Your archive still ranks. Search Console shows impressions holding steady on 800 old posts — and clicks down 40 to 70 percent since the AI Overviews rollout finished eating the top of the SERP. The traffic didn’t move to a competitor; it got answered above your link. Meanwhile ChatGPT, Perplexity, and Google’s AI mode are citing three sites in your category and yours isn’t one of them, because your 2019 posts are formatted for a crawler that reads whole pages, not a retriever that pulls quotable chunks. You can’t fix 800 URLs by hand, and rewriting all of them is a five-figure decision you can’t justify without knowing which ones make money.
This is for business owners and operators who already own a content archive — roughly 200 to 5,000 indexed URLs — and have a revenue number attached to at least some of them. You should be comfortable in Search Console and GA4, able to export a CSV, and willing to run a site crawler. You do not need to write code. Out of scope: building an archive from zero, technical SEO fundamentals like site speed and redirects, link building, paid acquisition, and anything requiring a developer sprint. This assumes WordPress or a comparable CMS where you control page templates and can add markup.
Honest assessment: AI is genuinely good at the middle of this job — restructuring a rambling 2018 post into retrievable passages, generating structured data, drafting variations at volume, and summarizing crawl output. It is bad at the parts that decide whether this works. It hallucinates statistics into your refreshed pages, invents citations, flattens your voice into the same beige prose every competitor is publishing, and confidently rewrites pages that were converting fine. Human review is non-negotiable on three things: every factual claim and number that goes live, every page carrying revenue or legal exposure, and the go/no-go call on whether a URL gets touched at all. The failure mode here is not a bad draft — it is publishing 100 pages of plausible-sounding wrong information under your own domain and getting quality-suppressed for it.
What This Guide Covers
- A clear read on what actually changed in 2025-2026 — so you stop diagnosing an answer-engine problem as a ranking problem and wasting budget on the wrong fix
- How retrieval and chunking really work inside answer engines, in plain business terms, so your refresh decisions are grounded in mechanics rather than guesswork
- An audit method for isolating the URLs that kept impressions but lost clicks — the exact pages where the traffic is recoverable
- How to measure your AI visibility today, including which paid tools are worth it and which free alternatives get you 80 percent of the signal
- A revenue-per-citation scoring model that ranks your archive by money instead of sessions, so you refresh the 40 pages that matter and skip the 760 that don’t
- A scaled crawling setup that pulls the inputs you need for a refresh in one pass, instead of opening pages one at a time
- An assembly-line rewrite process built around producing quotable, citable passages — the format answer engines actually lift
- The structured data and machine-readability layer that makes your pages easy for retrieval systems to parse and attribute
- Real cost math on 100 refreshed URLs, with the break-even point and the conditions under which this project does not pay for itself
- How to convert citation traffic into revenue, and why it tends to convert several times better than the organic traffic you lost
- The specific failure modes that get refreshed sites suppressed rather than surfaced, and the review gates that prevent them
- Before-and-after citation share from three real archives, including what didn’t work and the timeline to visible movement
- How to package this as a productized service and price it as a recurring monthly retainer, with scope boundaries that keep it profitable
- A measurement dashboard and automation plan so visibility is tracked continuously instead of rediscovered next time traffic drops
Delivered as instant online access the moment checkout completes — no waiting on email, no upsell sequence, no upgrade tier holding back the useful parts. You buy it, you read the whole thing.
Reviews
There are no reviews yet.