agentlinkopsHow to qualify listicle and review inclusions
Qualify a best-of list or review roundup before asking for inclusion: freshness, editorial versus paid entries, incumbents and what an editor needs to see.
A listicle or review inclusion is an entry in a page that ranks or compares several products, services or resources for one audience. The qualification job is to check that the list is still maintained, learn whether entries are chosen by an editor or sold, see which competitors already appear and on what basis, and work out what evidence would let the editor add you without lowering the page.
This guide is one of the discovery use cases. A roundup is different from a paragraph inside an article, which the link insertion review covers, and from a curated resource list, which the resource page checklist covers. The fields below are the ones a ranked list needs and the others do not: freshness, inclusion basis, incumbents and disclosure.
Editorial inclusion and paid inclusion are different products
Google's review guidance asks reviewers to "provide evidence such as visuals, audio, or other links of your own experience with what you are reviewing", and to make ranked lists strong enough to stand on their own. A list built that way adds an entry when the editor has tried the product. A list built the other way adds an entry when someone pays, or when a free product arrives.
Both of those are disclosures under the FTC's endorsement guidance: a connection "that a significant minority of consumers wouldn't expect and it would affect how they evaluate the endorsement" must be disclosed "clearly and conspicuously". Google's spam policies list "sending someone a product in exchange for them writing about it and including a link" as link spam unless the link is qualified, and its outbound-link guidance says paid placements should carry rel="sponsored".
So the first question for any roundup is the basis of its entries. Look for a sponsorship disclosure, a "featured" or "partner" label, an affiliate notice, and the rel tokens on the existing entry links. A paid list is an advertising channel. That can be a fine decision for the owner to make; the record just has to say it.
Find candidates with the files you already have
A competitor export, filtered to list pages. A supplier backlink export of two or three competitors will contain the roundups that cite them. AgentLinkOps imports seven supplier layouts and a generic CSV through presets, and anything else through an explicit column map. Filter on page titles and paths that look like lists (best, top, vs, alternatives, review) and keep the page URL.
A frozen-inventory comparison. If you already compared competitor inventories, the gap report tells you which list pages appear in a competitor's dataset and not in yours. Read that state literally. A gap is not_found_in_customer_dataset; the report says absence_claim: not_supported and whole_web_coverage: false. As the documentation puts it, "An inventory gap means the source did not return a row; it does not prove a link is absent." It also does not prove the list will take you.
An unlinked mention. A roundup that names your product without linking to it is the strongest listicle clue there is. The unlinked mention review covers the identity checks; the numbers below show why the "without linking" half has to be observed.
Your agent's own search. Queries for best [category] and [competitor] alternatives find the lists that rank today. Save the result page and the date. Ranking today says nothing about whether the list is maintained.
Qualify with four fields the other jobs do not need
| Field | What to record | Who supplies it |
|---|---|---|
| Freshness | The date the page states, and your own observation date, kept separate | Your agent reads, AgentLinkOps stamps the fetch |
| Inclusion basis | Editorial, affiliate, sponsored or unclear, with the disclosure text or the rel tokens that show it | Your agent reads the observation, you classify |
| Incumbents | Which competitors appear, in what position, and whether their entries are qualified | Your agent reads |
| Evidence the editor needs | What a reviewer could try: a free result, a public demo, a documented method | You |
The product's campaign templates return three verdicts and no score: satisfied, not_satisfied and agent_judgement. Every row in that table except the stamped dates is judgement. The software attaches the evidence and withholds the verdict, and that is the intended division.
Freshness deserves the separate date columns. A list page usually states a date, and a page can claim "updated" without changing. Your observation time is yours; the page's date is the publisher's claim. Keep both.
The contact rule. A contact counts only when the publisher explicitly published it. The hosted contact tool reads addresses and contact-page URLs from the latest saved copy of the page and never checks or implies deliverability. A reviewer's byline with a social handle is a lead for your agent, not a route.
