It’s the third day of month-end close.
You’re staring at a row in Excel that defies logic. Marketing spend is up 15% against the plan, but you don’t see a corresponding bump in leads. You dive into the ERP transaction logs. You message the VP of Marketing. You wait. By the time you piece the story together, the board deck is due and the opportunity to pivot is in the rearview mirror. This scramble is the standard operating procedure for too many finance teams.
You know variance analysis is vital. It acts as your radar. It tells you when reality drifts from your assumptions and signals what needs to change. But the traditional workflow fights against you. You wrestle with static downloads and disconnected spreadsheets. Your budget lives in one place while your actuals live in another. Context gets lost in email threads. Budget variance is your feedback loop that compares actuals to budget and highlights reality splits. When that loop takes a week to close, you are reporting the news instead of making it.
We built this guide to turn that retroactive reporting into proactive strategy. It serves as your roadmap and practical playbook. We will explore the core definition of variance analysis and review the comparison frameworks that actually work. You will learn how to isolate specific drivers and close the gap between finding a variance and re-forecasting your plan. We will also show you how Runway integrates variance analysis directly into your workflow. It helps model owners and cross-functional teams stop chasing numbers and start driving performance.
What is variance analysis? (A quick refresher)
Variance analysis compares your actual results to a baseline plan. This baseline could be a budget, a forecast, or last period’s numbers. You get two key numbers: absolute variance (the dollar difference) and percentage variance (the relative change).
We label variances as favorable or unfavorable, depending on profit. If revenue falls short, it’s unfavorable. If expenses go over, it’s also unfavorable. But context matters here. A $200K overrun could mean poor planning, or it could signal growth. You have to find the root cause.
Great variance analysis shows the gaps and explains them. It tells you what happened, why, and what you should do next. That’s your loop: detect, diagnose, decide.
Budget vs. actual and forecast vs. forecast: Choose your comparison
Most teams start with budget vs. actual. You lock an annual budget, then compare each month’s actuals to those targets. It’s good for accountability. It measures performance against what you set out to do.
But budgets get old fast. By Q3, your January assumptions about hiring, pricing, or the market might not fit. That’s where forecast comparisons come in:
- Forecast vs. actual compares actuals to your latest forecast. You see how you’re tracking against your current view, not what you thought months ago. It helps you course-correct sooner.
- Forecast vs. forecast puts two forecasts side-by-side. Compare last month’s forecast to this month’s, or look at base case vs. an upside scenario. You see how your outlook changes and the impact of new assumptions before they turn into actuals.
- Period-over-period (month-over-month, quarter-over-quarter, year-over-year) reveals trends and seasonality. These views help you spot patterns, not just measure results.
Modern teams layer in all these comparisons. Use budget vs. actual for board reports, forecast vs. actual for operations, and forecast vs. forecast to plan scenarios. Runway’s scenario model lets you map each comparison to a different scenario and run them together. No duplicate models. Simple side-by-side views.
Key variances: What you should track
- Revenue variances show if you hit top-line targets. But one number doesn’t tell the full story. Break it down into volume (did you sell more units?), price (did you charge more?), and mix (did you sell a different product lineup?). A flat revenue number can hide shifts, such as high volume with lower price, or a mix change affecting profit.
- Expense variances reveal where costs go over or under plan. Go granular. Drill down to the vendor level with synced actuals. A top-line COGS jump could reflect a one-time buy (that actually lowers future costs) or an efficiency gap.
- Volume variances tell you what changes when quantity shifts. More sales. More production. More people. Volume is often the clearest to explain because it’s tied right to business activity.
- Price variances capture shifts in unit economics. Did you raise prices? Did vendor costs jump? These are key for margin tracking.
- Mix variances matter when your business mix shifts. Selling more low-margin products hurts profit even if revenue stays steady. You need segmented data to spot this.
- Timing variances show when things happened, not just if they did. Deals that move from Q1 to Q2 appear as a swing in both time periods. If you only check monthly, timing can trick you.
A variance analysis workflow in Runway
Runway’s variance workflow has three steps: lock your baseline, anchor your actuals, and compare.
- Step 1: Lock your budget. Lock forecasts as standalone scenarios. Prevent changes with scenario settings. Budgeting and forecasting need flexibility, but a finished plan shouldn’t change. Lock your scenario and you have a solid reference for variance.
- Step 2: Set up your actuals. Runway anchors actuals on
Last close. You compare closed periods (actuals) to forward periods (forecasts). This keeps you from mixing partial actuals and forecasts. Actuals sync from your accounting system, so you’re always using up-to-date closed month data. - Step 3: Compare budget vs. actuals. Build a BvA table or chart. In your driver table or chart, turn on Compare → variance, variance %. See time series values from each scenario. Add variance columns. Adjust the layout. Show scenarios as rows or columns. Flip the color scheme so higher can be better (for revenue) or lower can be better (for expenses).
For future comparisons, scenarios do the heavy lifting. Each scenario is a parallel model where you play with different assumptions. Compare values, add variance and variance %, and see how plans shift key lines. It’s how you check forecast vs. forecast, or model a new hiring plan before you commit.
