What survives the cut

The problem

A founder spends a summer writing fifty careful pages, is told they get lost in the detail, and then has to produce twelve slides of twenty words each. Nothing on the market helps with that step. Deck templates assume the content is already decided, and generators produce a confident deck with no account of what was thrown away. The founder cannot see the loss, so they cannot judge it, and they keep the sentences they are proud of instead of the ones an investor needs. The real work is the series of cuts between plan and deck, and it happens invisibly in the founder's head.

Why now

Generative tools have made a polished deck cheap, which moved the scarce skill from layout to editing. Investors skim, and a deck that carries methodology or a five-year roadmap loses them on slide three. At the same time, founders are used to tools that show a diff, so a compression that shows what it removed and why is now a natural thing to ask for. The demo shows that view on one hand-written sample, because the visible loss is the part worth testing first.

What it is

One long document and a density control with four stops: full text, one page, twelve slides, one line. Moving it re-renders the same content at each density and marks every sentence kept or cut in the source, with a stated reason for each. Selecting a slide marks the sentences it came from. The demo runs on a single invented plan with compressions written by hand and labelled as scripted suggestions. It has no summarizer, accepts no pasted text, and does not claim to judge anyone's plan.

Who it's for

For founders with a long plan and a deadline, mostly technical and pre-seed, who over-explain. It is not for people who want a designed deck handed to them; layout, themes and export are deliberately absent, and the generator category already serves them. It is also not for investors or advisors reviewing decks, and it does not replace a human editor who knows the market. It helps the founder see their own cuts, nothing more.

How it makes money

Nothing yet, and that is correct while the core claim is untested: that showing the cut beats hiding it. A real version needs an actual summarizer, which brings per-document cost, so the likely shape is a small one-off fee per plan, around the price of a coffee-shop editing session, rather than a subscription founders would use for a fortnight. I reject a free tier funded by deck-template upsells because it pulls the product back toward design, which is the crowded part.

Go to market

Start with one place: accelerator cohorts in the weeks before demo day, where fifteen founders are all cutting a long plan at once and compare notes. Offer a cohort a worked example built from a public, consented plan, and ask the programme manager to share it, because it saves them repeating the same feedback to each team. The demo link is the artifact they forward. Reddit threads of founders asking exactly this question are a second source of first users.