Price elasticity is the most important unknown in any SaaS pricing decision. You can't raise prices confidently without knowing how customers will respond. Here are the practical methods for estimating elasticity without a large academic study.
Method 1: A/B test your pricing page (most reliable)
Show 50% of new visitors price A and 50% price B. Measure trial conversion rates. Run for at least 2 weeks and 500+ visitors per variant.
Elasticity estimate = (CR_B/CR_A − 1) ÷ (Price_B/Price_A − 1)
Example: price A = $99 (CR = 4%), price B = $129 (CR = 3.1%) = (3.1/4.0 − 1) ÷ (129/99 − 1) = (−0.225) ÷ (0.303) = −0.74
This elasticity of −0.74 means a 10% price increase reduces demand by ~7.4% — inelastic demand, price increase is revenue-positive.
Prerequisite: at least 1,000 visitors/month to get statistically reliable results. Use a sample size calculator to determine test duration.
Method 2: Van Westendorp Price Sensitivity Meter
Survey 100+ customers with four questions: 1. At what price is this product too cheap (poor quality)? 2. At what price is it a bargain (great value)? 3. At what price is it expensive but still worth it? 4. At what price is it too expensive (you'd stop buying)?
Plot the cumulative distributions. The intersection of "too cheap" and "too expensive" defines your acceptable price range. The intersection of "expensive" and "bargain" is the price point of marginal cheapness (PMC) — your optimal price.
Method 3: Analyze churn after a price increase
If you've raised prices before, look at the churn data: - How many customers cancelled within 60 days of the increase? - What % of your customer base churned specifically citing price?
Elasticity ≈ (Churn rate increase %) / (Price increase %)
This is a retrospective method — useful for calibrating future decisions.
Method 4: Competitor price comparison
If competitors are priced above you and growing, that's evidence your market is relatively inelastic. If you're priced above competitors and seeing churn to lower-priced alternatives, that's evidence of elastic demand.
What elasticity to assume when you have no data
If you have no data and can't run tests: - B2B SaaS with significant switching costs: assume −0.7 to −0.9 - B2B SaaS without lock-in: assume −1.0 to −1.3 - B2C subscription (consumer): assume −1.5 to −2.0 - E-commerce / physical goods: assume −1.5 to −3.0
Use the Pricing Elasticity Calculator to model revenue impact at different elasticity assumptions before running a live test.