The Product Is Never Just the Object

Raymond Loewy didn’t start from an abstract design theory. He started from a practical observation, repeated client after client across five decades of consulting: people want something new, but only up to the point where they stop recognizing it as their own. Loewy summarized it as two opposing psychological forces that coexist in every consumer — neophilia, curiosity for the new, and neophobia, fear of the too different — and a formula for resolving the tension between them: MAYA, most advanced yet acceptable.

The formula doesn’t describe an object; it describes a design decision. A completely familiar product generates no desire because it offers nothing the customer didn’t already have. A completely new product generates no trust because it gives the customer no reference point from which to evaluate it. The designer’s job — of product, not merely of form — is to find the exact point where the object feels like a step forward without ceasing to feel familiar.

This explains why Loewy never invented new product categories. He redesigned refrigerators that already existed, bottles that were already selling, packs of cigarettes that half the country was already smoking. His genius wasn’t in invention but in translation: taking a need already met mediocrely and solving it in a way that felt, at once, modern and obvious.

The rest of this chapter repeats that same question in contexts where the object is no longer an object — a loan, a bet, an entire streaming catalog — but where the tension between the known and the new remains the force that decides whether a product gets adopted or ignored.

In 2000, Blockbuster dominated video rental with thousands of stores around the world, and a revenue stream the company itself never wanted to let go of: late fees, which generated around 800 million dollars a year. That figure reveals the underlying mistake — Blockbuster had turned the very friction its customers hated most into a business model. The customer’s real need was never “being able to rent a video”; it was “watching a movie tonight without that turning into a punishment if I return it late.”

Reed Hastings and Marc Randolph founded Netflix in 1997 betting on exactly that poorly served need: a monthly subscription, no return date, no late fee, with the catalog arriving by mail instead of requiring a trip to the store. Netflix didn’t invent the DVD or improve movie quality — it attacked the psychological pain point that Blockbuster had normalized as part of the business.

Blockbuster had the chance to buy Netflix for 50 million dollars in 2000 and turned it down, convinced the business was still the physical store and the late fee. By the time it finally tried to copy the subscription model, it was too late: it had built its entire cost structure around a friction it now had to dismantle while a nimbler competitor scaled without it. Blockbuster filed for bankruptcy in 2010.

The lesson for any product, physical or digital, is the same one Loewy taught from another angle: when a company confuses its revenue model with the customer’s real need, any competitor who solves that need without the friction ends up winning, even if their product — in this case, a disc mailed to your door — is objectively simpler than the original.

A loan has no shape, weight, or color. It doesn’t exist as an object at any point in its lifecycle — it is, literally, a contractual promise of future money. And yet it meets absolutely tangible needs: a house to live in today, a car to drive this week, a surgery that can’t wait for the next paycheck. The product is intangible; the need it meets is as physical as any Coldspot refrigerator.

This inversion — a bodiless product solving a need with a body — is what makes credit a limit case of the MAYA principle. Accepting a loan requires a far bigger leap of trust than buying an object: the customer signs away their name in exchange for money that doesn’t even exist yet in their account, trusting a system — the bank, the score, the contract — they almost never fully understand. The more abstract the product, the more the acceptance side of MAYA (the “A”) outweighs the advancement side (the “M”): a loan that’s too novel in how it’s approved or disbursed generates distrust before desire.

That’s why the credit products that work best don’t compete solely on interest rate — they compete on shrinking the psychological distance between “borrowing money,” which triggers anxiety, and “solving a need today,” which triggers relief. A loan approved in minutes from a phone isn’t just faster than one processed at a branch: it changes the emotional meaning of the act of asking for money, from an uncomfortable negotiation into a transaction as simple as buying any other product.

The gambling case takes this same logic one step further: it’s no longer an intangible product solving a material need, but a product that doesn’t even need to solve a need — it only needs to trigger a psychological mechanism that already exists in the user.

A gambling product meets no material need at all — it doesn’t even try to. Its design rests on a principle of behavioral psychology described by B. F. Skinner in the mid-20th century: variable-ratio reinforcement. Unlike a reward that arrives every fixed number of attempts, a reward that arrives at an unpredictable interval generates a much higher response rate and far more resistance to extinction — the user never knows whether the next attempt will be the winning one, so they keep trying.

Skinner himself noted that if a gambling business can’t convince a customer to hand over money with no return at all, it can achieve the same effect by giving back part of that money under a variable-ratio schedule. That principle, formulated in the lab with pigeons and levers, is exactly the mechanism behind a slot machine, a roulette wheel, or any digital gambling product: the object — physical or on-screen — is only the surface on which a reinforcement pattern runs, designed to maximize repetition, not satisfaction.

