Refresh or Rewrite? Diagnose the Stage That Failed

A page that stopped showing up in AI answers did not expire. It is failing at one identifiable stage of how the answer gets built, and the stage decides whether you repair the page or replace it. Four stages can fail: the page is never fetched, it is fetched but never selected as evidence, it is selected but the answer states something wrong about you, or the question buyers now ask is not the question the page answers, and only that last one calls for a rewrite. The first three are repairs, and repairs are cheaper and easier to measure.

The stakes are the shortlist. 51% of software buyers now begin research inside an AI chatbot, the measured figure behind G2's account of chatbots as the single largest influence on B2B shortlists. Buyers also arrive knowing most of their options, since 6sense found buyers already know 3.8 of the roughly 5 vendors they evaluate before contact. Guessing wrong is expensive in both directions, and the top-ranking page now sees a 58% lower average clickthrough rate when an AI answer sits above it.

Why did the page stop showing up in AI answers?

Because one of four stages failed, and each fails independently of the others. An answer is assembled by fetching pages, selecting passages from what was fetched, composing claims out of those passages, and matching the result to the question asked. Our walkthrough of how AI reads your content covers the machinery; what follows is how each failure looks from the outside.

Stage 1: the page is never fetched

The symptom is absence from your own server logs: no AI user agent requests the URL, or one requests it and receives a shell whose content only appears after JavaScript runs. The crawler study Vercel ran with MERJ found that OpenAI's and Anthropic's crawlers fetch raw HTML and do not render JavaScript, while Google's AI features run through the same Web Rendering Service capabilities as Search. Google is explicit that pages must be indexed and snippet-eligible to appear in its AI features, and OpenAI publishes four relevant user agents you can check your logs and robots rules against. Cloudflare measured agent readiness across the 200,000 most visited domains, which is the scale at which this failure now gets tracked.

Stage 2: fetched, but never selected

The symptom is presence without participation: your URL is fetchable and indexed, and competitors' passages get quoted instead. Seer found 87% of 500 SearchGPT citations matched Bing's top results, while Ahrefs found 38% of AI Overview citations rank in the organic top 10 across 4 million AI Overview URLs in March 2026, down from roughly 76% in July 2025. Selection rewards extractable evidence: the GEO study published at KDD 2024 found statistics, quotations and citations lifted visibility in its benchmark, and an AirOps analysis of 548,534 pages mapped which traits correlate with being pulled in.

Stage 3: selected, but the answer gets you wrong

Your page can be cited in an answer that still describes you inaccurately, because the claim is assembled from several sources at once. Profound found 57% of AI citations point to sources brands do not control, so a stale directory entry or an old review can outweigh your own copy. Ambiguous identity compounds it, and entity-oriented retrieval research across 443 configurations shows how much retrieval quality depends on resolving who you are.

Stage 4: the question moved

The page is fetched, selected and accurate, and still loses, because buyers stopped asking what it answers. A page built around a workflow nobody runs now cannot be repaired into relevance. This is the only stage where a rewrite is the correct instrument.

How do you tell which stage failed?

Re-run a fixed set of buyer questions and watch where the page drops out. Keep the set stable across quarters, because a moving question set cannot tell you whether the page changed or the market did. We use ten runs per question as a working default, spread across several days and two engines, recording who was cited and which URLs carried each answer. The right number for you depends on the run-to-run variance you observe in your own sample.

Single runs prove nothing. A 2026 variance-components study found run-to-run noise large enough to swamp real differences in small samples, and SparkToro's research reports AI engines highly inconsistent when recommending brands. Search Engine Land covered a study finding AI recommendation lists rarely repeat, and work on quantifying uncertainty in AI visibility makes the same point with confidence intervals.

Then read the record for a fingerprint.

• Your URL never appears, and a JavaScript-disabled fetch returns an empty shell: stage one.

• Your URL is fetchable, and competitors' passages get quoted instead: stage two.

• Your page is cited, and the surrounding sentence is wrong: stage three.

• The answers are reasonable and address a different problem than your page does: stage four.

Which repair matches which stage?

Match the intervention to the stage, and stop there. Stage one is mechanical: server-render the content an engine needs, and check your robots rules and indexability against the documented user agents. Cloudflare recorded AI crawler share of requests rising from 2.2% to 7.7%, so log evidence of these fetches sits in your own server records.

Stage two is an evidence edit, not a new page. Put the number, the named source and the direct answer into the section that already covers the question, and keep a heading structure an engine can segment. C-SEO Bench later tested these same rewrite methods and found only 3 of 54 unilateral conditions produced statistically significant gains, so treat this edit as making an existing answer easier to extract, and expect selection itself to move for other reasons.

Google's structured data documentation describes what markup does and does not do, and its people-first content guidance still describes the substance being selected for. Watch the persuasion dial while you edit, because scarcity and exclusivity framing measurably reduces how often a model recommends a product.

