The cost has been worked out, a margin added, and a price written on the menu. Customers look at it, hesitate, and order the cheapest thing instead. Two months later a competitor with higher prices seems busier.
Prices work a bit like a shop window. People rarely judge one item on its own; they judge it against whatever stands beside it, against what they paid last time, and against the alternative they had in mind. For a new product there is no “last time”, so the options on the page do most of the work. This article is a way to design those options on purpose, and to test them honestly.
In short: A new product is not one price; it is a small menu. Cost sets the floor. What customers use today and what they say is reasonable set the range. Inside that range, decide which option most customers should choose, then build the other options around it so that choice is easy and fair. Test one change at a time, and never use an option that exists only to trick.
Cost tells you the floor, customers tell you the ceiling
Cost-plus pricing protects against losing money, but it says nothing about what the product is worth to a buyer. The useful questions are about the alternative: what would the customer use instead, what does that cost in money, time or hassle, and what makes this one worth more or less.
For the range, Peter van Westendorp's Price Sensitivity Meter (1976) is a simple tool that still works. It asks target buyers four questions about a clearly described product: at what price would it be so cheap that you would doubt the quality; at what price would it be a bargain; at what price would it start to feel expensive, though you might still consider it; and at what price would it be too expensive to consider. Plotted together, the answers show a band of prices that most people find acceptable. It does not give one right price, but it stops guessing in the dark.
How the options change the decision
Behavioural research shows that adding or removing an option can change which of the others people pick. Two effects matter most for a new menu.
- The decoy. Dan Ariely's Predictably Irrational (2008) describes a magazine offer with three options: web only, print only, and print plus web at the same price as print only. Almost nobody chose print only, but its presence pushed many more people towards the combined option. A third option that is clearly worse than its neighbour makes that neighbour look like obvious value.
- The middle. Itamar Simonson's “Choice Based on Reasons: The Case of Attraction and Compromise Effects” (Journal of Consumer Research, 1989) found that when people are unsure, many choose the middle option because it is easy to justify. For a new product, where buyers have no reference, the middle of three is often where the eye and the decision settle.
The honest use of these effects is to make the option that suits most customers easy to see and easy to justify. The dishonest use is an option with no purpose except to make another look good. Customers usually notice the second one, eventually, and trust goes with it.
A menu review: eight questions
These questions work on any price list, a menu, a services page or a set of packages. A weak answer to any one of them is a candidate for the first test.
| Question | Good answer | Warning sign |
|---|---|---|
| Which option should most buyers pick? | One, named before designing | “Whatever they like” |
| Is it where the eye lands? | Middle, or marked clearly | Last on a long list |
| How many options? | Two to four | Seven sizes and add-ons |
| Is each step's extra obvious? | Seen in a second | Needs a comparison table |
| Are options named by use? | “For one”, “for the family” | S, M, L or codes |
| Is the top option real? | Some people buy it | Exists only as an anchor |
| Is the price inside the acceptable band? | Checked with buyers | Copied from a competitor |
| Does every option cover its cost? | Margin checked per option | Only the average is known |
The method, step by step
- A floor for each option. Full cost per unit, including time, delivery, packaging and payment fees. Output: the lowest price at which each option still makes sense.
- A list of real alternatives. What target buyers use today and what it costs them, in money and effort. Output: three to five reference points customers already have in mind.
- An acceptable price band. Ask thirty or more target buyers the four van Westendorp questions about a clear description of the product. Output: a band, not a single number.
- A lead option. Decide which option should suit most buyers and carry a healthy margin. Output: one named option, with the reason.
- A draft menu. Build two or three options around the lead one: a real entry option for the cautious, and a real upper option for those who want more. Name them by use, and make the difference at each step obvious. Output: a draft page as the customer will see it.
- One test at a time. Change one thing, the order, a name, the gap between two options or an inclusion, for a fixed period of two to four weeks, and record what share of buyers choose each option. Output: a short log with the share per option before and after.
- A margin check. Multiply the share of each option by its margin. A menu that sells more of a low-margin option can look busy and earn less. Output: the total margin of the menu, before and after.
