Measuring Willingness
To Pay
The concept of price elasticity, how demand changes in response to price changes, appears deceptively simple in economic textbooks. A single coefficient that tells you whether to raise or lower prices to maximise revenue. In the messy reality of SaaS businesses, this simplicity evaporates immediately. Price elasticity is not a single number you can look up or calculate once. It’s a multidimensional landscape that varies dramatically across customer segments, usage patterns, purchase contexts, and time horizons. Understanding this landscape is the difference between pricing that captures value and pricing that leaves enormous sums on the table or, worse, actively repels the customers you most want to serve.
The challenge begins with recognising that your customer base is not a homogeneous group with uniform price sensitivity. Every SaaS company serves multiple distinct segments; sometimes knowingly, often accidentally, and each segment exhibits fundamentally different willingness to pay.
Elasticity Is Not A Single Number
The first and most consequential error in elasticity analysis is the assumption that price sensitivity can be captured in a single metric. In practice, elasticity varies along multiple dimensions simultaneously, and understanding these variations is essential to making sensible pricing decisions.
Segment-level variation is the most obvious dimension but also the most frequently mishandled. Small and medium-sized businesses exhibit radically different price sensitivity than enterprise customers, but not always in the direction you’ would expect. The conventional wisdom suggests that larger customers are less price-sensitive because they have bigger budgets and focus more on value than cost. This is sometimes true, but the reality is considerably more nuanced.
SMBs often have extremely high price sensitivity at low absolute price points but surprisingly low sensitivity once a price crosses a certain threshold. A £20 per month product competes with doing things manually or using free alternatives. A £200 per month product is being evaluated against other paid solutions, and the incremental difference between £200 and £300 may matter far less than the difference between £20 and £40. This creates non-linear elasticity curves where small price increases at low price points have dramatic effects on conversion, whilst larger increases at higher price points have modest impacts.
Enterprise customers exhibit different patterns entirely. Their price sensitivity often has less to do with the absolute price and more to do with procurement processes, budget cycles, and internal politics. A £50,000 annual contract might sail through approval whilst a £60,000 contract gets stuck because it crosses a threshold requiring additional sign-offs. The elasticity in enterprise segments is often about navigating organisational dynamics rather than fundamental value perception.
Power users versus casual users represent another critical dimension of elasticity variation. Power users, those who have deeply integrated your product into their workflows, who use advanced features, who have built dependencies, typically exhibit very low price elasticity. They’re willing to pay substantially more because switching costs are high and the value they receive is tangible and measurable. Attempts to extract this willingness to pay through aggressive pricing can work, but they require careful calibration to avoid crossing the line into perceived exploitation.
Casual users exhibit high price sensitivity precisely because they have not built these dependencies. Your product is useful but not essential, convenient but not irreplaceable. For these customers, even modest price increases can trigger re-evaluation of whether they need the product at all. The challenge is that casual users today might be power users tomorrow, so pricing them out entirely can eliminate your expansion opportunity.
The danger of averaging elasticity across these segments manifests in concrete ways. If you calculate that your average customer has moderate price sensitivity and decide to increase prices by 20%, you might lose 60% of your SMB customers whilst barely affecting enterprise retention. Your aggregate numbers might look acceptable, revenue held steady, customer count dropped but not catastrophically, but you have fundamentally altered your customer mix in ways that affect everything from support costs to expansion potential to product roadmap priorities.
Alternatively, you might keep prices low to accommodate high-elasticity segments and discover that you’re leaving vast amounts of money on the table with customers who would happily pay more. You have optimised for volume over value, which works for some business models but proves catastrophic for others. The low-elasticity enterprise customer paying the same price as the high-elasticity startup is not a victory, it is a failure to capture available value.
The sophisticated approach to segment-level elasticity is to measure it separately for each meaningful customer cohort and then make deliberate decisions about which segments to optimise for. This might mean accepting that you’ll have lower conversion in price-sensitive segments in order to maximise revenue from price-insensitive ones. Or it might mean creating differentiated pricing structures that allow you to serve multiple segments without leaving money on the table or creating excessive friction.
Measuring Elasticity Using Behavioural Data
Traditional elasticity measurement relies on controlled price experiments, change the price for a subset of customers, measure the impact on demand, calculate elasticity. This approach works but comes with significant limitations. Experiments are slow, require substantial sample sizes to detect meaningful differences, carry real revenue risk if you guess wrong, and measure only the narrow scenario you have tested rather than the broader elasticity landscape.
