Sign in

Blog · Territory, capacity and quota planning

Zero-based budgeting: how it works, with a worked example

What zero-based budgeting (ZBB) means, where it came from, and how it differs from an incremental budget. A finance department is worked through ten decision packages to a funded budget, with the check that the funded packages sum to the approved total, the pros and cons, and how to see whether the savings showed up.

The short answerZero-based budgeting (ZBB) builds every cost line from zero each cycle: each activity is described as a decision package with its cost, ranked, and funded only if it justifies itself against a spending limit. Incremental budgeting instead takes last year's budget and adds a percentage. ZBB finds savings that incremental budgets carry forward, at the cost of more work.

Zero-based budgeting (ZBB) builds every cost line from zero each budget cycle, instead of taking last year's figure and adding a percentage. Each activity is written up as a decision package with its cost, the packages are ranked, and money goes down the ranking until a spending limit is reached. In the example below, a finance department that would have received $1,030,000 on an incremental budget is funded at $860,000.

What zero-based budgeting means

In a zero-based budget, no cost is carried forward by default. A manager has to say what each activity does, what it costs, what the alternatives are, and what happens if it is not funded. Whatever is not justified this cycle is not in the budget.

The method came out of Texas Instruments. It gained attention in 1970 when Peter Pyhrr wrote about the company's experience with it in Harvard Business Review. Jimmy Carter then used it for Georgia's state budget as governor, and as president told federal agencies in April 1977 that the Office of Management and Budget had issued guidelines for zero-based budgeting in the executive branch. The federal version proved heavy to run and was dropped in the 1980s; companies revived it later, mostly for overhead and discretionary costs.

Incremental vs zero-based

Incremental budget Zero-based budget
Starting point Last year's budget or actuals Zero
Question asked How much more or less than last year? Should we do this at all, and at what level?
Unit of work The cost line The decision package (activity, cost, alternatives)
Effort Low: a few days per department High: weeks of manager and finance time
What it misses Costs nobody would choose today, carried forward year after year Benefits that appear later, if packages are judged only on this year

Incremental budget = Prior-year budget × (1 + uplift %)

ZBB budget = Sum of funded decision packages, taken in rank order up to the funding limit

Saving against incremental = Incremental budget − ZBB budget

How ZBB works, step by step

  1. Identify decision units. The teams or cost centers that will each prepare packages: here, the finance department.
  2. Write decision packages. One per activity, with its cost, what it delivers, and at least one alternative: a cheaper way of doing it or a reduced level.
  3. Rank the packages. Mandatory work first (statutory filings, audit), then by value to the business. Finance challenges the ranking rather than accepting each manager's order.
  4. Set the funding limit. The total the business is willing to spend on this unit.
  5. Fund down the ranking. Packages above the line are funded; those below are recorded as unfunded, with what they would have cost.

A worked example: one finance department

Last year's budget was $1,000,000. An incremental budget at +3% would be $1,000,000 × 1.03 = $1,030,000. Instead, the department submits ten packages, and leadership sets a funding limit of $860,000:

Package Requested (USD) Alternative chosen Funded (USD)
1. Annual financial statements and tax filings (mandatory) 190,000 Full 190,000
2. Accounts payable, 3 staff 180,000 Automated invoice capture, 2 staff plus software 145,000
3. Credit control 125,000 Full 125,000
4. ERP licenses and support 90,000 Full 90,000
5. External audit (mandatory) 95,000 Full 95,000
6. Monthly management reporting package 110,000 Full 110,000
7. Treasury and cash forecasting 85,000 Full 85,000
8. Ad-hoc analyst 75,000 Not funded 0
9. Process consultants 60,000 Not funded 0
10. Travel and training 40,000 Reduced level 20,000
Total 1,050,000 860,000

The packages asked for $1,050,000 at full level, more than the incremental budget. Two choices of alternative and two packages below the line bring it to the limit:

  • Accounts payable on automated invoice capture: $180,000 − $145,000 = $35,000 less.
  • Travel and training at the reduced level: $40,000 − $20,000 = $20,000 less.
  • Packages 8 and 9 not funded: $75,000 + $60,000 = $135,000.

