TL;DR: Price elasticity assumes a single posted price the market responds to. B2B software has no such price: every deal closes at a negotiated net price, so a curve fitted to your deal history measures your discount waterfall, not buyer demand. The fix is a constructed, margin-calibrated pricing surface, recalibrated on current transaction data as the market moves.
Price elasticity of demand measures how quantity changes when price changes. In B2B software, the concept misleads in a specific way: the curve you think you’re measuring is your own discount waterfall. Every won/lost deal in your CRM records a negotiated net price, not the single posted price the textbook assumes. Try to read elasticity off that data and what comes back is your discount authority, your quarter-end pressure, and which rep worked the deal. You have measured your sales motion, not your buyers’ price sensitivity.
The answer is to stop trying to read a demand curve and start constructing a pricing surface: a control surface you build, deploy, and tune, calibrated against margin rather than overfitted to historical deals. That gives you a foundation where net prices adjust continually to market feedback and price increases ship in step with the new value your roadmap delivers.
- What Price Elasticity of Demand Actually Measures in B2B Software
- In B2B Software, the Curve You Measure Is Your Sales Motion
- The Margin Math Hiding Inside Every Software Elasticity Estimate
- Measuring Elasticity with Surveys Makes the Error Worse
- Stop Reading the Demand Curve. Build a Pricing Surface.
- The Ceiling Moves Now, in Both Directions
- FAQs
What Price Elasticity of Demand Actually Measures in B2B Software
Price elasticity of demand is the ratio of the percentage change in quantity to the percentage change in price. A coefficient of -1 means a 10% price increase loses 10% of volume. Below -1 is elastic; above -1 is inelastic.
The definition is clean. The buried assumption is what creates the problem in B2B contexts.
Where Does the Elasticity Assumption Hold, and Where Does It Break?
The elasticity framework works when buyers face a single posted price and choose to buy or not. Consumer subscription markets are a reasonable fit: one price, one click, observable churn. B2B software is not that market.
In B2B software, nobody pays list. Every meaningful deal is negotiated. The “price” a buyer faces is a landed net price produced by a sequence of discounts, term adjustments, and channel commissions. There is no posted price for the market to respond to, so there is no stable demand curve to read.
The Posted-Price Assumption B2B Software Violates
In a B2B software deal, three distinct numbers exist where the textbook demand curve assumes one. List price is the published figure in the pricebook. Scheduled net price is the deterministic price the pricing surface produces for a specific commitment. Landed net price is what actually closes, after the discretionary concessions, channel fees, and implementation offsets a rep layers on to win the deal.
The seller builds the discount structure that opens the gap between list and landed net. Buyer resistance pushes a deal down within it, but so do rep authority and quarter-end timing, and the deal record cannot separate the buyer’s signal from the seller’s discretion. When you fit a demand curve to your deal history, you are fitting that blend, not clean demand.
In B2B Software, the Curve You Measure Is Your Sales Motion
Sort your last four quarters of closed deals by net price for the same SKU. The spread is not a clean read on willingness-to-pay. The gap between achievable and realized price is built from seller-initiated discount layers, so it traces to your own discounting and channel commissions more than to any buyer pushback. Buyers who paid more worked with a rep who held price, closed in Q1 with no quarter-end pressure, or came through a channel with less discount authority.
Why the Same Product Sells at a Dozen Different Net Prices
A standard discount waterfall runs several layers: a volume schedule, a term-length incentive, a pay-in-advance discount, a competitive concession, a quarter-end incentive, and a channel margin, plus the occasional one-time implementation offset to smooth a deal. Sellers initiate every one of them.
At the end of the waterfall, the landed net price looks like a clean buyer signal. Most of its spread is seller artifact, and the willingness-to-pay underneath is real but buried.
The Discount Waterfall Manufactures the Curve You Then “Measure”
When a pricing analyst fits a demand curve to that deal scatter, the curve describes the shape of the waterfall, not the shape of buyer demand. Refining the model with more data fits the waterfall more precisely. It comes no closer to real willingness-to-pay.
In the pricing engagements we run, we routinely see the same product sell across a wide net-price band in a single quarter to buyers with nearly identical profiles. The spread tracks which rep, which quarter-end, and which discount authority was invoked, not any difference in what the buyer would pay.
Without a scheduled net baseline to measure against, the real demand curve is not just distorted, it is unrecoverable: the price scatter blends deliberate segmentation with discretionary discounting, and no amount of refitting separates them. Most companies never see this, so they keep treating a structural blindness as a sales-discipline problem.
The Margin Math Hiding Inside Every Software Elasticity Estimate
Comparing relative price response across editions and customer groups has real portfolio value. The narrow mistake is different: fitting a single demand curve to negotiated B2B deals and treating it as buyer willingness-to-pay. On high-margin software, that mistake is expensive in a specific, arithmetic way.
What a High Gross Margin Does to the Price/Volume Tradeoff
At an 85% gross margin, the volume swing around a price change is smaller than intuition suggests. Work the arithmetic. Sell 100 units at $100 against a $15 unit cost and you earn $8,500 in gross margin. Cut the price 5% to $95 and each unit now earns $80, so you need about 106 units, a 6% volume gain, just to hold that same $8,500. Raise the price 5% to $105 and each unit earns $90, so you can sell about 94 units, a 6% volume loss, and still clear $8,500.
