Revenue function: the essential idea
A revenue function describes the money a seller receives from sales at different quantities or prices. If a business sells q units for a constant price p per unit, total revenue is R(q) = p × q. If the price changes with the number of units customers will buy, substitute the price relationship into that product: R(q) = q × p(q). This simple distinction is the key to understanding why some revenue graphs are straight lines while others bend into parabolas. Revenue is money received from sales before costs are subtracted; it is not profit.
This guide is for students of college algebra, introductory economics, and business mathematics who need to build a function from a word problem, interpret its graph, and avoid the most common mistakes. We will work through constant-price sales, price-dependent demand, maximum revenue, cost and break-even, and marginal revenue. Each example defines its units and a realistic domain before drawing conclusions. The examples are invented for teaching; they are not forecasts of an actual business.
The underlying definitions and algebraic approach agree with OpenStax’s college-algebra discussion of revenue, cost and profit. For an official open-textbook example of maximizing a quadratic revenue model, consult OpenStax’s section on quadratic functions. Those sources use their own examples; the worked numbers below are original so that you can practice the method rather than memorize a published solution.
Define the variables before writing an equation
Many errors in revenue problems are really unit errors. Let q be the number of units actually sold during a specified period, such as one week. Let p be dollars per unit. Then p × q has units of dollars for that week. If q is measured in thousands of units, the same numerical multiplication produces thousands of dollars unless you convert the units. If the item is a subscription, decide whether p is per month or per quarter and whether q represents subscriptions, subscribers, or subscriber-months. A formula is meaningful only when those units match the story.
The quantity sold is not necessarily the quantity produced. Inventory, returns, unsold stock and capacity can separate them. In a basic algebra exercise, a problem may assume everything produced is sold; if so, state that assumption before using the same q in both a revenue and a cost function. In a real business, gross sales, net sales, taxes, discounts and refunds create additional distinctions. Introductory revenue functions usually model the gross amount received from the transactions described in the question and leave those complications outside the simplified model.
A function takes an input and returns one output. R(50), for example, means the model’s predicted revenue at a quantity of 50, not the product of R and 50. If price depends on quantity, p(q) gives the per-unit price associated with selling q units. Then R(q) = q p(q) combines a per-unit amount with a unit count. If quantity depends on price, q(p), you may instead write R(p) = p q(p). Both formulations can describe the same demand relationship, but the input variable and the units on the horizontal axis are different.
Before optimizing or interpreting a graph, state the domain. Quantity normally cannot be negative. A store may have a capacity limit, and a linear demand formula may predict negative sales if extended too far. Those invalid values must be excluded. A graphing tool can draw a parabola forever; the business model cannot. The same caution applies to fractional values when items must be sold in whole units. Mathematics gives a continuous candidate, but an actual decision may require checking neighboring integers.
The constant-price revenue function
Suppose notebooks sell for $20 each and every notebook in the modeled range sells at that price. The revenue function is R(q) = 20q. At q = 0, revenue is zero. At q = 10, it is $200. At q = 75, it is $1,500. The graph is a straight line through the origin with slope 20 dollars per notebook. Every additional unit sold adds $20 to revenue as long as the price really remains fixed and the new unit is sold. The slope is not profit per notebook, because manufacturing, shipping and other expenses have not yet been deducted.
This line has no interior maximum if the model permits arbitrarily many units. If a question says the shop can sell at most 80 notebooks during the week, restrict q to whole numbers from 0 through 80. The greatest modeled revenue then occurs at the endpoint, q = 80, and equals $1,600. Without that capacity or another constraint, asking for a finite maximum from R(q) = 20q is incomplete. The line keeps rising as q rises, even though the underlying constant-price assumption would eventually become unrealistic for a real market.
The formula also explains how to work backward. If the store’s recorded notebook revenue is $1,200 at a fixed $20 price and no refunds are included, divide 1,200 by 20 to infer that 60 notebooks were sold. This inverse step depends on knowing the price. A total revenue number alone cannot reveal quantity if multiple products, promotions, or variable prices were involved. When a task asks for units, write the equation first, insert the known values and check the inferred quantity against the allowed domain.
Constant price is a useful first model for a price-taking seller or a short interval with a posted price. It should not become a universal rule about revenue. A quantity discount, a ticket-price change, or a downward-sloping demand relationship means p changes as q changes. Then the slope of a revenue graph is generally not equal to the current price. That is the crucial bridge to the next model.
