Your CEO heard at a conference that 3G Capital uses "zero-based budgeting" to find 20% of operational savings without sacrificing revenue. He calls you on a Monday and asks you to implement ZBB across the company before the next budget cycle. Four months. Ten departments. Three-person FP&A team.
Here you see two types of CFO: the one who says "yes" and starts immediately (and fails in 6 months when the team burns out and everyone reverts to "incremental + label it ZBB"), and the one who asks "in which cost categories and why?" before committing.
ZBB works — but only when applied with discipline and selectivity. Doing it on the WHOLE business every year is exhausting and ends up being theater. Doing it on 2-3 cost categories rotating each year finds the real value: savings where incremental had accumulated drift, and documented justification of each line for future board conversations.
In this module you learn the mechanics of honest ZBB, where it works, where it does NOT, and how to avoid the typical failure pattern. Let's go.
Your consultant promises that ZBB across the entire company will generate 15% recurring savings. What do you respond?
In plain language
Before the mechanics, the four basic questions.
Why do this at all?
For two distinct reasons: (1) Find REAL savings that incremental budgeting doesn't detect — accumulated drift, line items that no longer have purpose, retainers paid for idle capacity. Typically 10-15% in the analyzed cost categories. (2) DOCUMENT justification of each line for future conversations with CEO/board. Without ZBB, when they ask "why $X in marketing?", the answer is "it's what we spent last year +5%". With ZBB, the answer is "$X because we have N campaigns × $K each, justified by Y% expected IRR".
Who executes it?
Led by CFO + senior FP&A. The OWNER of each cost category (Marketing director, IT director, etc.) builds their own ZBB with FP&A guidance and challenge. NEVER "FP&A builds and presents to the department" — that guarantees the department rejects the numbers and reverts. The executive committee approves the consolidated proposal. Board audit committee (if it exists) reviews approach and results.
When does it show up?
Annual cycle with the AOP. BUT not in all cost categories every year — that's the fatal error. Rotation: year 1 marketing + travel, year 2 IT + general admin, year 3 supply chain + sales operations, year 4 back to marketing. Each cost category ZBB every 3-4 years. In between, incremental on the ZBB base from the last cycle.
What if we do it wrong?
Three modes of failure: (1) Theater — the team labels incremental budget as "ZBB" without actually rebuilding. CEO sees "we completed ZBB" in deck but the savings don't appear. (2) Burnout — full ZBB across the company burns out the team in 6 months; year 2 they silently revert. (3) Indiscriminate cuts — they interpret ZBB as "cut everything" instead of "rebuild from purpose." They cut valuable paid media, critical talent expenses, necessary tools. The business suffers in 12-18 months; the conversation becomes "ZBB destroyed the company."
Andina S.A. — selective ZBB on Marketing
Andina's CFO proposes to the board: year 1, ZBB on two cost categories — Marketing ($1.45M annual) and Travel ($380K). Why these two: Marketing because there's unexamined historical drift over 4 years; Travel because post-COVID the pattern changed but the budget returned to pre-COVID base by inertia.
For Marketing (the visual's example): five lines with incremental baselines. Team $525K, paid media $420K, events $168K, tools $84K, agency $252K. Incremental total: $1.45M.
ZBB rebuild led by the Marketing director with FP&A as challenger. Each line is rebuilt from purpose:
Team DECREASES $45K. Last year 5.5 FTE were approved; incomplete vacancy means incremental on actual already had unnecessary expansion. Explicit ZBB is 5 FTE × $96K loaded = $480K.
Paid media GOES UP $100K. Surprises the CEO. The last 3 years' marginal return analysis shows each incremental $100K in digital generates $400K in revenue. ZBB rebuild says: incremental was underestimating justified investment. Raise digital from $200K to $320K; keep traditional. ZBB is not just cutting.
Events DECREASE $93K. ZBB question: which events are critical for pipeline vs generic branding? 8 events × $20K (generic branding, hard-to-defend ROI) eliminated; 3 events × $25K for top accounts (clear ROI) maintained.
Tools DECREASE $22K. Inventory reveals 6 software licenses with no material use in the last 6 months. Cancel.
Agency DECREASES $72K. Historical retainer paid for idle capacity. ZBB rebuild: 4 big campaigns × $40K + 8 tactical pieces × $2.5K = $180K. Reduction.
Total ZBB: $1.32M. Variance −$131K (-9%). BUT the main value is not the nominal saving — it's that each line now has documented justification. When the CEO asks in 6 months "why $X on events?", the Marketing director answers with a table, not with "it's what we approved."
The visual below lets you see each line with its ZBB rationale. Tap any to expand.
Marketing, live
Five lines of Andina's Marketing budget. Left column shows "incremental" (last year + 5%). Right is ZBB rebuild.
Detail matters: click any line to see the specific ZBB rationale. Notice paid media GOES UP, not down — ZBB is not synonymous with "cut." It's synonymous with "rebuild from purpose."