The break-even swing for a 5% price move is only about 6% of volume. High-gross-margin software with deep integration and real switching costs gives buyers few near-term alternatives, so demand behaves as near-inelastic across the discount range buyers negotiate within, moving far less than that 6%.
So the two directions are not equal in practice. A 5% price increase on near-inelastic software loses far less than the 6% of volume it can afford to lose, which makes it almost pure margin. A 5% discount rarely buys the 6% of extra volume it needs to pay for itself. Chasing volume down the demand curve quietly destroys software margin.
Are You Fitting an Elasticity Curve to the Wrong Customer Segment?
The margin math breaks when you apply aggregate elasticity estimates to segments with fundamentally different value thresholds. We can assess whether your licensing metric is mapped to the right curve for each portfolio tier.
Measuring Elasticity with Surveys Makes the Error Worse
The standard route when deal data looks noisy is to run a conjoint study or a Van Westendorp price-sensitivity survey. Both measure hypothetical price response. Neither measures what buyers actually do when they face a real net price in a real negotiation.
Can You Survey Your Way to a B2B Software Elasticity Estimate?
You can produce a number, but it will not match your realized transaction data. A survey captures a stated preference at zero cost; a negotiation captures what a buyer accepted when the budget was real. Deal-level transaction data reveals real willingness-to-pay more reliably, because it records the net prices buyers accepted and the deals that went dark.
The better analysis decomposes the net-price band in your existing deal data, separates the seller-side discount contribution from buyer behavior, and treats the residual as your actual price-sensitivity signal. For why survey-based measurement fails here specifically, see why willingness-to-pay surveys fail B2B software.
Stop Reading the Demand Curve. Build a Pricing Surface.
A demand curve is an observed trendline you try to read off the market. A pricing surface is a control surface you construct: a multi-input, margin-calibrated function that produces a deterministic scheduled net price for any commitment a buyer might make. You do not fit it to history; you build it, deploy it, and tune it, calibrated on the prices buyers actually pay.
This is the core distinction in the software monetization frame: the commercial architecture of a software business is engineered, not discovered.
A Pricing Surface Is a Control Surface, Not a Demand Curve
A pricing surface takes multiple inputs. The primary one is the deal’s bill of materials, the configured set of software and service SKUs that can range from a single line to a complex solution. With value metric volume, contract term, customer group, and channel layered on, the surface outputs a scheduled net price that is deterministic and buyer-verifiable, because it does not vary by rep or by quarter-end desperation. You cannot tune a trendline; you can tune a surface.
The two are conflated constantly. In one engagement, a company fit a regression line to its deal data, used it to draw manual pricing tiers, then set each tier’s price to the average of the historical deals that landed inside it. What that produces is the discount waterfall bucketed and averaged, then relabeled a pricing surface. The historical distortion becomes the official price, and nothing in the process is calibrated to margin.
Margin-Calibrated Discounting Replaces the Waterfall That Made Your Curve
Once you have a pricing surface, the discount waterfall is replaced by scheduled net prices at every commitment level. Sales compensation anchors to the surface, so reps are not rewarded for deeper discounts. Discretionary concessions shrink because the surface already handles volume, term, and customer-group variation.
Margin-Calibrated Discounting is the practice that operates the surface. Three discounting approaches that slow SaaS growth describes what the waterfall looks like before the surface exists. The surface is calibrated on real transaction data, which is what LevelSetter records across sales-led and self-serve motions.
The Ceiling Moves Now, in Both Directions
A measured elasticity is stale on arrival, because the willingness-to-pay ceiling for B2B software now moves in both directions. New product capabilities push it up. Competitive commoditization pushes it down. An AI feature wave can shift the ceiling faster than any annual pricing review cycle.
Why a Measured Elasticity Expires the Moment You Take It
A fitted curve describes the market that produced the deals you fit it to. Using recent history is not the mistake; leaving the measurement in place is. Measure elasticity once and it decays immediately, because product versions, competitive conditions, and buyer budgets keep moving while the number sits still. Your costs move too: as generative-AI workflows grow more capable and more expensive to run, the margin the surface is calibrated against shifts under you. The failure is not the window you used, it is the interval before you look again, and that interval must shrink as the rate of change accelerates.
Price to last year’s elasticity and you leave money behind when the ceiling has risen, or accelerate churn when it has fallen.
Revising the Architecture on Your Product’s Cadence
The discipline I’ve called Continuous Monetization addresses this: revising the pricing architecture on the cadence your product changes, not on a fixed annual calendar. LevelSetter records real transaction behavior so each revision is calibrated on current data.
The elasticity number is not the input to a pricing decision. The pricing surface is the output of a monetization discipline. Each time the product changes in a way that shifts the value a buyer receives, the surface is revised to match. For a full treatment, see Continuous Monetization.
Stop reading demand curves. Build a pricing surface you can tune.
Your pricing surface should be calibrated on what buyers actually pay, not on a curve fitted to last year’s deals. LevelSetter records net prices across sales-led and self-serve motions, so each revision uses current transaction data. See how we approach it.
Explore the pricing surface approach · Book a working session