The alternative is to extract elasticity signals from the behavioural data your business generates continuously. Every customer interaction with your pricing contains information about willingness to pay, even when no purchase occurs. The art is learning to read these signals correctly and combine them into a coherent picture of price sensitivity across your customer base.
Upgrade patterns provide rich elasticity signals because they reveal customers actively choosing to pay more for additional value. The timing of upgrades; how long customers stay on lower tiers before upgrading, tells you about the strength of the value proposition at each tier. Customers who upgrade quickly have experienced sufficient value to overcome inertia and justify higher spend. Customers who delay upgrading despite hitting usage limits are revealing either price sensitivity or insufficient value differentiation between tiers.
The distribution of customers across your pricing tiers itself represents an elasticity signal. Heavy concentration at the lowest tier suggests either that your entry-level pricing is well-calibrated or that higher tiers are not creating sufficient perceived differentiation to justify their prices. Heavy concentration at mid-tier plans with low adoption of both entry-level and premium tiers suggests your packaging has created a natural sweet spot that captures most of your market.
Equally informative are the transitions customers do not make. Customers who repeatedly hit limits on lower-tier plans but never upgrade are revealing price resistance at the next tier. The question is whether this resistance comes from the absolute price being too high, the incremental value being too low, or the transition itself creating friction. Understanding which of these mechanisms is operating requires looking at the complete context, what features they are using, which limits they are hitting, how they respond when you engage them about upgrading.
Discount redemption behaviour provides perhaps the clearest window into price elasticity because discounts temporarily shift the price point and allow you to observe how demand responds. Customers who only purchase during promotional periods have revealed genuine price sensitivity; the discount is not just accelerating a decision that would have happened anyway but enabling a decision that would not have happened at full price.
The pattern of discount response tells you about the shape of the elasticity curve. If a 10% discount produces minimal lift but a 30% discount produces substantial conversion, you have identified a threshold effect, a price point at which the product transitions from “too expensive” to “acceptable” for a particular segment. If conversion increases linearly with discount depth, you are seeing continuous elasticity across the price range. If even deep discounts produce minimal lift, you are dealing with fundamental product-market fit issues rather than price sensitivity.
What happens after the discount expires is equally telling. Customers acquired at a discount who maintain their subscriptions at full price were likely going to convert eventually, the discount just accelerated timing. Customers who churn immediately when promotional pricing ends or who downgrade to free tiers have revealed that your full price exceeds their willingness to pay. The proportion of each type tells you how much genuine price resistance exists in your market versus how much you are simply shifting purchase timing around.
Abandonment at checkout represents a direct signal of price resistance, though interpreting it requires care. Some checkout abandonment reflects price shock; the customer did not fully internalise the price until the moment of commitment and balked. Some reflects comparison shopping; the customer is checking multiple options and will return later. Some reflects technical issues, concerns about terms and conditions, or simple distraction.
The most informative abandonment signal comes from customers who repeatedly visit the pricing page or initiate checkout but never complete purchase. This pattern reveals active consideration combined with unresolved hesitation. When you see this behaviour at scale, particularly concentrated at specific price points or tiers, you have identified friction that demands investigation. Are customers confused about what they are getting? Concerned about value? Comparing against alternatives? Lacking budget authority? Each underlying cause has different implications for how you should respond.
“Intent” signals (behaviours that suggest interest without completing purchase) provide leading indicators of price sensitivity before it manifests in conversion metrics. Repeated visits to the pricing page indicate active price comparison or hesitation about value. Requests for trial extensions suggest the customer has not experienced sufficient value to justify paying but believes more time might change that. Customer research calls, particularly those focusing on pricing questions, reveal active evaluation and often surface the specific concerns preventing purchase.
The sophistication in behavioural elasticity measurement comes from combining these signals into predictive models rather than treating each as an isolated data point. A customer who visits the pricing page repeatedly, initiates but abandons checkout, and then requests a trial extension is telling a clear story about their journey toward or away from purchase. When you see this pattern across dozens or hundreds of customers, you can identify the common characteristics; company size, usage patterns, feature adoption, that predict price sensitivity.
Expert help is only a call away. We are always happy to give advice, offer an impartial opinion and put you on the right track. Book a call with a member of our friendly team today.
Analytical Techniques
Beyond reading behavioural signals, several established analytical frameworks help structure elasticity measurement and translate observations into actionable insights. Each technique has strengths and limitations, and the most sophisticated pricing operations use multiple approaches in combination rather than relying on any single method.