A worked example
This example is hypothetical, and its numbers, in Saudi riyals, are illustrative. A new car-care service in Riyadh washes and cleans cars where they are parked, at home or at the office, booked through an app and WhatsApp. It launches with two options: a single wash at SAR 60, and a monthly plan of four washes at SAR 220. Nearly everyone books the single wash, once.
The alternatives. Conversations with target buyers show three references: the petrol-station wash, which is cheap but needs a trip and a queue; the worker who washes cars in the building's car park for a monthly fee, which is cheap and regular but basic; and the mall valet wash, which is thorough but expensive and happens only when shopping. The real comparison for the monthly plan is the building washer, not the single wash.
The band. The van Westendorp questions show most buyers find a monthly plan acceptable well above the building washer's fee, if it includes the interior and arrives on a fixed day. The plan's price was fine; its contents and its name were not.
The new menu. Three options, named by use: “One-time”, a single wash at SAR 60; “Every week”, four washes on a fixed day including the interior, at SAR 199, marked as most chosen; and “Every week plus full care”, adding a monthly deep clean of seats and mats, at SAR 329. The lead option sits in the middle. The upper one is real: some owners of family cars genuinely want it.
The test and the check. Over the next four weeks, with only the menu changed, the share of buyers choosing a monthly option rises clearly, and a small share takes the top one. The margin check matters: “Every week” costs less per wash to deliver, because the washer visits one building on a set route. The menu now sells more of what is cheaper to run.
Five pricing mistakes with new products
- Pricing from cost alone. It feels safe and fair. But it ignores what buyers compare with, which is often a different product altogether. Start from the alternative, then check the cost.
- Launching low “to grow”, planning to raise later. The first price becomes the reference. Raising it later feels like a loss to the people you won. A launch offer with a clear end date does the job without moving the reference.
- Too many options. More choice feels generous, but it slows decisions and pushes people to the cheapest or to “I'll think about it”. Two to four clear options usually work better.
- Copying a competitor's price. Their price carries their costs, reputation and mistakes. Use it as one reference point, not as an answer.
- Changing several things at once. A new price, new names and a new layout in the same week. Something changes, and nobody knows why. One variable per test.
When this is not the answer
Structuring options helps most when a business sells several versions, sizes or plans and customers compare on the spot. It helps much less in a tender or a technical B2B purchase decided on specification, and not at all where prices are set or capped by a regulator, as with medicines in several markets. And if the floor is above what any buyer will pay, the problem is the cost structure, not the menu. That is work for operations and finance, not for an agency, and we say so.
How OG works on pricing
We start with market research on the real alternatives and what they cost, and customer research to understand how buyers decide and to run the price-band questions. Pricing strategy then brings it together: the lead option, the menu around it, the margin per option, and the tests to run first. Finally, the menu has to be seen the way it was designed: on a landing page, in an online store, on packaging, or in the WhatsApp catalogue, with analytics recording what people choose. Scope follows the need; there are no fixed packages.
The strongest objection
“In our market, customers just want the lowest price.” Some do, and for some products price is the deciding factor. But even price-sensitive buyers compare, and what they compare is rarely the price alone; it is what they get for it against what they use today. In the car-care example, the cheapest option did not change. What changed is that the monthly plan finally made sense next to the real alternative. A clear menu lets the price-sensitive buy the entry option instead of leaving, and lets those who value more pay for more.
Built on: our pricing work with clients since 2013; Peter van Westendorp's Price Sensitivity Meter (1976); Dan Ariely, Predictably Irrational (2008); Itamar Simonson, “Choice Based on Reasons: The Case of Attraction and Compromise Effects”, Journal of Consumer Research (1989). The car-care service is a hypothetical example.
Before closing this page
- Which option should most customers choose, and is it where the eye lands?
- What do target buyers use today instead, and what does it cost them?
- Has the price band been checked with real buyers, or only with the team?
- Is every option on the menu one that some people genuinely want?
- Is the margin known for each option, not only on average?
