Snackz.ai

SnackzLAB generates the fields a German title needs to be found: Warengruppe, Thema codes, BISAC headings and qualifiers. It also suggests keywords and shows search volume on Amazon and Google, so you can see whether a term is worth using. The output is formatted for the VLB record, which cuts the copy-and-paste work between systems.

Instead of guessing which phrases readers type, the tool proposes keyword sets and attaches demand figures from two of the biggest search surfaces in book retail. That turns a subjective argument about wording into a comparison of numbers. You still decide which terms fit the book.

The marketing tool reads the text first, then drafts a book profile for people who will not read the whole thing, social media concepts and blog articles for publisher or author channels. Because the drafts come from the actual content, they tend to include specifics rather than generic praise. Editors can rewrite in place.

A separate press tool produces a short release, topic pitches with suggested outlets, and an interview concept with questions drawn from the book. For small houses without a press officer, this covers the first draft of a campaign. Journalists still need a human contact and a reason to care.

You can define unlimited custom text styles, so output sounds like your imprint instead of a neutral assistant. A literary fiction house and a guidebook publisher can run the same tool and get noticeably different prose. This is the feature that decides whether the copy is usable without a rewrite.

Initial metadata can be generated from a few notes before the manuscript is finished and updated later, which fits a publishing schedule where catalogues are planned months ahead. Backlist optimisation runs in batches, so a long list does not mean a long queue of single jobs.

SnackzAGENT is a book discovery chatbot that lives in a bookseller's webshop, talks to visitors about books and recommends titles from the shop's own catalogue. It captures email addresses, keeps people on the page longer and reports anonymised interest data. Setup needs catalogue access and an iframe, plus styling to match the shop.

Finished material leaves the platform in the format your systems already read, including ONIX for the book trade and XML for internal pipelines. Markdown covers the blog and social drafts. An API route exists for teams that want to push records automatically instead of downloading files.