Runway also gives you time comparisons for tracking periods. Run month-over-month, quarter-over-quarter, or year-over-year comparisons. Layer in custom rollups, like quarter-to-date or trailing twelve months. These quick views help you spot trends regular BvA misses.
Set thresholds: See what matters
A good variance report shows a hundred lines with differences, but not all deserve your focus. Use thresholds to focus on what counts.
- Most teams use both a dollar threshold and a percent threshold. Start by checking variances over $10,000 or 10% of budget. But your ideal limits depend on business size, stage, and line item. A 10% variance in rent may not matter. In revenue, it always does.
- You can set thresholds by category or team. For instance, marketing spend could push past payroll in volatility. Early teams look at cash burn first. Larger companies care more about margin swings.
Runway lets you set controls so variances stand out visually. Choose whether positive variance shows as green or red, depending on the metric. With revenue, higher is better. With expenses, lower is best. You set the goal at the driver, so your color coding shows what “good” means for every metric.
When you spot a material variance, drill-ins show the root data. One click. You see instantly how numbers flow through your model. If a driver comes from a single database, Runway surfaces those rows immediately. If you pull from several databases, Runway groups by source so it’s clear where data lives.
Break variances into drivers
A top-line variance is only the start. The real value comes when you know why it happened. Break the variance into component drivers (headcount shifts, price changes, deal timing, volume moves).
Say revenue drops $200K under plan. Why? It could be fewer closed deals (volume), lower average deal size (price), a shift to low-margin products (mix), or deals moved to next month (timing). Each cause drives a different decision.
Runway’s driver-based modeling makes this simple. Your model runs on human-readable formulas and dimensions. Drill into revenue and see the exact assumptions that missed, like deal count, average price, conversion rate. Drill-in variance rows add up cleanly, zeros drop to the bottom, and you focus on what moves the number.
Segment data by vendor, location, or team using dimensions in your databases. This lets you pinpoint budget variances by vendor. Is software spend over plan? See which vendors are driving the change, whether it’s a one-off or an ongoing trend.
Write variance commentary people actually read
Numbers drive insight, but stories drive action. Your team doesn’t want another static deck. They want to understand what’s happening, fast.
Effective variance commentary follows a clear structure:
- What happened? ("Revenue was $200K under budget.")
- Why did it happen? ("Three enterprise deals moved to next quarter.")
- What’s next? ("We’re moving up two mid-market deals to offset, and our Q2 forecast reflects the shift.")
Tailor your message to the audience. Board members care about strategy. Department leads want operations. Other teams need to know how the change affects their work.
Runway’s Pages let you blend data, narrative, and context. A Page mixes driver charts, tables, and live text blocks. Type @ and a driver name to reference live values. Your commentary stays connected and updates automatically when the data changes.
This turns static decks into living documents. Your variance report isn’t a snapshot. It’s your ongoing source of truth. Updated, current, and ready for action.
From analysis to re-forecasting: Close the loop
Variance analysis only adds value if it updates your forecast. Every variance serves as a signal. If you see repeats or patterns, they belong in your forecast.
Say customer acquisition costs run 20% above plan for three months straight. That isn't timing. It's a broken assumption. Update your driver formula in your core scenario. Create a new forecast to test the impact on cash runway so you can decide your next step for growth.
Runway connects actuals to modeling in real time. If a variance signals a broken assumption, update the driver. Make a new scenario and compare outcomes directly inside Runway. You don't need to duplicate models or export spreadsheets. You detect, diagnose, and decide in one place.
Plans connect your strategic changes to model results. Tag changes to your timeseries values with a plan to surface a timeline on a Page. You see when changes hit the numbers so you communicate clearly with your team.
Common mistakes and how to avoid them
- Don’t treat variance analysis as an autopsy. Top teams fold variance into their regular rhythm and check as soon as the books close.
- Never compare against a stale or unlocked budget. If someone changes your budget scenario after the fact, variance data loses meaning. Lock scenarios to keep plans clear.
- Separate timing from performance. Sometimes, a bad-looking variance in one month smooths out the next. Always compare year-to-date, not just monthly snapshots.
- View variance as strategic. Accounting can explain what happened. Operations show why. Put both together to get clarity. Collaborate in Runway, share Pages with teams, comment where the action is, and keep everyone on the same page.
- Act on what you find. If a variance doesn’t update your forecast, you haven’t closed the loop. The cycle from detection to action should take days.
Make variance analysis your edge
Variance analysis keeps your plan connected to real results. You’ll know the moment an assumption breaks or performance shifts. You’ll know what to do, fast.
Teams that win make variance analysis part of how they work, not just another report. They lock baselines, sync actuals, compare scenarios, drill into drivers, write clear takeaways, and close the loop, all in one place.
Runway handles the details. AI highlights discrepancies. Drill-downs trace each variance to the source. Scenario planning lets you test decisions before they’re real. Stop firefighting and start leading. Turn variance analysis into your planning superpower.
Want to see it in action? Talk to a real person and see how you can move from detection to decision in days, not weeks.