Why some companies charge twice for the same product

Two machines with practically identical performance. Two software products whose feature lists overlap almost completely. One asks a certain sum, the other asks twice as much. And the market tolerates this for years, not weeks.

The first reaction is usually that the expensive one is simply trading on how little the buyer knows. I offer the second explanation as an observation from practice rather than a measurement: the more expensive supplier tends to be the one that negotiates less and loses fewer deals on price. Its product is not better. It has something else.

Price is not purely a function of production costs. The buyer is not paying for what the product costs to make, but for what they expect from it and for the risk they take on when they buy. This follows on from the article A brand is not a logo. It is the reason you can charge more. That piece dealt with the definition, this one deals with the mechanism: how pricing power comes about and why no single intervention will ever produce it.

Why the market tolerates twice the price for a functionally similar thing

When buyers decide, they do not see quality. They see signals of quality. In the used-car market, George Akerlof showed what happens when the seller knows the quality and the buyer does not: the buyer offers a price matching average quality, owners of good cars withdraw, quality falls further and the market can unravel. This is known as adverse selection.

The second half of the paper matters more. Akerlof names the institutions that counteract information asymmetry: guarantees, brand names and reputation, repeat business, licensing and certification. These are not marketing ornaments, they are instruments that reduce the buyer's risk. And risk carries a price. Two functionally similar products are therefore not two similar purchases, they are two different degrees of uncertainty. Lower uncertainty is paid for in the same way as an insurance premium.

Why the model of the perfectly rational buyer is not enough

The model of the perfectly rational buyer assumes that the purchaser knows every alternative and picks the best ratio of price to performance. It fails not because buyers are stupid, but because verification costs time and time costs money. Take a typical situation: a managing director choosing a supplier has no capacity to run a technical audit of five proposals. They decide on signals, because under uncertainty that is the economically sensible thing to do. Michael Spence, who introduced signalling into economics, described the same structure in the labour market: an employer hires without knowing actual productivity, and therefore decides like an investor under uncertainty.

Price does not only change the decision, it changes the experience. In a study published in PNAS, Plassmann, O'Doherty, Shiv and Rangel scanned 20 participants using magnetic resonance imaging while they tasted wine. The participants believed they were tasting five different wines. In reality there were three, two of them served twice with a different price tag. The same wine was labelled once as a $5 wine and once as a $45 wine. The higher price raised both the taste ratings and activity in the area of the brain associated with experienced pleasantness. Eight weeks later, in a tasting without price tags, the differences between the wines disappeared. A small sample and laboratory conditions do not support the claim that price changes the quality of the wine. They support something more fundamental: price changes the experience itself.

Perceived value, signalling and social proof

The technical term for this is perceived value: what the buyer attributes to a product before the purchase, not what the product objectively is. It is created by signalling. Spence distinguishes between the attributes a market participant cannot change and the signals they can change, but only at a cost.

Here is the condition most companies overlook. Spence argues that a signal separates the strong player from the weak one only if the cost of signalling is negatively correlated with genuine ability. Translated into business language: the signal has to be more expensive for the weaker player than for the stronger one. If anyone can imitate it cheaply, it stops carrying information. Cost here should be read broadly, including time and effort, not merely money.

This is why an attractive logo on its own creates no pricing power: practically anyone can obtain one, so it carries no information. A traceable record of delivered projects, a guarantee that actually costs the company something, a response time honoured even on a Friday evening or a transparent pricing structure are signals a weaker player cannot sustain. In a review article in the Journal of Marketing, Amna Kirmani and Akshay Rao carry this logic into marketing: price, brand and marketing spend are not merely a cost, they carry information. A meta-analytic review by Akshay Rao and Kent Monroe adds that the relationship between price and perceived quality is positive and statistically significant, with the brand effect somewhat stronger. The authors also warn that part of the measured effect is an artefact of study design: when the same person compares several prices at once, the effect comes out larger.

The third layer is social proof, the evidence that others have already trusted you. Paul Resnick and colleagues ran a controlled experiment on eBay: an established trader sold matched pairs of identical collectible postcards, once under a well-established identity with a high reputation and once under new identities that he also controlled. The difference in willingness to pay amounted to 8.1 per cent of the sale price in favour of the established reputation. This covers one seller and one product category, so it is not a universal premium for trust, but it is a clean measurement: for an identical item, people pay more to a seller they know.

Michael Luca combined Yelp ratings with tax data on restaurant revenues and found that a one-star increase in rating leads to a rise in revenue of 5 to 9 per cent. The effect did not hold for outlets belonging to a chain. What is measured here is one city and revenue rather than price, so it says nothing about how far a price list can be raised. According to Luca, reviews are a substitute for more traditional forms of reputation: anyone who has not built one by other means depends on them.

Scarcity, exclusivity and the line into manipulation

Scarcity is a legitimate positioning tool, positioning being the place a brand occupies in the customer's mind. The condition is that it is true. Consider a studio with the capacity for only a handful of projects a year: it genuinely does have to turn work away. A countdown on a website that resets itself after midnight does not simulate limited capacity, it lies about it. The difference is not aesthetic, it is informational.