$1,050,000 − $35,000 − $20,000 − $135,000 = $860,000. Against the incremental budget, the zero-based budget is $1,030,000 − $860,000 = $170,000 lower, a 16.5% reduction ($170,000 / $1,030,000). Against last year's $1,000,000 it is $140,000, or 14.0%, lower.

The check: funded packages sum to the budget

Three things must reconcile before the budget is approved:

  1. Funded packages add to the approved total. $190,000 + $145,000 + $125,000 + $90,000 + $95,000 + $110,000 + $85,000 + $20,000 = $860,000, exactly the funding limit.
  2. Every funded package maps to budget lines. Package 2 at $145,000 is two salaries plus a software line in the ledger's cost centers, so actual spend can be compared with it later.
  3. Unfunded packages are kept on record. Packages 8 and 9, $135,000 in total, are the first call on any money that frees up during the year, and the first place to look if the work turns out to be needed.

Keep the approved ZBB budget as its own plan version, so later reforecasts do not overwrite the baseline the savings were measured against.

Pros and cons

For ZBB:

  • It removes costs that an incremental budget carries forward because nobody asked about them.
  • It forces a choice between ways of doing the work, as with automated invoice capture here, rather than funding the current way by default.
  • It ties each dollar to a stated activity, which makes later variance analysis specific.

Against ZBB:

  • The effort is large. Carter's own 1978 memorandum said it helped "the difficult task of allocating limited resources among competing and worthy programs", but the federal process was still dropped as too burdensome.
  • It favors short-term cuts. Training and maintenance are easy to rank low, and the cost comes back later.
  • Rankings can be gamed: a manager who ranks a must-have package low, expecting it to be rescued, protects a pet project above it.

Where it goes wrong

  • Rebuilding every line every year. The effort exhausts managers. Many firms run full ZBB every few years, or each year on selected cost pools only.
  • Cutting what pays back later. Training, maintenance and controls look discretionary in one year's ranking and return as cost the next.
  • Managers ranking their own pet packages first. The ranking needs challenge from finance, not just collation.
  • Counting the budget cut as the saving. A lower budget is a plan. Check actual spend against each package during the year; the method is in savings claimed against savings realized.
  • Applying ZBB to volume-driven costs. Freight, commissions and payment fees move with sales; they need a driver-based budget, as in cost to serve by industry, not a fixed package.

After the budget: were the savings realized?

The $170,000 only exists if actual spend follows the packages. Each month, compare the ledger with each funded line: is accounts payable running at two staff plus software, or did a temporary hire bring back the third clerk? Spend booked against packages 8 and 9, which were not funded, is spend outside the budget; maverick spend by requester shows who is incurring it, and spend under management covers the procurement side of holding the cost base.

Covirage compares actual spend from the ledger with each funded line in the uploaded budget, with tools computing every variance, so a ZBB saving is checked in the actuals rather than assumed; the external AI model explains the variances and never calculates them. Upload the budget and the actuals, and see which cost lines are running over and why, with the rows behind each: cost variance analysis. To build the budget itself, start from the business budget template, and track it during the year with the budget vs actual template.

Questions people ask

What is an example of zero-based budgeting?

A finance department that would have received last year's 1,000,000 plus 3% instead lists each activity with its cost, ranks them, and funds the top ones up to 860,000, choosing automated invoice capture over a third clerk and dropping external consultants.

What is the difference between incremental and zero-based budgeting?

Incremental budgeting starts from last year's budget and adjusts it. Zero-based budgeting starts from zero and requires every cost to be justified each cycle. Incremental is faster; ZBB is better at removing costs nobody would choose to incur today.

What are the disadvantages of zero-based budgeting?

It takes much more time and management attention, can encourage short-term cuts to training or maintenance, and depends on honest ranking. Many companies limit it to discretionary costs or run it every few years rather than annually.

Who invented zero-based budgeting?

Peter Pyhrr developed it at Texas Instruments in the late 1960s and described it in Harvard Business Review in 1970. Jimmy Carter adopted it as governor of Georgia and later for the US federal budget.