moment the project ends and the savings evaporate within 18 months, the classic "ZBB bounce-back."

3. Build cost packages and force the trade-off conversation

For each decision unit, require the owner to describe funding at multiple service levels, say, 80%, 100%, and 120%, and articulate what output changes at each level. This is where judgment enters. The question is never "can we cut this?" (you always can). The question is "what specifically do we lose, when, and is that acceptable given strategy?" A well-run ZBB session sounds like: *"If we consolidate to two travel vendors, we save $3.1M but lose the ability to book premium-cabin exceptions for the M&A team during live deals, is that trade worth it?"* Documenting these trade-offs is what protects you from the Kraft outcome.

4. Set stretch targets grounded in benchmarks, not ambition

Targets should come from external benchmarks, what best-in-class companies spend on this category as a percent of revenue, not from a CEO's arbitrary "take out 20%." Arbitrary targets are precisely what turn ZBB into demoralizing box-ticking: managers who can't hit an unjustifiable number stop reasoning and start gaming, sandbagging their submissions so they have room to "concede." Benchmark-anchored targets keep the exercise intellectually honest.

5. Make it a capability, not an event

Durable savings require that ZBB become a recurring, lightweight process, category owners re-examining spend annually or continuously, rather than a brutal one-time reset. The one-time version produces a spike and a bounce-back. The embedded version produces a lower cost trajectory that holds. The difference between these two is the difference between a diet and a metabolism.

Where ZBB becomes demoralizing theater

Now the failure modes, because knowing when *not* to deploy ZBB, or how to stop it from rotting, is the higher-order skill.

Failure mode 1: The re-justification tax on stable, well-run spend. Forcing a team to rebuild from zero a budget that is already efficient, well-governed, and directly tied to output is pure deadweight loss. If your manufacturing line's direct materials budget is driven by a bill of materials and volume, making someone "justify it from zero" annually generates enormous documentation effort and zero savings. Reserve ZBB rigor for categories with genuine slack. Applying it universally signals that leadership can't distinguish waste from value, and your best operators notice.

Failure mode 2: Savings targets divorced from strategy. When the number comes down from on high with no logic, managers experience ZBB not as reasoning but as a compliance ritual. They produce elaborate justification decks whose real purpose is defensive theater. The tell: submissions get longer every year while decisions get no better. You've built a bureaucracy that consumes the savings it generates.

Failure mode 3: One-and-done with no ownership. The project team descends, extracts savings, declares victory, and leaves. Eighteen months later spend has crept back because no one owns the categories permanently. This is the most common quiet failure, the savings were real but temporary, and the organizational trauma was permanent.

Failure mode 4: Cutting the load-bearing walls. Discussed above, but worth restating as a failure mode: when ZBB is applied to strategic-differentiation spend under financial-sponsor pressure, the numbers improve while the enterprise hollows out. Kraft Heinz is the cautionary monument.

The through-line across all four: ZBB fails when it becomes about *the number* instead of about *the reasoning and the ownership*. The moment your people are producing justification for its own sake rather than making genuine resource-allocation decisions, you have converted a value-creation tool into a morale tax. As CFO, your job is to police that line continuously, to ask, in every cycle, "are we still learning something from this, or are we just performing rigor?"

Key Takeaways

1. ZBB is a transparency-and-ownership system, not a zeroing ritual. The durable savings come from assigning single owners to cross-cutting cost categories that no one was watching, consolidating fragmented indirect spend, not from managers heroically re-justifying line items.

2. Segment your cost base before you begin. Attack fragmented, low-strategic-differentiation overhead (travel, indirect procurement, facilities, IT infrastructure). Protect strategically load-bearing spend (R&D, brand, key talent), cutting it produces this year's savings and next-decade's decline. That is the Heinz-vs-Kraft lesson.

3. Anchor targets to external benchmarks, never to arbitrary edicts. Ambition-driven numbers with no logic are exactly what turn ZBB into demoralizing theater and teach your best people to sandbag and game the process.

4. Recycle savings into growth to signal intent. ZBB that only drops money to the bottom line reads as pure cutting; ZBB that reinvests waste into brand and capability builds durable competitive advantagecompetitive advantageA lasting edge over competitors: a resource, capability or position they cannot easily replicate, letting a firm earn above-average returns over time.View full definition → and organizational buy-in.

5. Build a metabolism, not a diet. Embed ZBB as a lightweight recurring capability with permanent category owners. The one-time slash produces a spike and an 18-month bounce-back; the embedded version produces a cost trajectory that holds.