What the mention scan found on 140 real pages
On September 11, 2026 the mention module re-fetched 140 pages from an owned corpus through the verifier and scanned them for four targets, each with a brand string. Seven pages mentioned a target. Four of those seven already linked to it. Three were reclaimable.
Two limits from that run apply directly to listicle work. The bare domain found nothing on its own; every hit came from a brand string. And a single-word brand is refused unless you force it, because a brand that is also an ordinary word matches every sentence that uses the word. The scan skips text inside script, style, template and comments, so a brand that only appears in a page's tracking code is not a mention anyone read.
The four-of-seven number is the reason the "without linking" half must be observed rather than assumed. A tool with no mentioned_and_linked state would have reported seven opportunities and been wrong about most of them, and the fastest way to lose an editor is to ask for a link they already gave you.
What your agent does and what you decide
Agent work, with your tools: read the list completely; record the stated date and the observation date; classify each entry's link by its rel tokens; find the disclosure text; list the incumbents; find the published contact route; draft the case for inclusion, with the evidence a reviewer could test, for your review.
Your decisions: whether a paid list is worth paying for; whether the audience fits; whether your product would hold its place beside the incumbents; whether the evidence you offer is honest; and whether the request goes out. The local next-action vocabulary cannot express a send. The publisher relevance guide covers audience fit; the outreach handoff shows the reviewed draft.
How AgentLinkOps records the result
If the editor adds an entry, register the list page and your destination as a source-and-target pair. The candidate starts not_checked. One explicit selection creates a paused watch and a single metered check. The observation stores a hash of the fetched bytes, the robots posture, the redirect chain and the HTTP status, with the standing note that JavaScript execution and visual visibility were not checked. The recall, verify and monitor pipeline carries a whole shortlist through that check in one bounded batch and converts a present result into a watch that keeps its candidate lineage.
The rel tokens in that observation are the record of the inclusion basis. If the entry carries sponsored, your record says the placement is an advertisement, whatever the page's wording. If a later check shows the tokens changed, the change is in the history. Read the check placement guide for the sequence.
What can go wrong
The list loads its entries with JavaScript. Comparison tables are often rendered client-side. The check returns unknown with possible_render_required; unknown never fails a run and never means anything was removed. Your agent's browser is the next step, and its observation needs a date.
The page is behind a challenge or a login. A 403, a bot challenge or a paywall is unknown with a named reason. It is not evidence about the list.
A competitor's entry is treated as your eligibility. The competitor campaign template names this rejection: "competitor link is assumed to guarantee our eligibility". An incumbent shows what the editor accepted once, on their evidence, under their terms.
The list is abandoned. A stated date from two years ago, dead entry links and no author is a list nobody maintains. An inclusion there is a placement on a page no one updates. Record it and move on, or ask the owner.
Concealed affiliate terms. If entries are affiliate links without disclosure, the list is already outside the guidance it would need to follow to add you honestly.
Where this sits in the product today
Imports, selected candidate checks and placement monitoring run in the invitation-only pilot. Comparing inventories you already hold works; requesting a new paid competitor inventory is disabled in this build. Live supplier discovery stops with PROVIDER_NOT_CONFIGURED. The mention scan and campaign qualification are local research modules; check the capability status before choosing a workflow. AgentLinkOps sends nothing.
The niche directory guide covers the other kind of list, where you submit a listing yourself, and the opportunity library's agent-tooling directories niche shows what such a list looks like before any route has been read: no records, a disclosure and a take-down route.
Sources
- Write high quality reviews · Google Search Central · 2025-12-10
- FTC's Endorsement Guides: What People Are Asking · Federal Trade Commission · 2026-09-15
- Spam policies for Google web search · Google Search Central · 2026-08-28
- Qualify your outbound links to Google · Google Search Central · 2025-12-10
- From publisher research to a campaign brief · AgentLinkOps documentation · 2026-09-15