Here the MAYA principle flips in an unsettling way: a well-designed gambling product carries no tension between the new and the familiar, because the familiarity of the ritual — the same machine, the same gesture, the same wait — is precisely what sustains the hook. What’s “advanced” isn’t the shape of the object but the sophistication with which the system calculates when to deliver the next reinforcement to maximize time on device.

Understanding this mechanism isn’t an invitation to use it without limits — quite the opposite: it’s the reason any product team working near variable-reward mechanics (a notification, a streak, a surprise box) carries a design responsibility different from that of a refrigerator: the same principle that generates desire can, with no additional friction, generate dependency.

A fast-food chain wanted to sell more milkshakes and tried the obvious path: it surveyed customers, asked whether they wanted them thicker, sweeter, with more fruit chunks, and adjusted the recipe based on the answers. Sales didn’t move. Clayton Christensen’s team took a different path — they stood outside the restaurant and asked every person walking out with a milkshake in hand a single kind of question: “what job did you hire this milkshake to do?”

The answer revealed two completely different jobs depending on the time of day. In the morning, buyers were solitary commuters on a long, boring drive: the milkshake competed against a banana, a donut, or a coffee — not against other milkshakes — because the job was “something that can be held with one hand, that lasts the whole commute, and that keeps me full until lunch.” In the afternoon, the job was entirely different: a parent buying their kid something to keep them calm, competing not against breakfast but against a toy or a piece of candy.

That finding is the essence of the Jobs to Be Done framework: people don’t buy products for their attributes — they “hire” them to make a specific kind of progress at a specific moment in their lives. A user isn’t looking for “a streaming app”; they hire something for the job of “filling an evening with no decision effort” or “sharing an experience with my partner without leaving the house” — jobs neither Netflix nor Blockbuster would have identified by simply asking “which movies do you prefer to rent?”

A well-written Job to Be Done has three layers: the functional (what task it solves), the emotional (how the user wants to feel while solving it), and the social (how they want to be perceived by others while solving it) — the same three layers Loewy, without naming them that way, was already solving with the Coldspot, the bottle, and the pack.

Eliciting a need — bringing it to the surface, in the literal sense of the verb — almost never works by asking for it directly, because most users have no conscious language for their own job. The techniques that do work share one principle: observe real behavior before stated opinion. Contextual interviews inside the environment where the product gets used, shadowing the full flow of a task, usage diaries the user fills in the moment — not afterward, once they’ve already rationalized their decision — and the “job” question, exactly as Christensen’s team used it outside the restaurant.

Once a need has been elicited, the next challenge is deciding how advanced or how familiar the solution should feel — the same MAYA tension, now turned into a design-decision tool: the Desire and Usability matrix (D&U). The Desire axis measures how much the user wants this specific solution compared with the alternatives they already know; the Usability axis measures how much cognitive effort adopting it demands. A product with high desire and low usability generates curiosity but adoption friction. A product with high usability and low desire gets adopted effortlessly but generates no competitive advantage at all.

The quadrant a product team is after is the same sweet spot Loewy used to sketch by hand: high desire held up by a usability that doesn’t force the user to learn a new language to get the job done. That quadrant is not a fixed point — it shifts as the market matures and the user accumulates experience — which is why MAYA is not a design rule but a moving target that has to be re-evaluated every time the product, the competitor, or the user changes.

Once a product exists and runs in the market, the designer’s job changes: it is no longer about finding the need but about deciding what to do with the answer that already works. The Boston Consulting Group matrix — developed in the 1970s but still useful as a first-pass diagnostic — organizes that decision around two axes: market growth and the product’s relative share within that market.

A Star is a product with high growth and high share: it dominates a market that still has room to grow, and the right move is to invest to sustain that leadership before the market matures. A Cash Cow generates steady revenue in a mature market where share is high but growth has slowed: the right move is to harvest its margins and use them to fund the Stars and Question Marks. A Question Mark has high market growth but low share: it’s a bet that hasn’t paid off yet, and the team has to decide fast whether to invest aggressively to build share or exit before the window closes. A Dog has low growth and low share: it consumes resources without strategic return, and the right move is almost always to divest or discontinue.

The matrix doesn’t describe a product’s identity — it describes a product’s position at a specific moment, which changes. A Star matures into a Cash Cow. A Question Mark becomes a Star if the investment is right, or a Dog if it isn’t. The key decision isn’t which quadrant a product is in today, but which quadrant it is moving toward — and whether the team’s investment, attention, and design energy match that trajectory.

When a product reaches Cash Cow status — stable, generating margin, but with limited growth — the classical move is to have children: derivative products that expand into adjacent segments using the same core platform. The four-color phone, the refrigerator in three sizes, the streaming app with a family plan and a premium tier — all of them are product families built from a single core, each targeting a slightly different job while sharing the investment already made in the platform.