The problem is a blank comparison slot
When a buyer reads your homepage, they are doing one thing before anything else: filing you. They are looking for the shelf you go on, because the shelf tells them what questions to ask, which internal budget line you come out of, who else they should look at, and roughly what you should cost. This happens in seconds and it happens whether or not you have given them any help.
If your product does something the market already has a name for, that filing is free. You say data warehouse, or applicant tracking, or endpoint security, and the buyer's whole apparatus for evaluating one of those switches on. You are then arguing about which one, which is a much easier argument than whether.
A genuinely new category means that slot comes up blank. The buyer has nowhere to file you, so they do the thing humans do with unfamiliar objects: they file you under the nearest familiar thing anyway. Usually that is the incumbent you are least like. You do not get to skip the filing step. You only get to decide whether you supply the answer or let the buyer improvise one.
Three tests before you claim you are new
Most companies that believe they are creating a category are not. They are building a better instance of something that already exists, with a genuinely novel mechanism inside it. That is a good business and a much cheaper story to tell. Run these three tests honestly before committing.
- The line-item test. When a customer buys you, which budget does the money come out of, and what did that budget previously pay for? If there is an existing line item, you are in an existing category and you are the replacement for whatever is on it.
- The displacement test. Ask five customers what they would do if you disappeared tomorrow. If they name a competitor, you are in a category. If they name a spreadsheet, an internal tool, a contractor or nothing, the category may genuinely be absent.
- The job title test. Is there a person whose job is to buy this? Categories tend to arrive with roles attached. If nobody in your buyer's company has a remit that obviously covers your product, you are creating one, and part of your work is making that role exist.
What creating one actually costs
A new category is not a positioning statement, it is a market education programme, and it is paid for in time rather than in a campaign budget. You are funding the period during which buyers learn a new word, learn why the old approach was a problem, and learn who is credible in the new space. That period is longer than most Series A runways.
It also changes your sales motion. Selling into an existing category, the deal is a comparison. Selling a new one, every deal contains a mini education, which means longer cycles, more stakeholders, more champions who have to defend an unfamiliar purchase internally, and a much heavier demand on your content and sales collateral. Your marketing has to carry work that a category would otherwise carry for free.
The reason to accept that cost is when the alternative is worse. If filing yourself in the nearest existing category makes you look like an expensive, feature-light version of the incumbent, then the category is doing you active harm and paying to leave it is rational. That is the actual test. Not whether the category is boring, but whether it prices you and frames you wrongly.
The move most companies should make instead
There is a middle option and it is the right one far more often than either extreme. You accept an existing category as your filing label, and you take a strong, specific position inside it that the incumbents cannot copy without breaking their own business.
Concretely: you keep the noun the buyer already knows and you own an adjective in front of it. The noun does the filing. The adjective does the differentiation. Buyers get the shelf for free, and you spend your entire marketing budget on the one thing that is genuinely yours instead of splitting it between teaching a word and selling a product.
This is unglamorous and it works. It is also reversible: a company that owns a distinctive position inside a known category has a much stronger platform for renaming the space later, once it has customers who will repeat the new word. Category creation from a position of strength is a different activity to category creation as a substitute for one.
If you really are creating one, write these four things
The homepage of a genuine category creator has a specific job, and it is not to describe the product. It is to install a way of thinking that makes the product obvious. Four artefacts do most of that work.
- The named problem. Give the old way a name and make it unattractive. Buyers cannot want a solution to a problem they have not agreed exists, and a problem without a name cannot be discussed in a meeting you are not in.
- The reason now. Something changed that made the old approach stop working: a shift in how software is built, a regulation, a cost that used to be small. Without a because-now, a new category reads as a preference rather than a necessity.
- The comparison you are inviting. Pick the thing you want to be compared to and put it in your own copy. Buyers will make a comparison anyway. Choosing it is cheaper than correcting it.
- The proof that the new way works. Not testimonials about how nice you are to work with, but evidence that the mechanism does what you say. In a new category this is the scarcest asset you have, and it should decide which claims you are allowed to make at all.
A worked example
Take an invented company, Harbour, selling software that watches a build pipeline and rewrites failing configuration automatically. The founders believe this is a new category and want to call it autonomous delivery.
Run the tests. The line-item test says the money comes out of a developer tooling budget that already pays for CI. The displacement test says customers would go back to their existing CI plus an engineer's afternoon. The job title test says the buyer is a platform engineering lead, a role that already exists and already buys tools of this shape.
Two of three say there is a category, and it is developer tooling for build pipelines. So Harbour keeps the noun and takes the adjective: it is the build pipeline that fixes itself. The filing is free, the differentiation is specific, and the mechanism is the claim. The word autonomous can come back in two years when there are customers to say it for them.
What to do this month
Write down which category you are in, in the words your buyer would use, not the words your investors would use. Those are frequently different documents and only one of them is on the homepage.
Then take the ten most recent closed-won and closed-lost deals and write down what each buyer compared you to. If the answers are consistent, you have your category and your positioning problem is a sharpening problem. If the answers are scattered, you have a filing problem, and no amount of design work will fix it until somebody decides the answer.