thoughts on product design

I think a product should leave its user in a better place than it found them. This sounds like a moral statement, but I mean it first as a design test. Something should have changed in the user's favor. They should know more, be able to do more, have saved time, avoided a mistake, made something, solved a problem, or enjoyed an experience that was worth the attention it required.

The standard is deliberately broad. A product does not need to cure a disease or reorganize an industry to be worthwhile. A small tool can be worthwhile. A game can be worthwhile. Music, film, and conversation can leave a person better without producing a measurable return. Utility is not limited to productivity. But a product should be able to explain what it gives, not only what it takes.

This is the idea that sits underneath how I think about cleanliness, usefulness, business models, and even the pleasure of building. Product design is not only the arrangement of screens. It is the arrangement of the relationship between the builder and the user.

The clean line

Clean products are often confused with visually minimal products. White space, restrained color, and fewer controls can help, but cleanliness is not a style. A beautiful interface can still be conceptually dirty. It can hide costs, blur important states, make cancellation difficult, or place five secondary ideas in the path of one primary task.

Cleanliness means that the shape of the product is understandable. The user can tell what the product is for, what is happening, and what will happen next. Important actions are not disguised. Errors are named. Exits exist. The product does not require the user to learn the company's internal organization before they can complete a simple task.

This kind of clarity is difficult because companies naturally accumulate interests. Growth wants another prompt. Marketing wants another surface. Finance wants a new tier. A new feature arrives with its own terminology and settings. Each addition is locally defensible, while the product as a whole becomes harder to understand.

Good product design has to remove as actively as it adds. It decides which interests are allowed to enter the interface and which must stay outside it. The result may still be complex, but the complexity should belong to the problem rather than to the company.

The invitation

Another useful test is whether a person would seek the product out of their own accord. Distribution is part of business, and there is nothing wrong with showing people that a useful product exists. The concern is what happens when the distribution mechanism becomes stronger than the reason to use the product.

A product can be bundled into another service, forced into a workflow, kept alive through constant notifications, or made difficult to leave. It may report excellent numbers while very few users would independently choose it. The company has acquired access without earning preference.

This matters because voluntary use is evidence. It suggests the product has utility that survives outside the company's ability to capture attention. If the ads stopped, the notifications disappeared, and the product had to be remembered rather than repeatedly inserted into a person's day, would they return?

Not every product needs daily use. In fact, many good products should disappear for long periods because they solved the problem they were meant to solve. Frequency is not the same as value. A fire extinguisher can be extremely useful while almost never being used. Product metrics become misleading when they treat every return visit as success and every completed need as churn.

Value before the toll

A business needs to capture some value in order to survive. Employees must be paid, infrastructure costs money, and useful products require maintenance. I am not against revenue or profit. A product that cannot support itself may disappear regardless of how good its intentions are.

The order matters. The product should create value before the company becomes clever about extracting it. When extraction is designed first, the user is treated as a resource: attention to sell, behavior to shape, data to collect, or friction to monetize. The interface becomes a tool for converting human impulses into company metrics.

Advertising is not automatically disqualifying. It can fund products that users could not otherwise afford, and it can help people discover things they genuinely want. But a product supported by advertising should still have utility beyond the ads. If the advertisements were removed, there should be a reason for the product to remain.

The same question applies to data. Collecting information can make a product better. It can remember preferences, detect failures, and reduce repetitive work. The test is whether the collection serves the product the user chose or quietly creates a second product in which the user is the raw material.

The house and the crowd

My discomfort with betting products and much of crypto is not mainly that they are immoral. That framing is too easy and often not very useful. Adults can choose risk, speculation can play a legitimate role in markets, and entertainment does not need to be educational to justify itself.

The problem is that many of these products struggle to show how the user is better off after using them. The business can do well precisely when the user continues to place bets, trade more often, or remain inside a cycle of anticipation and loss. The strongest product behavior and the user's long-term interest can point in opposite directions.

Crypto sometimes promises utility in the future while offering price movement in the present. The immediate product becomes the possibility of selling the asset to someone else at a higher price. There may be real technical work underneath it, but the ordinary user is often introduced first to speculation and only later to the proposed use.

Betting is more direct. It sells the feeling attached to uncertainty. That feeling can be enjoyable, just as a difficult game can be enjoyable. But the company earns more when the user repeats the wager, and repetition can gradually turn entertainment into harm. A product designed around that loop has a special responsibility to place limits on itself.

The issue in both cases is alignment. Does the company prosper when the user gets what they came for, or when the user remains unsatisfied enough to continue? A product does not need perfect alignment, but persistent opposition between company success and user wellbeing eventually shapes every design decision.

The cost of returning

Products are often described as addictive as if addiction were simply a compliment. Usually the speaker means that the product is compelling. Compelling products are not necessarily harmful. A book can keep someone awake. A game can absorb an entire afternoon. People can become deeply attached to exercise, music, work, or conversation.

Intensity alone is not the problem. The important question is what the intensity does to the person. Does it damage their health, reduce their control, exploit a vulnerability, or make stopping feel like a failure? Does the product respect sleep, money, concentration, and the ability to leave?

Builders cannot control every pattern of use, and users remain responsible for their own choices. But design changes probabilities. Infinite feeds, artificial urgency, loss streaks, variable rewards, and deliberately inconvenient exits are not accidents. They are decisions about how much pressure the product is willing to apply.

A product can be delightful and habit-forming without being detrimental to health. It can create natural stopping points. It can make time and cost visible. It can allow a person to return because they want to, rather than because the product has made absence uncomfortable.

The part worth making

I care about joy in the process of making products. This is not because users owe builders an appreciation of the work. They do not. Most users should never have to think about the internal effort at all.

But the builder's attitude enters the product indirectly. Work made with attention tends to contain small acts of respect: a clear error message, a sensible default, a fast path through a common task, a detail that exists only because someone wanted the experience to feel complete. These decisions are difficult to produce through indifference.

Joy is also protective. When building is reduced entirely to metric optimization, it becomes easy to treat users as numbers moving through a funnel. Enjoying the craft keeps part of the work attached to the object itself. The question is not only whether a change increases conversion, but whether it makes the product better.

This does not mean indulging every idea that is fun to build. The user should not be forced to carry the builder's experiments. Joy still needs discipline. The best form of it is visible as care, not as unnecessary cleverness.

The simple question

The test I return to is uncomplicated: after using the product, is the person in a better place?

They may have completed a task, understood something, been entertained, connected with another person, avoided a risk, or simply had a good hour. The improvement does not need to be permanent or profound. It should only be real enough that the product can point to it without first pointing to its own growth.

A clean product makes that value easy to reach. A respectful product allows the user to choose it and to leave it. A durable business captures enough value to continue without turning the user into the thing being consumed.

Product design is therefore not neutral decoration around a business model. It decides what the company asks from people, what it gives them in return, and how much pressure it applies in between. The interface is where those decisions become visible.