Van Westendorp price sensitivity modelling, also known as the Price Sensitivity Meter, identifies acceptable price ranges through four key questions: At what price would the product be so expensive that you would not consider buying it? At what price would you consider it expensive but still worth considering? At what price would you consider it a bargain? At what price would it be so cheap that you’d question its quality?
The power of Van Westendorp is that it identifies not just an optimal price point but the acceptable range around that point and the boundaries where price creates negative signals. The technique reveals both the ceiling (where price kills demand) and the floor (where price undermines credibility). For SaaS businesses, this helps avoid the trap of pricing so aggressively low that you attract customers who will never pay meaningful amounts and repel customers who associate higher prices with better quality and support.
The limitation of Van Westendorp is that it relies on stated preferences rather than revealed preferences; what customers say they’d pay rather than what they actually pay. Stated preferences systematically underestimate willingness to pay because customers have incentives to claim lower thresholds than their true limits. The technique works best when used to establish rough boundaries and identify relative differences between segments rather than to set precise price points.
Conjoint analysis addresses the stated preference problem by forcing trade-offs between features and prices, revealing which attributes customers actually value most. Rather than asking “How much would you pay for Feature X?”, conjoint analysis presents realistic product configurations with different feature sets and prices, asking customers to choose between them. The resulting data reveals not just willingness to pay but the relative value customers place on different capabilities.
For SaaS businesses, conjoint analysis is particularly valuable for packaging decisions; which features to gate at which tiers. You might discover that customers place high value on Feature A but low value on Feature B, even though your internal stakeholders consider B more important. This information transforms packaging discussions from opinion-driven debates to evidence-based decisions about which features justify tier differentiation.
The challenge with conjoint analysis is methodological complexity and the risk of overwhelming respondents with too many attributes or unrealistic combinations. The technique requires careful design to avoid creating artificial scenarios that don’t reflect actual purchase decisions. When done well, conjoint provides insights that are difficult to obtain through other means. When done poorly, it produces precise-looking numbers that bear little relationship to actual customer behaviour.
Behavioural elasticity curves constructed from historical experiments represent the gold standard when you have sufficient data. Rather than asking customers hypothetical questions, you observe actual purchase behaviour at different price points and map the relationship between price and conversion. This might come from A/B tests where different customers saw different prices, from natural experiments where you changed prices over time, or from geographic variation where different markets have different price levels.
The advantage of behavioural curves is that they reflect actual decisions rather than hypothetical ones, capturing all the contextual factors and irrationalities that influence real purchase behaviour. The disadvantage is that they require substantial data to construct reliable curves, they measure past behaviour in specific contexts rather than future behaviour in different contexts, and they cannot tell you about price points you have not tested.
The most sophisticated approach combines historical behavioural data with forward-looking models. You use past experiments to establish baseline elasticity for segments you understand well, then use conjoint or Van Westendorp data to estimate elasticity for segments or price points where you lack historical data. You validate these estimates with small-scale experiments before making large pricing changes.
Predictive models using usage patterns to forecast lifetime value represent a different approach to elasticity, rather than measuring price sensitivity directly, you identify customer characteristics that predict high or low willingness to pay over time. Customers who rapidly adopt advanced features typically exhibit lower price elasticity than customers who use only basic capabilities. Customers who integrate your product into automated workflows show lower churn sensitivity to price increases than customers who use it sporadically.
These predictive models allow you to segment customers by expected elasticity even before you have directly tested their price sensitivity. A new customer exhibiting high-value usage patterns can be routed toward premium tiers or targeted with expansion campaigns, whilst customers exhibiting low-engagement patterns might require different nurturing before they are ready for price increases.
The analytical sophistication required for effective elasticity measurement should not be understated. It is not enough to run a single Van Westendorp study or A/B test and declare the elasticity question settled. Elasticity varies by segment, context, and time. Building genuine understanding requires multiple analytical approaches, continuous measurement, and humility about the limitations of each technique.
Elasticity In The Real World
The theoretical relationship between price and demand; lower prices increase demand, higher prices decrease it, breaks down quickly when you examine real SaaS purchase behaviour. The actual elasticity landscape contains anomalies, discontinuities, and counterintuitive patterns that challenge conventional pricing wisdom.
The most striking anomaly is that cheaper pricing does not always produce higher elasticity. Conventional logic suggests that price reductions should reliably increase demand, but this assumes that price is the primary barrier to purchase. In reality, many SaaS purchase decisions are constrained by factors orthogonal to price: limited attention, evaluation overhead, switching costs, risk aversion, and simple inertia.
