Split a mixed cost into fixed and variable parts using your highest and lowest activity levels
💡 Why This Matters Many costs are a mix of fixed and variable — rent stays flat, but utilities scale with production. The high-low method gives you a quick way to split the two apart using just your highest and lowest activity periods.
The high-low method only ever looks at two data points, no matter how many months of cost history you have — which makes it fast, but also means one unusual outlier month can throw off the whole estimate.
As an Amazon Associate we earn from qualifying purchases.
We do our best to link products that we genuinely believe are organic, no disposable junk, sustainable, and built to last.
The high-low method uses only your highest and lowest activity levels — not the highest and lowest costs — to split a mixed cost apart. First, variable cost per unit = (high cost − low cost) ÷ (high units − low units). Then, fixed cost = high cost − (variable cost per unit × high units). Once you have both, total cost at any volume = fixed cost + (variable cost per unit × units).
Variable cost per unit tells you how much each additional unit of activity adds to your total cost — things like materials or hourly labor. Fixed cost is what you'd pay even at zero activity — rent, salaried staff, insurance. Together they form a simple cost model you can use to forecast total cost at any production level.
This method only uses two data points out of however many months or periods you have, so it's vulnerable to outliers — an unusually expensive or unusually cheap month at either extreme skews the whole estimate. It also assumes costs behave in a straight line, which breaks down outside the range you actually observed. For more precision, methods like regression analysis or a scatter plot review use every data point instead of just two.
Use the fixed and variable cost split to build budgets, price products, or forecast costs at new production levels. Treat it as a fast estimate for planning purposes — if the numbers will drive a major pricing or budgeting decision, cross-check them against a more detailed costing method first.
Example: Highest production level costs $1,000 at 100 units; lowest costs $700 at 50 units. Variable cost per unit = ($1,000 − $700) ÷ (100 − 50) = $6 per unit. Fixed cost = $1,000 − ($6 × 100) = $400. So the cost model is $400 + $6 × units — at 100 units, total cost = $400 + $600 = $1,000, which checks out against the original high-activity cost.
Variable cost per unit = (highest cost − lowest cost) / (highest units − lowest units). Fixed cost = highest cost − (variable cost per unit × highest units). Total cost at any volume = fixed cost + (variable cost per unit × units).
It oversimplifies the relationship between cost and production activity by only using the highest and lowest data points, ignoring everything in between and any outliers those two points might represent.
No — other methods include the scatter plot method and regression analysis, both of which use every data point rather than just the extremes. The high-low method is simpler and faster but less precise.