Interactive visual
Marketing budget · Incremental vs ZBB
Five lines of Andina's Marketing budget. Left column is "incremental" (last year + 5%). Right is ZBB rebuild — each line reconstructed from purpose. Tap each row to see the rationale.
What you are seeing
Three critical lessons: (1) ZBB does NOT mean "cut." In this example, paid media GOES UP $100K because the ZBB analysis found it was worth investing more there. Cutting everything is the caricature; rebuilding from purpose is the real practice. (2) Total saved ($131K) is NOT the main value. The main value is that every line now has documented justification — defense before the CEO/board asking "why $X on events?". (3) Implementation: do NOT do ZBB on all departments at once. Pick 2-3 cost categories per year (marketing year 1, travel year 2, technology year 3). Rotating avoids exhaustion and maintains discipline. The company that does "full ZBB annually" on everything always reverts to incremental in year 2 from exhaustion.
The critical reading of the visual: the nominal variance of −$131K is secondary to the exercise. The real value is that five implicit conversations no one was having before ("why 5.5 FTE?", "why $200K digital when it's worth $320K?", "why retainer paying for idleness?") are now explicit, documented, and defensible before any internal audit or board question.
The second critical reading: doing this exercise over Marketing took the director and FP&A 3-4 real weeks of work. Multiply by 10 departments = 30-40 person-weeks. That's 7-10 people for 4 weeks, or 2-3 people during the full budget season. That's why "ZBB across the company" burns out the team and ends in theater. The opposite discipline: 2-3 cost categories per year, rotating. Still catches the drift; doesn't burn the team.
And critical: ZBB should NOT be used on cost categories with little drift and trivial justification. If your base is "legal minimum salaries × FTE needed to operate" — ZBB on that is zero value. Apply ZBB where there's accumulated drift, where components are discretionary, or where no one has questioned assumptions in 3+ years. Marketing, travel, agency, discretionary IT, training, consulting — those are the targets.
The mechanics: how to do ZBB without theater
- Pick 2-3 cost categories per year, not 10. Characteristics of a good candidate: material amount (>$500K annual), unexamined historical drift, discretionary components, or business change that invalidated prior assumptions. Marketing, travel, discretionary IT, consulting/agency, training — typically good candidates.
- The cost center OWNER builds their own ZBB. Marketing director builds marketing; IT director builds IT. FP&A does NOT impose — challenge, validate, push, but the operational owner is in the cost center. Without this, the department rejects the numbers and silently reverts.
- Each line is rebuilt from purpose + volume + unit cost. Team: # FTE × loaded. Events: # events × unit cost. Software: # licenses × price. Retainer: # deliverables × unit price. The key question: does this component have identifiable purpose and justified volume? If not, it goes.
- Document justification BEFORE approving the amount. Each ZBB line has a 1-2 paragraph note. It's not bureaucracy — it's the defense when the CEO asks in 6 months "why $X?". Without a note: you'll revert to incremental's "because."
- Compare ZBB rebuild with incremental. Document material variances (>10%). Variances are not automatically "cuts." Some are growth (paid media in the example). The report to CEO/board shows: by category, what changed and why.
- Once approved, ZBB is the new base. For next year of this cost category, incremental goes on adjusted ZBB — don't return to "pre-ZBB historical". If you do, the exercise was theater.
- Rotate cost categories every 3-4 years. Year 1: Marketing + Travel. Year 2: IT + General Admin. Year 3: Supply chain + Sales ops. Year 4: back to Marketing — and you'll find new drift. The perpetual rotation cycle is what maintains discipline without burning out the team.
- WORKS in: cost categories with historical drift (4+ years unexamined), discretionary components (events, retainer, consulting, training, paid media), or where the business changed (post-COVID travel, post-pandemic digital paid media, post-M&A integration of duplicates).
- Does NOT work in: legal minimum salaries, debt and debt service (depends on structure, not budget), taxes (legally determined), direct production costs (structural).
- Optimal frequency: each cost category, ZBB every 3-4 years. More frequent burns the team; less frequent allows drift reaccumulation.
- Optimal coverage: 2-3 cost categories per year in mid-market. Rotate each year.
- Who builds it: cost center operational owner, not unilateral FP&A. FP&A is challenger, not author.
- Practical rule: if after 6 months you can ask "why $X on line Y?" to any owner and get a quantitative documented answer, ZBB worked. If you get "because it's always been so" or "we approved it at start," ZBB was theater.
Adversarial check
Adversarial check
1.Your CEO wants ZBB across the company next cycle. Do you proceed?
2.Your Marketing director builds their ZBB and proposes raising paid media from $200K to $320K. What do you do?
3.After the first successful ZBB on Marketing (cut $131K), your CFO asks: "should we do ZBB on Marketing again next year?". What do you respond?
Exit checklist
Suggested re-review: annually at the start of the budget cycle. Define which 2-3 cost categories will go to ZBB this year and publish to the executive team 3 months in advance — they need the time to build properly.
Optional
Go deeper
Sources and books to dig into the original material