Carrd is a genuinely excellent piece of software with an unusual property: it does one thing and has resisted growing. A single page, hand-arranged, hosted for nineteen dollars a year on the Pro Standard tier that unlocks forms. Mailchimp behind it holds the addresses and sends the eventual launch email, and its free tier covers the volume any pre-launch list will reach.

For the first idea this beats everything, including this. You will have a page live before you have finished reading a comparison of the alternatives, and the page will look better than most, because Carrd's constraint is the reason its output has taste.

The combination is designed for a launch: one page, one list, one send. That design is exactly what fails when the job is a comparison rather than a launch.

The first is the second page. A landing page alone is invisible to search: nobody types your product's name because it does not exist yet. Traffic comes from the calculator, the comparison, the template — pages about the problem rather than the product. Building those by hand, per idea, is where the evening becomes a fortnight.

The second is the second language. Duplicating a Carrd site into Portuguese is straightforward; keeping reciprocal hreflang correct across two languages, sixteen products and two hundred pages is not a design task, and doing it by hand produces exactly the kind of silent error that takes six months to notice.

The third is the list itself. Mailchimp is built to hold addresses and send to them later, which is a reasonable thing to want and a slightly uncomfortable thing to have when the product may never exist. A signup that goes to a Slack channel and nowhere else cannot be accidentally mailed in eighteen months by a version of you who has forgotten the promise.