Stage three is a claims repair that mostly happens off your own domain. Correct the fact on your page, then correct it where engines are reading it: the directory listing and the comparison article carrying the outdated description. Keep substantiation for every claim, which is the standard the FTC's advertising guidance already sets.

When is a rewrite actually the right call?

A rewrite is right when the question itself changed. The signal is in the answer text: engines are addressing a problem statement your page does not contain, using vocabulary your page never uses, and citing pages that frame the topic differently. That is a different article wearing the same URL.

Two secondary cases qualify. A page built for a product you no longer sell has nothing left to repair. A page whose central argument you now believe is wrong needs replacing, because a corrected sentence inside a flawed frame produces an incoherent source. Everything else is a repair carrying a rewrite's price tag.

A rewrite also resets whatever history the page had accumulated. You are trading a known position for an unknown one, which is defensible when the known position is worthless and reckless when it is merely diminished.

Is a superseded number a reason to rewrite the page?

No, and this is the most common real case. A page carrying one stale figure inside otherwise sound structure fails at stage two or three, and the repair is scoped to the figure and its citation. Replace the number and its link, date the claim in the sentence, and leave the URL and headings alone.

Superseded figures are ordinary. Ahrefs first measured a 34.5% lower clickthrough rate for top-ranking pages on informational keywords across 300,000 keywords, and its December 2025 re-run superseded that with 58%. Any page citing the earlier figure needs one paragraph edited. Treating that as a rewrite is how a content calendar fills with work that changes nothing, which is the failure mode described in a visibility gap is not a content brief.

Does publishing something new help on its own?

There is no reliable public measurement showing that recency by itself decides which pages AI engines cite, so treat freshness as a hypothesis to test inside your own category. Changing a publish date repairs none of the four stages. If you believe recency matters for your buyers, measure it: hold the question set fixed, change one variable, and compare appearance rates across enough runs to clear the variance documented above.

The evidence on rewriting for engines is unflattering to tactics generally. C-SEO Bench, published at NeurIPS 2025, tested conversational-SEO rewrite methods and found only 3 of 54 unilateral conditions produced statistically significant gains, and the benchmark's authors tested those methods across two tasks and six domains. Phrasing edits are a weak instrument, whatever a refresh checklist implies.

Structure the measurement so the result stays legible. The IAB's framework separates presence, prominence, portrayal and persuasion, and those four move for different reasons. A repair that improves portrayal while presence stays flat is still a win, and a blended score would have hidden it.

How does Trovance decide refresh versus rewrite?

By reading which of the four stages the preserved record says failed, and proposing a rewrite only for stage four. Trovance holds the question set still so that diagnosis stays honest. You define the market questions your buyers ask, and the platform runs them on a repeating analysis cycle, preserving each answer run as a snapshot with its full citation set. Because the same questions are asked over time, a page dropping out of answers shows up as a change in the record.

Each stage leaves a different fingerprint, and the platform compares those records against each other. Answer coverage shows which tracked questions your pages participate in at all. Answer snapshots preserve which URLs carried each answer and what the engine said about you while citing them, which is how a stage-three claims problem separates from a stage-two evidence problem.

The output is a recommended action with a stage attached. Your Brand Core holds the claims you are entitled to make and the proof behind each, so a repair names the figure or source the evidence record says is missing, and a rewrite is proposed only when the question itself has moved. Drafts come from approved claims, and a person reviews everything before it publishes. The next cycle reruns the same questions to verify the answer moved.

Trovance will not promise that a repaired page gets cited, ranked or recommended. Engines are probabilistic, and most of the sources that decide an answer sit outside your domain, so no vendor controls what a model says. What the system does is preserve the evidence and name the stage that failed. The next cycle shows you whether the change you shipped altered the answers you were already tracking.

What should you do this week?

Pick the ten pages you would defend in a sales conversation and diagnose those only. Re-run your buyer questions on the schedule you set, record the URLs the answers cite, and sort the pages into the four stages. Most will land in stages one through three, which means most teams will spend an afternoon on repairs.

Then set the standard for done. A repair is finished when the same question set produces a measurably different answer pattern over enough runs to clear normal variance, never when the document simply looks better. Anything you cannot re-measure is a preference. Semrush's index covers 126 million AI search prompts, which is the scale your fixed question set is deliberately not trying to represent.

If you would rather have the question set run for you and the failed stage named against preserved evidence, start a free Trovance analysis and see which of your pages needs a repair and which needs replacing.

Decide what to fix

• A visibility gap is not a content brief - why a missing answer does not automatically justify a new page.

• Why AI citations drop - the causes behind a page losing its place in answers.

• A relevant page can still miss the proof - the stage-two evidence gap in detail.

• How to repair a wrong AI answer about your brand - the stage-three claims repair, mostly off your domain.

Check the mechanics

• How AI reads your content - the stages an answer passes through before it names anyone.

• AI crawlers don't run your JavaScript - the fetch failure that makes every later repair pointless.

• Do headings help AI retrieve long documents - what structure actually contributes to selection.

• How long until AI cites you - realistic timing before a repair shows up in answers.

FAQs