Unit Demand Analysis- Essential Techniques for Market Research
What Unit Demand Analysis Actually Is
Unit demand analysis measures how much of a single product customers will buy at different price points. That's it. No fluff, no complicated theory.
You want to know: if I price this at $50, how many units sell? What about $45? $55? This data tells you where your revenue peaks and where you're leaving money on the table.
Most businesses guess at pricing. Unit demand analysis gives you actual numbers to work with.
Why This Matters for Your Market Research
Without understanding demand curves, you're flying blind. You either underprice and hemorrhage potential revenue, or overprice and watch customers vanish.
Unit demand analysis answers three questions you can't afford to ignore:
- What's the price that maximizes total revenue?
- At what price do you stop selling meaningful volume?
- How sensitive is your market to price changes?
Skip this step and you're making million-dollar decisions based on gut feelings. That's not research—that's gambling.
The Techniques That Actually Work
Gabor-Granger Method
This one's straightforward. You ask respondents if they'd buy a product at a specific price. Then you change the price and ask again. Repeat until you find the cutoff.
Best for: Products with little competition, new launches where you have no historical data.
The downside is it doesn't account for alternatives. Respondents might say yes to $100 when it's the only option, but grab a competitor's product at that price in real life.
Van Westendorp Price Sensitivity Meter
You ask four questions:
- At what price is this too expensive?
- At what price is this a bargain?
- At what price is too expensive but you'd consider it?
- At what price is too cheap (questionable quality)?
Plot the answers and you get an "acceptable price range" with an "indifference price point" where buyers stop caring about cost.
Best for: Consumer products where perceived value matters. Works well when customers have reference points for pricing.
Conjoint Analysis
This is the heavyweight. You present products with multiple attributes—price, brand, features, warranty—and see which combinations customers prefer.
It simulates real buying decisions better than single-question surveys. People don't evaluate price in isolation; they weigh it against what they get.
Best for: Complex products with multiple variables. Tech gadgets, insurance plans, subscription services.
The catch: it requires larger sample sizes and more sophisticated analysis. Cheap to do wrong, expensive to do right.
Brand-Specific Price Elasticity Models
Use historical sales data combined with price changes to calculate elasticity. If you dropped price 10% last year and sales jumped 15%, your elasticity is -1.5.
Best for: established products with enough sales history to analyze.
This method ignores market changes—competitor moves, seasonal shifts, economic conditions—so treat it as one input, not the whole picture.
Technique Comparison
| Method | Setup Time | Cost | Best For | Accuracy |
|---|---|---|---|---|
| Gabor-Granger | Low | $ | New products, simple pricing decisions | Moderate |
| Van Westendorp | Low | $ | Consumer goods, value positioning | Moderate |
| Conjoint Analysis | High | $$$ | Complex products, feature trade-offs | High |
| Elasticity Models | Medium | $ | Established products, historical context | Varies |
Common Mistakes That Ruin Your Data
Asking the wrong people. Your current customers will give different answers than potential customers. A loyal buyer might tolerate higher prices than someone who's never tried your product.
Ignoring competitive context. Survey respondents exist in a market. If you don't show them alternatives, their answers won't match real behavior.
Testing too narrow a range. If you only test $48-$52, you miss the $40 point where volume explodes or the $60 point where you lose everyone.
Treating estimates as facts. These techniques give you informed ranges, not precise predictions. Build in margins for error.
Getting Started: Your First Unit Demand Analysis
Step 1: Define your objective. Are you setting initial price, testing a change, or evaluating a new product? The goal shapes everything else.
Step 2: Pick your method. New product with no history? Start with Gabor-Granger or Van Westendorp. Established product with data? Build an elasticity model and validate with survey research.
Step 3: Design the survey. Test at least 5-7 price points spanning a realistic range. Include attention checks. Keep it under 10 minutes or completion rates tank.
Step 4: Collect responses. Aim for 300+ respondents minimum. More is better but diminishing returns kick in around 500 for most use cases.
Step 5: Analyze and interpret. Calculate revenue at each price point (price Ă— expected volume). Find where revenue peaks. That's your starting point for pricing strategy.
When to Combine Methods
No single technique gives you the full picture. Smart market research stacks approaches:
- Run Van Westendorp to establish a reasonable price corridor
- Use conjoint to test feature combinations within that range
- Validate with historical elasticity data if available
This triangulation catches weaknesses that single-method approaches miss.
The Bottom Line
Unit demand analysis isn't optional if you're serious about pricing. Pick a method that matches your product complexity and budget, run the study, and use the data. Your gut is not a substitute for actual customer feedback on what they'll pay.
Start simple. Gabor-Granger costs almost nothing and gives you workable numbers. Upgrade to conjoint when you have the resources and the product complexity demands it.