Features versus benefits: the distinction copy needs

A feature is a property of the thing: what it is made of, what it contains, what it does, how it is built. A benefit is what changes for the person who owns it. Stainless steel blade is a feature. Keeps its edge through months of daily use is a benefit — and note that the benefit is only worth writing if it is true, because a duration in copy is a promise the buyer will test.

The distinction is stated in every guide to writing copy and still gets ignored constantly, which suggests the problem is not that people fail to understand it. The problem is that features are what the people making the product think about all day. They know the material, they chose it for good reasons, and those reasons were interesting. The buyer has none of that context and no reason to acquire it. Writing benefits requires deliberately setting aside the most interesting knowledge you have about your own product, which is genuinely unnatural, and no amount of understanding the rule makes it automatic.

The mechanical way to convert a feature is to ask "so what?" repeatedly until the answer is about the buyer's life rather than the product. Take a feature: the fabric is pre-shrunk. So what? It does not shrink when washed. So what? It still fits after the first wash. That last answer is the benefit, and it is the one worth writing, because it corresponds to a thing the buyer has been annoyed by before.

The procedure has a natural stopping point that is worth respecting. Pushed too far, every benefit converges on the same handful of abstractions — saves time, saves money, gives peace of mind — and those are as generic as the filler phrases they were supposed to replace. Peace of mind is where the chain has been followed one step too many. Stop at the specific, concrete consequence: it still fits after washing, you can drink from the tap, you know what you owe before the deadline. Concrete consequences are believable and memorable. Abstractions about wellbeing are neither, and every competitor is already claiming them.

The advice to write benefits rather than features gets over-applied, and the result is a page that is warm, readable and impossible to buy from. Buyers compare specifications, particularly for anything technical, anything with a size, anything that has to fit or connect to something they already own. A buyer who cannot find the dimensions, the capacity, the material, the compatibility or the power rating does not conclude that those things are unimportant. They conclude that the page is unhelpful and go to the one that lists them.

The resolution is that features and benefits do different jobs at different moments and belong in different parts of the page. Benefits do the persuading, so they go where a reader who is still deciding whether to care will encounter them. Features do the verifying, so they go lower, in a form built for scanning rather than reading — a table or a list, not prose. Someone hunting for a dimension is not reading sentences; they are looking for a number, and burying that number inside a benefit statement is a small, common cruelty that costs sales from the most motivated visitors on the page.

There is a failure mode in benefit writing that is worse than dull feature copy, and it comes from the fact that benefits are inherently claims about outcomes. "Stays sharp for two years" is a warranty in everything but name. "Cuts your bookkeeping time in half" is a measurable assertion somebody will measure. Written casually as copy, these get treated as marketing language; read by a buyer, they are what was promised — and a promise that fails produces a return, a complaint and a review.

So the discipline is to write only benefits you would stand behind if challenged. Where you cannot substantiate a specific figure, write the mechanism instead of the number: not "cuts bookkeeping time in half" but "receipts are entered once instead of copied between two places". The mechanism is often more persuasive anyway, because it explains why the improvement happens rather than asserting a magnitude the reader has no reason to believe. A specific number you cannot support is the least defensible sentence on any page, and it is usually the one a dissatisfied customer quotes back.

One feature usually produces several benefits, and which one to lead with depends on who is reading. A same-day dispatch feature benefits an impatient consumer as speed, and benefits a business buyer as the ability to hold less stock. Those are different sentences aimed at different worries, and writing both in one paragraph produces a muddle that speaks precisely to nobody.

This is where knowing your actual buyer stops being a marketing platitude and becomes an editing decision. If most of your sales come from small retailers rather than end consumers, the stock-holding version is the one that belongs at the top. Where a page genuinely serves both audiences, it is usually better to segment visibly — a short section addressed to each — than to write one paragraph attempting to cover both, because a reader can skip a section clearly labelled for somebody else without concluding the page is not for them. What does not work is averaging the two into copy that is technically true for both and compelling to neither.

Take any page you already have and mark every sentence as feature, benefit, or neither. The "neither" pile is usually the largest and the most interesting: company history, self-praise, transitions that carry no information, and filler like premium quality that could appear on any competitor's page. That pile is where the space for the benefits comes from, and deleting it costs nothing because it was communicating nothing.

Then check the order. Are the benefits above the features, and is the first benefit the one a buyer would actually decide on? Finally, check the features are findable — that somebody looking for a dimension or a compatibility can locate it without reading prose. This takes about twenty minutes per page and tends to produce a shorter page that performs better, which is a slightly deflating result if you enjoyed writing the original. It is also the most reliable improvement available, precisely because it requires no new insight about the product, only a willingness to remove the sentences that were there to fill space.