When price depends on quantity
Imagine a small event organizer estimates that it can sell q tickets if it charges p(q) = 40 − 0.10q dollars per ticket. The model says the price compatible with selling more tickets falls by ten cents for each extra ticket. This is an inverse-demand relationship, not a promise that changing price will cause exactly that many sales. The model is a simplified assumption built from observations or a problem statement. Its reasonable domain has 0 ≤ q ≤ 400, because beyond 400 the formula would imply a negative price. Capacity might restrict the domain further.
Multiply price by quantity: R(q) = q(40 − 0.10q) = 40q − 0.10q². The negative coefficient on q² makes the graph a downward-opening parabola over its practical domain. At q = 100, the model price is $30 and revenue is $3,000. At q = 200, the price is $20 and revenue is $4,000. At q = 300, the price is $10 and revenue returns to $3,000. Selling more tickets does not always raise total revenue, because reaching those additional buyers requires a price reduction on every ticket under this model.
At q = 0, the mathematical price is $40 but there are no sales, so revenue is zero. At q = 400, the price is zero, so revenue is also zero. Neither endpoint is an attractive selling strategy; they mark the boundaries of the simplified demand relationship. The vertex between them is the modeled maximum revenue. In a real venue, demand may not be perfectly linear, customer groups may pay different prices, and the event has a seat limit. Use the vertex as an answer to the mathematical model, not as an unqualified commercial recommendation.
You can visualize the pattern by making a small table of q, p(q) and R(q), but keep the columns distinct. Price falls as quantity rises; revenue first rises and then falls. A common mistake is to graph p(q) and call it the revenue graph. Another is to multiply q by the original $40 intercept instead of the changing p(q). Write the units next to each column: tickets, dollars per ticket, and dollars. That simple habit makes the product relationship visible.
Build a revenue function from two observations
Word problems often provide two price-and-quantity pairs instead of a ready-made p(q). Suppose a seller estimates that at $10 it can sell 120 tickets and at $12 it can sell 100. Under a linear-demand assumption, quantity changes by −20 tickets for a $2 price increase, or −10 tickets per dollar. Therefore q(p) = 220 − 10p: substituting p = 10 gives 120, and substituting p = 12 gives 100. Checking both observations catches many intercept errors.
Because p is the input in this version, revenue is R(p) = p(220 − 10p) = 220p − 10p². If p is in dollars per ticket and q(p) is tickets, R(p) is dollars. A nonnegative price and nonnegative predicted quantity give 0 ≤ p ≤ 22. If tickets can only be priced in whole dollars, the candidate set is the integers in that interval; if cents are allowed, the pricing options are more numerous. State which assumption you are using before reporting a precise optimum.
The two data points do not prove demand is linear outside the observed range. The equation is an extrapolation. If you were advising a business, you would test sensitivity to that assumption, consider capacity and competing events, and account for costs. In a classroom problem, you normally accept linear demand because it is specified or implied, but good mathematical communication still labels it as a model. This is why a correct algebraic answer should include a sentence interpreting the input and the circumstances under which the result makes sense.
Find maximum revenue with a quadratic vertex
A downward-opening quadratic R(x) = ax² + bx + c, with a < 0, has its maximum at x = −b/(2a), provided that x lies in the allowed domain. For the ticket price model R(p) = −10p² + 220p, the vertex is p = −220/[2(−10)] = 11. At $11 per ticket, q(11) = 220 − 110 = 110 tickets, and revenue is 11 × 110 = $1,210. Compare the two original observed points: $10 × 120 = $1,200 and $12 × 100 = $1,200. The vertex gives $10 more revenue than either within this model.
The same outcome can be seen by completing the square: R(p) = −10(p − 11)² + 1,210. A squared term is never negative, so subtracting ten times that square makes revenue no greater than $1,210, with equality at p = 11. This form also shows symmetry: prices one dollar below and above $11 give the same modeled revenue. Symmetry is a feature of the assumed quadratic, not a claim that actual customers behave identically around a chosen price.
For the quantity-input model R(q) = 40q − 0.10q², the vertex is q = −40/[2(−0.10)] = 200. The price associated with that quantity is p(200) = $20, and revenue is $4,000. Be careful not to confuse these two examples: one uses p as the input and peaks at $11, while the other uses q as the input and peaks at 200 tickets. In both, the calculation is only the beginning. You must report the relevant input, its units, the corresponding output, and any domain or integer adjustment.
If the vertex falls outside a restricted domain, it is not the answer for the practical problem. For example, if a hall in the second model seats only 150 people, you cannot sell 200 tickets even though the unrestricted parabola peaks there. Within 0 ≤ q ≤ 150, revenue is still rising toward q = 150, so the constrained maximum is at 150, with price $25 and revenue $3,750. If the vertex is at 199.6 but q must be an integer, evaluate q = 199 and 200 rather than claiming that six-tenths of a ticket can be sold.
