Budget vs actual analysis: how to run a variance review

Budget vs actual analysis compares what the company planned with what actually happened, line by line, for a given period, and variance analysis explains the differences that matter: what caused them, whether they will recur, and what to do about them. Most companies produce the report every month; far fewer use it to change anything. The report arrives late, the commentary restates the numbers ("revenue was $90K below budget"), and the meeting ends without a decision. This playbook is for the CFOs, controllers and finance leads who own that monthly review.

What budget vs actual analysis is

Budget vs actual (often shortened to BvA) is one of the core monthly routines of FP&A, and it works at the line level. The difference on each line between plan and result is a variance, and variance analysis is the work of explaining the ones above the threshold.

The report is the easy part. A BvA that ends at "the company missed" is bookkeeping. One that ends at "the company missed because two go-lives slipped, they land next month, and the forecast reflects it" is management.

Budget vs forecast vs actual

The same numbers support three comparisons, and each answers a different question.

• Budget vs actual measures results against the plan agreed at the start of the year. It is the accountability view: did the company deliver what it committed to?
• Forecast vs actual variance analysis measures results against the latest forecast. An actual vs forecast variance shows how well finance and budget owners predicted the period, which is how forecast accuracy gets tracked and improved.
• Budget vs forecast variance analysis compares the latest full-year forecast with the budget. It shows where the year is now expected to land against the commitment, months before the actuals arrive.

Use budget vs actual for the monthly review and the board, forecast vs actual to judge how far the forecast can be trusted, and budget vs forecast to catch drift while there is still time to act. To analyze budget performance properly, keep the budget fixed and the forecast current; the report layout below carries the month and year to date against budget alongside the full-year forecast against budget.

Favorable, unfavorable and sign conventions

A variance is favorable when it helps profit and unfavorable when it hurts it. Revenue above budget is favorable. Expenses above budget are unfavorable.

The arithmetic differs by line type, which is where reports get confusing:

• For revenue and profit lines: variance = actual minus budget.
• For expense lines: variance = budget minus actual.

With that convention, a positive number is always favorable and a negative number, shown in parentheses, is always unfavorable. Readers never have to work out whether a positive expense variance is good or bad.

Some companies use actual minus budget on every line instead, which makes expense overruns positive. Either works if the report states the convention and never switches.

Types of variance

Price, volume and mix

A revenue or cost variance usually has more than one cause. Splitting it shows which lever moved.

• Volume variance: more or fewer units than planned (customers, seats, hours, orders), valued at the budgeted price.
• Price variance: a different price or rate per unit than planned, applied to the actual units.
• Mix variance: a different blend of products, plans or customer types than planned, each with its own price or margin.

A small net variance can hide two large offsetting ones. Revenue on budget might mean more customers at deeper discounts. That is a pricing conversation, not a non-event.

Timing vs permanent

The second split is about whether the variance will reverse.

• Timing variances shift money between periods. A project that goes live a month late, an invoice booked early, a conference paid in March instead of April. Over the year, they net to roughly zero.
• Permanent variances will not come back. A lost customer, a price cut, a new contractor, a one-time legal bill. They change the full-year outcome.

Calling a variance timing is a claim about the future, so name the month it reverses. If it has not reversed by then, it was permanent.

Materiality thresholds

Not every variance deserves commentary. A materiality threshold decides which lines get explained, so the review spends its time on what matters.

There is no standard threshold. Each company sets its own based on its size, margins and what leadership and the board care about. The rule can be a dollar amount, a percentage of the budget line, or a combination. Write it down, apply it every month, and revisit it once a year. Still look below the threshold where offsetting variances net to something small, and let owners flag smaller variances that signal a trend.

How to build the BvA report

Structure

Build the report from the summary down:

1. One-page summary: revenue, gross profit, operating expenses, operating income, cash, and the few operating metrics that drive them.
2. Income statement by line, in the same format as the budget.
3. Owner views: each budget owner sees the lines they control.
4. Commentary: cause, type, owner and action for every variance above the threshold.

Columns

Each line should answer three questions: how did the month go, how is the year going, and where will the year end.

The full-year columns matter most and are most often missing. A year to date can look fine while the full year drifts, and only the forecast column shows it.

Budget vs actual report template

A budget vs actual report template that works month after month has the same pages in the same order, the same columns on every financial page, and a commentary page that ends in decisions. The layout below is a starting point for a company with a single entity; add a consolidation page for each extra entity.

Report layout, page by page

Set the sign convention once, at the top of the summary page, and hold the format fixed. Changing the layout every few months hides the trends the report exists to show.

Commentary template

The commentary page is where most reports fall short. A fixed row format forces each explanation to reach a decision:

Illustrative figures, not client data.

