Pricing a digital product with Van Westendorp (with real numbers)

I have priced products by gut feel and I have priced them by Van Westendorp. The Van Westendorp results were closer to where the market actually was, by a margin large enough to make the survey worth running every time.

This is the workflow I actually use, the math behind it, and the real numbers from one of my products at toolgenx.com.

What Van Westendorp is

Peter van Westendorp published this in 1976. It is a four-question survey about a single product or service:

1. At what price would you consider this product so expensive that you would not consider buying it? (Too expensive)
2. At what price would you consider this product to be priced so low that you would feel the quality could not be very good? (Too cheap)
3. At what price would you consider this product starting to get expensive, so that it is not out of the question, but you would have to give some thought to buying it? (Expensive)
4. At what price would you consider this product to be a bargain — a great buy for the money? (Cheap)

You collect responses from real prospective buyers. You plot four cumulative distribution curves. The intersections of the curves give you four meaningful numbers.

The four numbers the intersections produce

• Point of Marginal Cheapness (PMC): where "too cheap" meets "expensive". Below this, more people doubt the quality than think it is expensive. This is your floor.
• Point of Marginal Expensiveness (PME): where "cheap" meets "too expensive". Above this, more people refuse than see value. This is your ceiling.
• Indifference Price Point (IPP): where "expensive" meets "cheap". Equal numbers of people see it as expensive vs as a bargain. This is the closest thing to "the market's median expected price".
• Optimal Price Point (OPP): where "too cheap" meets "too expensive". This is the price that maximizes the number of acceptable buyers — neither too few who think it is cheap-tier nor too few who think it is overpriced.

For most digital products in the $10-$100 range, OPP and IPP sit within $5 of each other. PMC is usually 30-50% below OPP. PME is usually 50-100% above OPP.

The real survey from one of my products

In March 2026 I ran this survey on a planned content-craft product, eventually priced at $29. Sample: 87 respondents collected over 11 days from a mix of Twitter DM responses, mailing list opt-ins, and a small banner on the legacy Gumroad shop.

Raw data summary (rounded to nearest dollar):

Question	Median	25th percentile	75th percentile
Too expensive	$49	$39	$69
Expensive	$35	$29	$45
Cheap	$19	$15	$25
Too cheap	$9	$5	$15

Plotted as cumulative curves, the four intersections gave:

• PMC (floor): $14
• PME (ceiling): $42
• IPP (indifference): $29
• OPP (optimal): $25

The acceptable range is therefore $14 to $42, with the indifference point at $29 and the optimal at $25.

I shipped at $29 (the indifference price) rather than $25 (the optimal) for one reason: indifference is where the price perception is neutral — half think it is fair, half think it is on the expensive side. Optimal is where the number of acceptable buyers is highest. For a small-volume digital product I prefer indifference, because the marginal sale at $29 vs $25 is worth more than the slightly higher acceptance rate.

That was a judgment call, not a math call. The survey gave me the range. The decision within the range was mine.

What the survey saved me from

Pre-survey, my gut said this product was a $39 product. I had drafted the landing page at $39. The price felt right because I had spent so many hours building it.

If I had shipped at $39:

• The PMC and PME bands were $14 and $42, so $39 was still inside the acceptable range — but only barely.
• The indifference point was $29, which meant the market median expected price was ten dollars less than what I was about to charge.
• Realistically, my conversion rate would have been 25-40% lower at $39 vs $29, based on the slope of the "expensive" curve between those two prices.

The cost of running the survey was about 6 hours of work. The cost of shipping at $39 would have been every sale I did not get in the first six months at that price. The math is not close.

How to run the survey (the practical version)

Five steps. Total time: about 6 hours over 10-14 days.

1. Write the four questions clean

The exact wording matters. Use Van Westendorp's original phrasing — paraphrasing changes the answers. My questions for the $29 product were copy-pasted from the 1976 paper with one word swap: "this product" became the actual product name and one-line description.

2. Decide your sample source

The three sources that have worked for me, in order of result quality:

• Warm list: Twitter followers, mailing list, or existing customers who would plausibly buy a related product. Highest signal, lowest scale.
• Free-tier audience: people who downloaded a free resource from the same landing page. Decent signal, moderate scale.
• Cold paid audience: small ad budget ($20-50) targeted at the keyword cluster the product serves. Lowest per-respondent signal but easiest scale.

For most solo founder products, the warm list of 50 respondents beats a cold list of 200.

3. Run the survey

A single Google Form with the four questions takes 5 minutes to build. Each respondent finishes in 90 seconds. Include one optional open text field at the end for context — you get qualitative signal that informs how to position the price.

I send the survey link via a single sentence: "Quick four-question survey about pricing for [product]. Takes 90 seconds. I will share the results."

4. Plot the curves

Four cumulative distribution functions. For 50-100 respondents this is a 15-minute task in Google Sheets or Excel. Plot all four on the same axes. The intersections are the four numbers above.

There is a free Van Westendorp template in AI CFO Toolkit that does this automatically — paste in the raw responses, the four numbers fall out. For a one-off survey, doing it by hand is fine.

5. Read the result against your positioning

The survey gives you a range, not a price. The price within the range depends on:

• Volume vs margin: lower price (closer to OPP) means more units; higher price (closer to PME) means more per unit. Pick the side that matches your distribution strategy.
• Brand position: premium positioning works at PME-adjacent prices; "the honest fair-price option" works at IPP.
• Competitive context: if a direct competitor is at $X and you are credibly better, you can sit $5-10 above them; if you are credibly equivalent, sit at parity; if positioning unclear, default to indifference.

For my products at toolgenx.com I default to indifference (IPP) because the brand positioning is "honest, no fluff" and that maps to "fair price, no premium upcharge".

Three things the survey will not tell you

1. Whether the product is good. Van Westendorp measures price perception, not quality. A product that nobody wants will produce a survey result that nobody acts on.

2. Whether to bundle. The survey is per-product. If you are planning a bundle, run a separate survey for each component price and a third for the bundle. Bundle premiums vary wildly by category.

3. Whether to discount strategically. The optimal price for a one-time payment is different from the optimal price for "regular price with a launch discount of X%". If your product will have anchored discounting, survey at the post-discount price.

When to use Van Westendorp vs other methods

Van Westendorp is the right tool when:

• You have a real audience to survey (50+ warm prospects)
• The product is similar to other products buyers can mentally anchor against
• You are pricing a discrete one-time-payment item

It is the wrong tool when:

• You have no audience yet (use competitor benchmarking instead)
• The product is novel enough that respondents cannot mentally anchor (use willingness-to-pay laddering instead)
• You are pricing a subscription or usage-tier (use feature-bundle conjoint analysis instead)

For the bulk of indie digital product pricing in 2026, Van Westendorp is the cheapest, fastest, most reliable method I know. I run it on every new product before I write the landing page price.

---

The full Van Westendorp template and the broader founder pricing framework are in Startup Launch Arsenal. The financial-modeling side (unit economics, scenario planning, pricing impact on runway) is in AI CFO Toolkit. For the wider context of why pricing is one of the four levers, see the solo founder hub post.