The site’s quadratic-functions lesson is a useful review of vertex form and graph shape. Use it after building the revenue equation correctly; a graphing method cannot rescue a mistaken demand equation. The OpenStax Algebra 1 modeling lesson also emphasizes interpreting the domain, vertex and zeros in context rather than treating them as free-floating numbers.
Zeros, slope and graphs in context
In a constant-price model R(q) = 20q, the revenue graph passes through (0, 0) and has a constant positive slope of 20. The point (50, 1,000) means selling 50 notebooks produces $1,000 of revenue at the specified price. Do not read the horizontal coordinate as dollars or the vertical coordinate as units. Label axes before plotting. If quantity is in thousands, label it accordingly and interpret the vertical scale with equal care.
The variable-price model R(q) = 40q − 0.10q² crosses the horizontal axis at q = 0 and q = 400. The first zero means no units are sold. The second means the assumed price p(400) is zero; it does not mean 400 tickets are given away profitably. Between those values the graph rises to its vertex and falls. Outside them the algebraic expression can return negative revenue, but that part of the polynomial is outside the intended business domain. A graphing calculator can help confirm the shape and intercepts, but the domain comes from the story.
The slope interpretation changes when price varies. At a constant price, the rise per additional unit equals p. For R(q) = 40q − 0.10q², each extra ticket changes revenue by a different amount depending on q because the associated price changes across all sales. At lower quantities, selling more can more than offset the lower per-ticket price; near the vertex, the effects balance; beyond it, the price reduction dominates. Avoid the claim that “the slope of every revenue function is price.” That is true for R(q) = pq only when p is constant.
You may use the site’s quadratic equation calculator to check a vertex or roots after showing the setup, but keep the meaning of each input and solution in view. A calculator can find a negative root; the business question still requires rejecting an impossible negative quantity. A numerical answer without a unit and interpretation is incomplete, even when the arithmetic is perfect.
Revenue is not cost, profit or cash in the bank
Revenue measures sales receipts in the model. Cost measures the expense of operating and producing or supplying those sales. Profit is P(q) = R(q) − C(q). For a simple notebook business with a $300 weekly fixed cost and $8 variable cost per notebook, C(q) = 300 + 8q. At a fixed $20 selling price, R(q) = 20q and P(q) = 20q − (300 + 8q) = 12q − 300. Selling 50 notebooks produces $1,000 in revenue, costs $700, and therefore yields $300 in modeled profit. Reporting $1,000 as profit would ignore the cost function entirely.
Fixed cost is incurred even when q = 0 in this example. Variable cost rises as more units are sold or produced. At q = 0, revenue is zero and modeled profit is −$300. This is why a revenue graph beginning at the origin does not imply the business is breaking even at zero units. A break-even point is where revenue equals cost, not where revenue itself happens to be zero. In a real accounting statement, expense timing and inventory treatment may be more complex; the simple functions are instructional approximations.
The same lesson applies to a maximum. The quantity that maximizes revenue need not maximize profit. If a price cut generates more sales but also adds substantial variable costs, profit may peak earlier. If demand, costs and capacity are all modeled, optimize the profit function for a profit question and the revenue function for a revenue question. The words in the prompt determine the target. “Highest sales dollars” and “highest earnings after costs” are not interchangeable goals.
Break-even analysis step by step
For the fixed-price notebook example, set R(q) = C(q): 20q = 300 + 8q. Subtract 8q to obtain 12q = 300, so q = 25. At 25 notebooks, revenue is $500 and cost is $300 + $200 = $500. Profit is zero. For q below 25, the simplified model shows a loss; above 25, a profit. This is a whole-number break-even point because a notebook cannot usually be sold in fractions. If the algebra instead yielded 25.4, the first whole-unit quantity above break-even would be 26, assuming the cost and price formulas remain valid.
A graphical method places the revenue and cost lines on the same axes. Their intersection is the break-even point. The horizontal coordinate gives units, and the vertical coordinate gives the common revenue and cost at that quantity. This differs from the x-intercept of either line alone. In an algebra problem, the intersection can be found by solving a system of equations; a graph then confirms and interprets it. The site’s break-even analysis guide explores the broader business setting, while the equations here show the college-algebra mechanics.
If price depends on quantity, R(q) may be quadratic and C(q) may be linear. Their intersection equation can then have two roots, one root, or no real root, depending on the assumptions. Two break-even quantities are not automatically an error: profit might be negative at very low sales, positive at intermediate sales, and negative again if the price required to sell very high quantities becomes too low. Still, only roots in the valid domain have practical meaning. A mathematical root outside capacity or at a negative price must be rejected.