Reducing your price from £100 to £80 per month might increase conversion by 5% or it might have no measurable effect at all. The customers who weren’t buying at £100 often weren’t making a marginal price-value calculation. They were uncertain about whether they needed the product, concerned about implementation effort, waiting for budget cycles, or simply hadn’t gotten around to making a decision. For these customers, a 20% price reduction does not address their actual barrier to purchase.
More perversely, lower prices can sometimes reduce demand by sending negative quality signals. In categories where customers associate price with quality, support, and reliability, aggressive pricing can trigger concerns about whether the product is adequately resourced, whether the company will survive, whether support will be responsive. Enterprise buyers in particular often view suspiciously low pricing as a red flag rather than an opportunity.
The “prestige pricing” phenomenon; where premium tiers convert better at higher prices than at lower prices, appears regularly in SaaS businesses with complex products or sophisticated target markets. When you price a premium tier at £200 rather than £500, you might see lower conversion rather than higher. The higher price serves as a signal of capability, seriousness, and target market. Customers evaluating premium tiers are often not primarily price-sensitive; they’re evaluating whether the product can handle their complex requirements. Low pricing suggests it cannot.
This creates the counterintuitive situation where raising prices on premium tiers actually improves conversion by attracting customers who view high prices as a proxy for high capability. This does not work for every product or every tier, trying to apply prestige pricing to an entry-level plan targeting startups will fail spectacularly. But for products targeting enterprises, for premium tiers, for categories where complexity is high and direct feature comparison is difficult, higher prices can paradoxically reduce price resistance by signalling that you serve sophisticated buyers with sophisticated needs.
Elasticity cliffs represent another real-world complication that simple models do not capture. Rather than a smooth curve where conversion gradually declines as price increases, you often see relatively stable conversion across a price range followed by sudden collapse when you cross a threshold. Trial-to-paid conversion might be 15% at £49, 14% at £69, 13% at £89, and then suddenly 7% at £99.
These cliffs typically correspond to psychological price thresholds or organisational authority limits. Breaking through £100 per month requires a different level of justification than staying under it. Moving from “a few hundred” to “a few thousand” crosses procurement thresholds in many organisations. The cliff is not about the incremental value between £89 and £99, it is about crossing a boundary that changes the purchase context entirely.
The practical implication is that you cannot just optimise price continuously by testing small increments. You need to identify where these cliffs exist, through data analysis, customer research, and systematic testing, and make conscious decisions about whether to price just below them (maximising conversion at some revenue cost) or well above them (accepting lower conversion to serve customers who clear the threshold easily).
Geographic elasticity variation adds yet another layer of complexity. The same product priced identically in different markets produces wildly different conversion rates, not just because of income differences but because of competitive contexts, cultural attitudes toward SaaS purchasing, and payment infrastructure. A price point that seems reasonable in San Francisco might be prohibitively expensive in Bangalore, not because the absolute value is different but because the competitive alternatives and reference prices are different.
The sophisticated response is not necessarily to maintain global pricing but to recognise that effective price in different markets includes more than just the number you display. It includes payment method friction, currency conversion concerns, local tax implications, and competitive positioning. A seemingly expensive product that accepts local payment methods and provides transparent local pricing might convert better than a cheaper competitor that forces international transactions and currency uncertainty.
When Elasticity Conflicts With Strategy
The most difficult pricing decisions arise when elasticity analysis points in a direction that conflicts with strategic objectives. You can measure elasticity with perfect accuracy and still make the wrong pricing decision if you optimise purely for short-term conversion without considering longer-term strategic implications.
High-elasticity segments that are not worth serving represent the most common conflict. You might discover through careful analysis that students, hobbyists, or very small businesses are extremely price-sensitive, if you drop your price to £5 per month, conversion would skyrocket. But these customers might also generate disproportionate support costs, have limited expansion potential, and create noise in your product feedback that pulls you away from enterprise priorities.
The temptation is to capture this demand because revenue is revenue and customers are customers. The reality is that not all revenue is equally valuable and not all customers are equally strategic. A thousand customers paying £5 per month generates £60,000 annually. Fifty customers paying £200 per month generates £120,000 with lower support costs, clearer expansion paths, and better strategic alignment if your goal is to move upmarket.
The elasticity might tell you that the high-volume, low-price segment is available to you. Your strategy needs to tell you whether pursuing it makes sense given your resources, competitive positioning, and long-term objectives. Sometimes the right answer is to deliberately price yourself out of high-elasticity segments to focus on segments that align with your strategic direction, even if it means leaving apparent revenue on the table.