The line does not have to be found by intuition. Daniel Kahneman, Jack Knetsch and Richard Thaler described it through the principle of dual entitlement: in the public's view, a firm is entitled to its customary level of profit and the customer is entitled to the price they are used to. Passing a rise in costs through to the price is therefore seen as fair, while raising the price because demand has increased is seen as unfair. In their survey, roughly 82 per cent of respondents judged it unfair when a hardware shop raised the price of snow shovels from $15 to $20 after a snowstorm. Price is judged against a reference point, not against costs. A premium justified by what the customer gets in return is legitimate. A premium justified by manufactured panic is profit bought with legitimacy.

Pricing power is produced by a system of touchpoints

Perceived value is not created in one place. It is assembled from everything the customer experiences of the company, and the weakest point decides. This follows from the logic of signals: what a company can polish is whatever it prepares for in advance, so the place where nobody made an effort reveals the most about how it operates day to day. The points that decide are mostly dull: whether visitors to the website immediately grasp what the company does and for whom. Whether the quotation is intelligible without a phone call. How packaging, handover of the work and complaints are handled. Whether the invoice arrives in the same visual world as the presentation that won the customer over.

Inconsistency is therefore an expensive mistake. A company with a premium price and amateurish onboarding is not sending a mixed signal, it is sending a negative one. What happens when it is the website that fails is covered in a separate piece, When visitors do not understand your website, you pay for traffic you never use.

Pricing power rests on a chain that has to hold along its whole length: the brand is a promise, the experience is its delivery, and processes and tools decide whether it can be delivered repeatedly as the company grows. If they do not hold, the promise breaks with the third customer in a single week and the premium disappears before it has paid for itself. This is why it pays to look at a company as one product, which is the subject of the article A company is one product. And every bad process is its technical debt.

How premium brands protect their price

Protecting a price is a discipline, not a campaign. Several recurring practices follow from the mechanisms above. These companies do not negotiate on price, they negotiate on scope: anyone who wants to pay less gets less rather than a discount, so the reference price per unit of value stays intact. They shift risk onto themselves through a guarantee, a fixed price instead of an hourly rate and an unambiguously defined deliverable, which is a direct application of Akerlof's mechanism. They control who they sell to, because an unsuitable customer turned away is cheaper than an unhappy reference.

Why discounting damages positioning

A discount presents itself as a neutral instrument for lifting volume in the short term. It is not. Carl Mela, Sunil Gupta and Donald Lehmann analysed 8.25 years of panel data on frequently purchased packaged consumer goods using a model that separates medium-term from long-term effects. The result: consumers become more sensitive over time to both price and promotional offers, as a consequence of reduced advertising and increased promotional activity. Discounts, in other words, train customers into price sensitivity. The company manufactures for itself the very market in which it can no longer command a premium.

The effect can also be immediate. Baba Shiv, Ziv Carmon and Dan Ariely showed that a discounted product delivers less real benefit to the customer. In the main experiment, 125 participants drank an energy drink and then spent thirty minutes working on fifteen anagrams. One group paid the regular price of $1.89, the other a discounted $0.89, justified by a bulk purchase. The discount group solved 7.7 anagrams on average against 9.5 at the regular price, while a control group of 31 people who had no drink solved 9.1. When the researchers deliberately strengthened expectations of the drink, performance under the discount fell to 5.8. A mediation test gave no indication that participants were aware of the effect of price, though the authors themselves note that such a result is difficult to interpret. These are laboratory experiments on small samples, yet the direction matches the other work cited here: a discount is not merely a lower margin, it is information about the product that shapes how the customer experiences it.

When a premium strategy makes no sense

A premium strategy is not universally right, and for a large share of companies cost leadership is the more honest route. It makes no sense where the product is a genuine commodity and the buyer knows the full specification, because there is nothing to signal, nor in public tenders where price is the only criterion. It makes no sense where the company cannot sustain the promised level of service, because an unkept promise damages the brand more than a lower price would. And it makes no sense when the company lacks the capital to hold its signals in place for years while a reputation settles.

Not every problem with perceived value is solved by a new website, a redesign or bespoke development. Very often it is enough to rewrite the quotation into an intelligible form, introduce a guarantee, shorten the response time or assemble off-the-shelf tools the company is paying for anyway. If the cheaper solution is sufficient, it is the right solution. Bespoke development only becomes worthwhile when off-the-shelf tools block the way the company works, and that constraint can be quantified in time, error rates or lost customers.

A price cannot be raised by a decision taken in a meeting. It can be raised once the company knows which risk it is taking off the customer's hands, which touchpoint is spoiling its promise and which signal would genuinely be expensive for a competitor to imitate. That is diagnostics, not creative work.

Key takeaways

Take one specific risk off the customer. A fixed price instead of an hourly rate, a written guarantee, a defined deliverable. The fastest route to a higher price without changing the product.
Fix the weakest touchpoint. Your price is not held up by the best point in the customer journey but by the worst. Walk through your own sales process as a stranger would.
Test every signal by asking what it would cost a competitor to imitate it within a month. If the answer is very little, it is not a signal, it is decoration.
Offer a smaller scope instead of an across-the-board discount. You protect the reference price per unit of value and you avoid training customers to wait for the next promotion.
If your market is purely price driven, do not fake a premium. With commodities and tenders, cost leadership is a legitimate strategy, it is simply executed differently: through process, volume and stripping out anything unnecessary.