Break-even analysis itself does not tell a manager which price to charge. It states where the modeled income and costs are equal. Demand uncertainty, fixed-cost estimates, taxes, returns, labor constraints and competitor responses all affect a real decision. For exam purposes, solve the equality, verify by substitution, and explain whether the solution is a minimum feasible whole-unit sales target or one of several intersections. For real planning, treat it as a scenario model whose assumptions deserve scrutiny.
Maximum profit can occur at a different quantity
Return to the event model with p(q) = 40 − 0.10q and R(q) = 40q − 0.10q². Suppose the event also has fixed costs of $500 and an additional cost of $8 per attendee, so C(q) = 500 + 8q. The profit function is P(q) = R(q) − C(q) = −0.10q² + 32q − 500. Its vertex is q = −32/[2(−0.10)] = 160, not the 200-ticket quantity that maximizes revenue. At q = 160, the model price is $24, revenue is $3,840, cost is $1,780, and profit is $2,060.
At q = 200, revenue reaches its $4,000 maximum, but costs are $2,100, leaving profit of $1,900. The seller gains $160 more modeled profit at 160 tickets even though total revenue is $160 lower. This is not a contradiction: the 40 additional attendees add $320 in modeled cost while the price changes. Comparing those two scenarios makes the distinction between objectives concrete. If a problem asks for the maximum profit, optimizing R(q) alone answers the wrong question.
The price at q = 160 follows from the same demand model: p(160) = $24. Do not report $160 as a price because the vertex variable is quantity here. A good final sentence names the quantity, associated price, revenue, cost and profit, then states the assumptions: linear inverse demand, a constant $8 variable cost, and the stated fixed cost. If tickets must be whole numbers, 160 is already an allowable integer. If the vertex were not integral, check the nearest feasible quantities.
This example also shows why maximizing a function is not a complete business plan. A real venue might have a capacity below 160, minimum attendance obligations, price tiers or fixed costs that change with staffing. The algebra faithfully solves the stated model; it does not validate the assumptions automatically. Teachers often value that distinction because it demonstrates understanding beyond symbol manipulation.
Marginal revenue and the next unit
Marginal revenue asks how revenue changes when output or sales quantity changes slightly. In a calculus course, marginal revenue is the derivative dR/dq. For R(q) = 40q − 0.10q², the derivative is 40 − 0.20q dollars per additional ticket near q. At q = 100, the derivative is $20; at q = 200, it is zero; beyond 200 it is negative. The zero derivative agrees with the vertex found by algebra. The derivative is a rate from the model, not the actual ticket price, which at q = 100 is $30.
If calculus is not part of your course, compare consecutive whole-number revenues instead. R(q + 1) − R(q) equals 39.90 − 0.20q for this quadratic. At q = 100, moving from 100 to 101 tickets increases modeled revenue by $19.90. The derivative at 100 is $20, a close but not identical local approximation because one ticket is a finite step. Both calculations reinforce the same idea: the revenue added by the next unit declines as q grows under this inverse-demand assumption.
For the profit model P(q) = −0.10q² + 32q − 500, marginal profit is 32 − 0.20q. It becomes zero at q = 160, matching the profit vertex. Equivalently, marginal revenue is 40 − 0.20q and marginal cost is $8; setting them equal gives 40 − 0.20q = 8, or q = 160. This calculus shortcut is valid only after the underlying functions and feasible domain are defined. It is a later extension of the basic revenue-function idea, not a replacement for it.
How to check a model built from data
Before trusting a result, substitute each supplied observation into the demand equation. For q(p) = 220 − 10p, the model must return 120 at p = 10 and 100 at p = 12. If either substitution fails, revisit the slope and intercept before building R(p). Next check units: dollars per ticket times tickets should give dollars. Then check the domain: a price that makes q(p) negative is not a meaningful sales scenario. Finally, evaluate the candidate optimum in the original revenue formula, not just the vertex expression, so a sign error is less likely to survive.
For a real dataset, two observations provide only a very fragile linear estimate. One unusually busy weekend, a promotion, or a competitor’s discount can distort it. Demand may be seasonal or nonlinear. The formula also assumes a single uniform price, while many businesses sell at multiple prices. State these limitations when the task asks for interpretation. You need not abandon algebra; you simply distinguish what the model establishes from what it does not.
