Case study · The Bookish Elf

A book publication engineered to be found.

The Bookish Elf is a book review publication, running since 2018. This is the platform and search work behind it, from structured book data to speed, mobile and indexing.

Visit bookishelf.com
Challenge
A large archive can bury its best work: pages go unlinked, archives stay thin, and templates slow down.
What we built
A structured book database, designed templates and an SEO programme on top of WordPress.
Result
817K Google impressions in 28 days, with search clicks up 55% on the period before.
bookishelf.com
The Bookish Elf homepage scrolling inside a browser frame
Type
Content platform
Status
Live
Running since
2018
Platform
WordPress with a custom plugin
Work
SEO, web development, content

817K

Google search impressions in 28 days

+55%

search clicks on the previous 28 days

3,657

published reviews

203

author interviews

200K+

readers, including social followers

646 → 0

posts with no internal links

Search figures from Google Search Console, 28 days to 28 October 2026. Review and interview counts from bookishelf.com.

01

The problem

A review site's value sits in its archive, and an archive only works if people can find it.

Thousands of reviews had built up on a general WordPress theme. Book details lived in free text, so one publisher could have several spellings, and an author, series or year had no page of its own. 646 posts had no links from other posts at all.

Phones and speed lagged too. Mobile visitors were redirected to a stripped-down page with no menu, and every page carried hidden theme markup it never used.

The goal was to give every book, author and award a page worth landing on, on a site quick enough to keep the reader there.

02

Who it's for

The site serves readers first, and the same work applies to any business with a large content site.

Readers

People choosing their next book, who want honest reviews, reading order and what's coming next.

Authors and publishers

Authors looking for reviews, interviews and book marketing, with a page that does their book justice.

Content businesses

Teams with large sites facing the same problems: thin archives, slow templates and pages Google ignores.

03

What we built

How one review feeds the whole siteInterface illustration

New review

Middlemarch

George Eliot · Blackwood · 1871

  • Author pageBio, ratings and where to start
  • Series orderEarlier and upcoming books
  • PublisherEvery title from the imprint
  • Release monthWhat else came out
  • AwardPrize coverage in one place

Structured book data

Author, publisher, series, year, translator and narrator moved from free text into linked data across about 3,000 reviews.

Designed archive pages

Real pages for every author, publisher, series, year and award, replacing generic theme archives.

A new review template

Ratings, reading notes, series order and structured data on every review, with genre-led styling.

Richer author pages

1,701 authors enriched with bios, achievements and trivia, plus a suggested book to start with.

Release calendar

Month pages for new releases, and an anticipated-books hub with countdowns and a calendar feed.

Homepage and site chrome

A new homepage, header and footer, with a read-alike finder that suggests similar books.

Mobile redesign

The full site on phones, replacing a stripped-down mobile version.

Book marketing services

A dedicated page and enquiry flow for authors and publishers.

Most anticipated books hub with counts of books, authors and months, and a release countdown
The anticipated books hub, with countdowns, a release calendar and an editorial take on each book.
Kristin Hannah author page with a bio, average rating and a suggested book to start with
Author pages turn reviews into a useful landing page.
October 2026 new releases page showing book counts by week and genre
Month pages list every new release, with a verdict on the ones reviewed.
Booker Prize award page showing books reviewed, winners and the 2026 shortlist
Award pages bring all the coverage together in one place.

04

Architecture

WordPress stays as the publishing system, so editors keep their workflow and existing URLs keep their rankings. A custom plugin adds the book database, templates and search work on top.

What readers see

  • Custom templatesReviews, authors, awards, months
  • Site chromeShared header, footer and homepage
  • Structured dataReview, book and breadcrumb schema

Platform

  • WordPressEditing, media and publishing
  • Custom pluginBook taxonomies, routes and pages
  • Feature switchesBuild off, preview, then go live

Data and automation

  • Book databaseAuthors, series, publishers, years
  • Weekly release importKeeps upcoming books current
  • Audit toolsLinks, overflow and indexing checks

Delivery and search

  • CloudflareEdge delivery in front of the site
  • SitemapsOnly pages worth indexing
  • Search ConsoleMeasures what changed

05

Key engineering decisions

1

Extend WordPress instead of replacing it

Why

Editors keep their workflow, and thousands of existing URLs keep the rankings they've earned. New features live in a plugin we control.

Trade-off

We work around an older theme and caching plugin, which adds conflicts a fresh build wouldn't have.

2

Turn book details into real data

Why

Linked data creates a page for every author, series and publisher, and those pages answer searches single reviews can't. 478 publisher spellings became 413 clean pages.

Trade-off

Merging needs judgement. We kept imprints separate, because merging them would have been wrong for readers.

3

Keep thin pages out of the index

Why

Year, publisher and series archives with too few reviews stay out of Google until they have something to offer. Author pages stay in, so debut authors can be found.

Trade-off

Some pages don't appear in search until they fill out, which makes indexing reports look worse than they are.

4

Link from real mentions

Why

Internal links come from places a post already names a book, matched strictly, so each link helps the reader instead of padding the page.

Trade-off

It's slower than automatic link blocks, and every plan needs checking before it's applied.

5

Ship everything behind a switch

Why

Each feature goes live off, gets previewed on the real site, then turns on. If something misbehaves, it turns off without a deploy.

Trade-off

More code to carry, and old paths to remove once a feature settles.

06

The hardest problems

Cleaning up what Google saw

Problem

Search Console reported over 21,000 pages not indexed. Most were noise: feed URLs, stray plugin files and page numbers that no longer existed. Separately, 117 of 390 month URLs in the sitemap returned errors.

What we did

Feeds now redirect to their page, out-of-range pages redirect to the last real page, and junk files return Gone. Month pages in the sitemap now follow the same rule the pages use.

Result

All 273 month URLs now in the sitemap return a working page, and Google spends its time on real content.

07

What shipped

Live today

  • Structured data across reviews, with pages for authors, publishers, series and years
  • New review, author, award and archive templates
  • New homepage, site header and footer
  • Release months and an anticipated-books hub
  • Mobile redesign and speed work
  • Indexing clean-up and zero orphaned posts
  • Book marketing services page

Still to come

  • Testing review titles on pages that rank well but earn few clicks
  • Refreshing older posts that still attract search demand
Next case studyBlurbBio

Let's build.

Tell us about your project. We respond within 24 hours with a detailed plan of attack.

Start a Project