The full-year effect column ties the commentary to the forecast: 35 a month for three months is 105 (35 × 3). If the commentary says a variance is permanent, the forecast should already show it.

To see this layout filled in, with an executive summary, revenue plan, cash and runway, and actuals vs budget pages, get the sample budget board deck, an excerpt of the budget package Decipher builds for clients.

The monthly review meeting

Who attends. The finance lead, the CEO or general manager, and each budget owner whose lines crossed the threshold. Owners on plan can skip.

When. As soon after the close as the numbers are reliable. Every week of delay means decisions get made without it.

A workable agenda:

1. Headline: full-year forecast vs budget, and what changed since last month (5 minutes).
2. Material variances, largest first, each explained by its owner, not by finance (20 minutes).
3. Open actions from last month: done, late or dropped (5 minutes).
4. Decisions and changes to the forecast (10 minutes).

Owners and actions. Every material variance ends with one of three outcomes: no action (timing, with a named reversal month), a forecast change, or an action with an owner and a date. Finance keeps the action log and opens the next meeting with it.

Reforecasting vs re-budgeting

When results drift, the temptation is to rewrite the budget. Usually that is the wrong move. The budget is the yardstick agreed at the start of the year. It stays fixed so performance can be measured against a stable commitment. The forecast is the best current view of where the year will land, updated monthly or quarterly as actuals come in.

Reforecasting keeps both, and a rolling forecast is the usual way to do it. Re-budgeting erases the original commitment and makes variances look smaller than they are.

Re-budgeting is reasonable after a genuine reset: a financing, an acquisition, a major strategic change, or a shock that makes the original plan irrelevant. When it happens, keep the original budget on file, label the new version clearly, and tell the board why.

A worked example

A software company is six months into its fiscal year. It has chosen to comment on any line where the monthly variance exceeds $20K. The convention is the one above: positive is favorable, parentheses are unfavorable.

The month

Illustrative figures, not client data.

Total operating expenses landed within $3K of budget. A review that stopped there would miss three material variances inside it.

Subscription revenue: price and volume

Subscription revenue is below the threshold, but it nets two larger movements.

Illustrative figures, not client data.

• Volume variance: 40 more customers × $2.1K = $84K favorable.
• Price variance: $0.1K lower per customer × 1,040 customers = $104K unfavorable.
• Net: $84K minus $104K = $20K unfavorable.

The company is winning more customers than planned at lower prices, from discounting on new deals or a shift toward smaller plans.

From variances to decisions

The sales and marketing line is the one to slow down on. The $60K underspend looks like savings, but those two sellers were meant to be building pipeline. Their late start will show up as lower bookings next quarter, so the favorable cost variance comes with a revenue cost later. Tying each role to the revenue it supports, as in headcount planning, makes that link visible before the month closes.

Year to date and full year

Illustrative figures, not client data.

Year to date, operating income is $10K better than budget. That looks like a company on plan. The full-year columns tell a different story: the revenue gap widens from $280K to $800K because the delayed hires and discounting carry forward, while the expense savings stay modest. The year is forecast to end $350K below budget on operating income.

That is the headline for the leadership team and the board: a company on plan to date, drifting off plan for the year, with the causes identified and the actions assigned.

Common mistakes

• Commentary that restates the table. "Revenue was 3.8% below budget" adds nothing; explain cause, type and action.
• Netting offsetting variances. Split the drivers on lines that matter.
• Permanent variances labeled timing. Name the reversal month and check it.
• No forecast column. The review only looks backward.
• Finance explaining everyone's numbers. Owners explain; finance tests the explanations.

How an investor or board reads your BvA

Boards and investors see the BvA less often and with less context than management, and they judge the finance process as much as the results.

• The full-year forecast trend first. They want to know where the year lands and whether that view moved since last time. A forecast that steps down a little every month without explanation erodes trust faster than one honest reset.
• Whether timing variances reverse. They remember last quarter's "timing" items. If those variances never come back, every later timing explanation gets discounted.
• Commentary that leads somewhere. Actions named in one meeting should show up as results, or as explained changes, in the next.
• Consistency. The same format, definitions and sign conventions every period, with the original budget still visible. A format that changes every month hides the trends they are looking for.

In diligence, including the quality of earnings review that comes with most sales, an acquirer or new investor may ask for a history of budgets, forecasts and actuals to judge how well management plans. A company that has kept those versions, with commentary, has an easy answer. Decipher's FP&A Lead engagement runs this monthly cycle under an existing CFO or controller, including the report, the reforecast and the board-ready commentary. To see where the monthly finance process stands today, the finance scorecard takes about two minutes.