A useful sensitivity check changes one assumption and recalculates. If the venue capacity is 150 instead of 250, the maximum-revenue q = 200 is impossible. If variable cost rises from $8 to $10, the profit vertex shifts because the linear term in P(q) changes. If the demand slope changes, both the optimal price and quantity can move. This reasoning is often more educational than presenting one unqualified “best” number.
Common mistakes and quick repairs
The first mistake is confusing revenue with profit. Repair it by writing R, C and P as separate functions before substituting values. The second is treating price as constant after the word problem gives a demand relationship. Repair it by explicitly writing R(q) = q p(q) or R(p) = p q(p). The third is mixing units, such as multiplying a monthly price by an annual quantity without converting. Repair it by annotating every variable with its unit and time period.
Another error is using the vertex formula on a linear function. R(q) = 20q has no parabola and no interior vertex; a bounded domain may put its maximum at an endpoint. Students also forget that a parabola’s algebraic maximum can lie outside the practical domain. Always draw or state the domain first and compare endpoints if it is restricted. If a formula yields a negative quantity or price, reject that part of the graph as outside the model instead of forcing a business interpretation.
Be precise about slope. For a constant-price linear revenue function, slope equals price per unit. For variable-price revenue, slope is a change in total revenue associated with quantity, which can be positive, zero or negative. Similarly, the break-even point is where R = C, not where R = 0. Checking a proposed answer by substituting into both sides of the relevant equation is a fast way to catch a mislabeled point.
Finally, avoid copying an equation rendered multiple ways from a webpage or calculator. Write one clean expression, define its variables, and use it consistently. A duplicated formula with mismatched units makes an explanation harder to follow, not more rigorous. In a written answer, an ordinary line such as “R(q) = 40q − 0.10q² dollars, for 0 ≤ q ≤ 400 tickets” often communicates more than a page of unlabelled symbolic manipulation.
Practice problems with answers
Practice one: A club sells shirts for $18 each and expects to sell at most 90 during a fundraiser. Write the revenue function and find revenue at 40 shirts and the maximum within the given domain. The function is R(q) = 18q dollars for whole-number q from 0 through 90. At 40 shirts, revenue is $720. Because the line rises throughout the allowed range, its constrained maximum is $1,620 at q = 90. There is no interior quadratic vertex to calculate.
Practice two: A museum predicts q(p) = 300 − 15p visitors at a ticket price p dollars, within 0 ≤ p ≤ 20. Write revenue as a function of price and find the modeled maximum. Multiplying gives R(p) = 300p − 15p². The vertex is p = −300/[2(−15)] = $10. The model then predicts 150 visitors and $1,500 in revenue. Check the endpoints: at p = 0, there is no revenue; at p = 20, the predicted visitor count is zero. The vertex is within the stated domain.
Practice three: A seller’s revenue is R(q) = 25q and costs are C(q) = 240 + 9q. Find break-even quantity and interpret it. Set 25q = 240 + 9q, giving 16q = 240 and q = 15. At fifteen units, revenue and cost both equal $375, and profit is zero. If whole units are required, fifteen is exactly the first break-even quantity. Selling fewer yields a modeled loss; more yields a modeled profit, assuming the functions continue to apply.
Practice four: In the original event model, compare q = 160 and q = 200 when C(q) = 500 + 8q. At 160 tickets, price is $24, revenue $3,840, cost $1,780 and profit $2,060. At 200 tickets, price is $20, revenue $4,000, cost $2,100 and profit $1,900. The second quantity maximizes the revenue quadratic, while the first maximizes the profit quadratic. State which objective a question asks you to optimize before using either result.
For extra practice, invent a two-point price-and-quantity table, derive a linear q(p), and test that it reproduces both observations. Multiply to form R(p), restrict the domain to nonnegative price and quantity, and identify a vertex or boundary maximum. Then introduce a cost function and see whether profit peaks at the same point. This sequence combines the skills without relying on one memorized formula.
A reliable final checklist
Every complete revenue-function solution should answer five questions. What is the input variable and its unit? How does price relate to quantity in the stated model? What is the resulting revenue equation and what are its output units? Which input values are actually feasible? Finally, does the problem ask about revenue, profit, break-even or marginal change? Those questions tell you whether to multiply, solve an intersection, find a vertex, compare endpoints or take a derivative.
The broad lesson is that R = price × quantity is a starting identity, not always a final linear function. Once price depends on quantity, the product becomes a richer model with a graph, a practical domain and a possible maximum. Interpret every algebraic answer in the language of the problem, check it against the assumptions, and keep revenue separate from what remains after costs. That approach transfers from classroom ticket and notebook examples to more complex business and economic models without pretending a simplified equation is a guaranteed forecast.
