Content: answer first, date every number, sound like a person

The Book 9 page on The Crawler's Cookbook leads with its answer, says when that answer was last checked, and drew 13.8K Google clicks from Jul 11 to Oct 8, 2026. Source: Google Search Console daily and page export.

The public Book 9 page on a phone, with its short answer that no release date has been announced
The live page at 390px, captured October 11, 2026.

How the page is put together, top to bottom

  1. The reader's question is the heading. “When is Dungeon Crawler Carl Book 9 coming out?” sits in bold at the top of the short answer.
  2. The answer comes first. No release date has been announced yet, said in the first line, before any background.
  3. Rumors get named and ruled out. Dates and titles that circulate without coming from the author are called out as not from him.
  4. The status carries a date. “Release date: not announced. Last checked October 8, 2026.”

A search result or an AI answer can lift the first lines of a page and show them on their own. When those lines are the answer, the source and the date, the page is still right when it's quoted out of context, and a reader who clicks gets the detail the summary left out.

The goal is no longer to be the first answer a user sees. The goal is to remain useful after the first answer has already been given.

Every number carries its date and where it came from

A number without a date can't be checked, so the numbers on this site are rendered from data files that name their source and measurement window. The click count above comes from one of those files, not from a sentence someone typed.

My post on how AI picks which brands to name runs on the same rule: every figure in it comes from raw exports pulled on one stated date, July 18, 2026.

How those files are checked on every build: the systems page.

It should sound like a person wrote it

These lines are mine, quoted exactly from my old site. The build fails if a template changes a word of them.

  • I was going to Autzen Stadium before I could walk.
  • I have made so many mistakes.
  • I had a linktree (why, dumb, I was younger…)

The writing and the case studies, and what each one shows

  1. Structuring content for answer-first searchHow a page stays useful after a search result or AI answer has already given the short version.
  2. The recommendation layerA long post where every figure carries the date it was pulled.
  3. Why impressions rise while clicks stay flatReading a Search Console pattern before deciding what it means.
  4. Search Console as a visibility dashboardWhich Search Console reports to read, and how to turn them into a report a client understands.
  5. Personal audit: where AI output failsMy own mistakes, ranked by how much damage they could do.
  6. CheckMyTapA case study that reports a decline it can measure and says plainly what it can't explain.
  7. The Crawler's CookbookThe answer-first page above, told as a full case study with its method.

The thinking behind the writing: how I think.

Read the writing itself
Ask Bella

Ask Bella Hayden's corgi. Answers from this site, every fact sourced.

BELLA

WOOF. I'm Bella, Hayden's corgi, and he's my human. Ask me about him: what he does, how he works, what he's built. Every answer here was written ahead, checked against this site, and cites numbered sources you can tap.

How it works: answers written ahead and checked against this site's pages, plus a search of the site that quotes it word for word. Every claim has a numbered source you can tap. Nothing is generated while you wait, nothing is made up, and nothing you type leaves your browser. Every answer, in plain HTML

Top