Mapping 10,000+ keywords to lesson pages at Study.com

01 The role

Study.com publishes lesson content at scale across dozens of subject areas. That scale creates a specific problem: with a very large library and an even larger pool of keywords, terms drift onto pages that cannot satisfy them. A page targeting the wrong query does not simply fail to rank — it competes with the page that should have had it.

I joined on the link building side and moved into keyword and content operations, working mostly across the maths and test-prep parts of the library.

• First — Link Building Associate

  Prospecting and vetting

  Prospected sites by building relevant keyword sets in Ahrefs, vetted candidates against client criteria in Buzzstream, sourced contacts for every approved site, and reviewed peers’ vetting for consistency.

• Then — SEO Specialist

  Keyword operations, content, and audit

  Keyword research and mapping at volume, page audits, interlinking, on-page optimization, content production from briefs, video scripting, metadata compliance auditing, and peer review across the contractor team.

02 Keyword mapping and information architecture

The core of the work was deciding, keyword by keyword, whether an assignment was correct and whether the term was worth targeting at all. Across the engagement that ran to more than 10,000 keywords.

One audit batch gives a sense of the shape of it: 1,489 keywords across 93 lesson pages, ranging from head terms down to long-tail queries. Of those, 266 were flagged as priority targets and 81 as incorrectly mapped. Each rejection carried a written reason, so the decision could be reviewed rather than taken on trust.

The rejections fell into recognisable categories. Naming them mattered more than counting them, because a category is something you can check the next batch against.

From mapping to architecture

The mapping fed a broader architecture question. Alongside it I matched subject nodes to the lesson that best fitted them, or confirmed that no matching page existed at all. That second outcome is the useful one: a documented gap is a content brief waiting to be written, and collectively those gaps shaped the information architecture of the wider library.

The surrounding on-page work followed from the same decisions — primary keyword identification and optimization, retitling lessons to best practice, close match audits across existing pages, interlinking related lessons, and concept matching to find openings for new ones. I also ran relevancy scoring tests using ChatGPT to sanity-check keyword-to-page fit at volume, as a second opinion rather than a decision-maker.

03 Content and scriptwriting

I produced keyword-optimized lesson summaries from structured briefs — an outline, a transcript, and a set of target keywords — across subject areas including earth science, biology, psychology, and social science. The brief allowed editorial latitude to merge redundant sections, resequence content, and use keyword variants where an exact match would not read naturally. In practice the work was closer to editing than transcription.

I also wrote and revised video scripts for test-prep content. These were built around exam topic breakdowns, cited primary sources from the testing body, linked internally to the relevant course pages, and closed with calls to action — content designed to sit inside a funnel rather than stand alone.

04 Metadata and credential compliance audit

A separate project involved auditing published video metadata against the videos themselves, checking that the expert credentials stated in each description matched what was actually claimed on screen and in the audio. Across 100+ published videos, that meant reviewing each one against its description, flagging inaccurate credential claims for correction, and documenting discrepancies between description text and on-screen overlays separately, since the two were fixed by different people.

Production faults surfaced along the way — unclear audio, missing branded elements, uncropped framing — and were tracked through to resolution rather than handed off as a list.

This is author credential compliance, which is the practical side of E-E-A-T. Getting expertise signals to match reality across a large content library is now a mainstream SEO concern. This project ran before it became one.

05 Process and peer review

I documented the content production workflow end to end — from reading the transcript and identifying mergeable sections through keyword incorporation, drafting, and final quality checks — and used it as the basis for training material. I also peer-reviewed other contractors’ work against that same standard, first on link vetting and later on SEO output.

Client URLs, keyword lists, and per-page volume data are omitted as client data — happy to walk through the methodology in an interview or under NDA.

06 What this means for your site

Most of this transfers directly. If any of the following sounds like your situation, it is the same problem in different clothes.

Got a keyword map nobody trusts?

Most large sites have one. Tell me what you need and I will come back with what I would check first.

See what I do →

← Read the technical SEO case study