Low-elasticity enterprise buyers justify a different kind of strategic complexity. When you identify customer segments that are relatively insensitive to price, they would pay the same amount whether you charged £50,000 or £75,000 annually, the temptation is to charge as much as possible. But simplistic exploitation of low elasticity can backfire in several ways.
First, even low-elasticity enterprise buyers compare value to price. They might not be sensitive to incremental price differences within a range, but if you price egregiously above value delivered, you create resentment and vulnerability to competitive displacement. The customer who tolerates paying £75,000 when your product is worth £80,000 to them will switch immediately when a competitor offers 90% of your value for £40,000.
Second, enterprise pricing often involves bespoke negotiation, custom contracts, and relationship-driven terms that cannot be captured in standard pricing logic. The measured elasticity might suggest a single optimal price, but the reality is that different enterprise customers have different budget sources, different procurement constraints, and different internal politics. Trying to enforce uniform pricing on low-elasticity enterprise segments leaves money on the table with customers who could pay more and creates unnecessary friction with customers who face genuine constraints.
The sophisticated approach to low-elasticity enterprise segments is typically to establish a starting price that reflects genuine value but to maintain flexibility for negotiation within boundaries. You’re not trying to extract every last pound of willingness to pay, you are trying to find sustainable pricing that captures fair value whilst building long-term relationships that enable expansion.
Freemium users who will never pay regardless of price represent perhaps the most challenging elasticity scenario. You can measure infinite elasticity, these customers exhibit no willingness to pay at any price point, but that does not necessarily mean they are worthless. Some provide valuable feedback, some drive network effects, some might convert years in the future, some might refer paying customers.
The question is not whether freemium users exhibit price elasticity (they do not) but whether they provide sufficient indirect value to justify their costs. This calculation depends entirely on your business model and growth strategy. If you are building a network product where free users create value for paid users, keeping them makes sense. If you are building an enterprise product where free users dilute your brand and distract your team, converting them is impossible and irrelevant, you should focus elsewhere.
The strategic error is confusing elasticity measurement with strategic decision-making. Elasticity tells you how customers respond to price changes. Strategy tells you which customers to serve, which objectives to optimise for, and which trade-offs to make. Sometimes these align neatly, high-elasticity segments that fit your strategy, low-elasticity segments that do not. Often they conflict, and resolving these conflicts requires judgment that goes beyond the numbers.
Conclusion
Price elasticity is not a single coefficient to be calculated and applied uniformly. It is a multidimensional landscape that varies across customer segments, usage patterns, competitive contexts, and time horizons. The customers you serve exhibit radically different sensitivities to price, driven by factors ranging from budget constraints to perceived value to organisational procurement processes to psychological thresholds.
Understanding this landscape requires moving beyond simple elasticity calculations toward rich behavioural analysis. Every interaction customers have with your pricing generates signals about willingness to pay; upgrade patterns, discount behaviour, checkout abandonment, intent signals. Reading these signals correctly, combining them with structured analytical techniques like Van Westendorp and conjoint analysis, and validating them through carefully designed experiments produces a nuanced understanding of how different segments respond to different price points.
But measurement alone does not produce good pricing decisions. The real world contains anomalies that challenge simple elasticity logic, prestige pricing effects where higher prices increase demand, elasticity cliffs where small price changes cause sudden conversion collapse, segments with high price sensitivity that are not worth serving. Navigating these complexities requires combining elasticity analysis with strategic judgment about which customers to serve, which objectives to optimise for, and which trade-offs to make.
The goal of elasticity analysis is not to find a single optimal price that maximises short-term revenue. It’s to understand how different customers perceive and respond to value at different price points, so you can structure your pricing to capture value responsibly across your target segments. This might mean accepting lower conversion in price-sensitive segments to capture more value from price-insensitive ones. It might mean creating differentiated pricing structures that serve multiple segments simultaneously. It might mean deliberately pricing yourself out of segments that don’t align with strategic priorities.
What elasticity analysis tells you, ultimately, is what customers value and what they are willing to pay for it. Your job is to structure pricing that captures this value in ways that are sustainable, defensible, and aligned with your long-term strategic objectives. This requires both analytical rigour in measuring elasticity and strategic sophistication in applying what you learn. The companies that get this right don’t just optimise pricing, they build pricing systems that grow more intelligent over time, learning continuously from customer behaviour and market feedback.
Get In Touch
Our friendly team are always on hand to answer questions, troubleshoot problems and point you in the right direction.
