Bing Indexing: The Inbound Links Warning

What Bing's inbound links warning actually means

Bing's recommendation that a site lacks inbound links from high-quality domains describes an indexing decision, and most readers treat it as ranking advice. Bing's own interface copy is explicit that link signals feed the choice of which pages enter the index in the first place.

The string that says so is in Bing Webmaster Tools itself. Explaining a site that has dropped in ranking, Bing writes that pages may have not accrued sufficient ranking signals (such as links, traffic) to be selected into the index. It adds that it weighs those signals relative to all other sites on the web as part of our selection criteria.

Two words in that sentence carry the weight. Selected places links before ranking, at the gate rather than in the ordering. Relative means the comparison runs against all other sites on the web rather than against a published figure, so the bar moves as the web does.

Bing closes the passage with a prediction rather than an instruction: as the popularity of a site's pages increases, more of those pages will be indexed. That is the mechanism the recommendation is pointing at, stated by Bing in its own product.

What the recommendation does not tell you

The recommendation is graded Moderate, reports one total error and names no affected pages. It is a site-level observation with no page list attached, so nothing in it identifies which pages are missing or how many referring domains would clear it.

Whether Bing raises it on a threshold or shows it broadly is not documented. The wording goes back to at least March 2025, when a site owner quoted it on Microsoft's own question site and asked exactly the question this article answers: whether it explained why an indexed site was not appearing. Treat the recommendation as a prompt to measure rather than as a measurement.

Why Bing indexes fewer of your pages than Google does

Bing applies a selection step that it describes in comparative terms, and Google's public guidance on indexing does not use the same framing. A site can therefore clear Google's bar comfortably and sit below Bing's, which is the usual shape of a Google-yes, Bing-no gap on a newer domain. Google's own equivalent verdicts are covered in how to check when Google indexed a page.

Bing Webmaster Tools lists six reasons a site is not being crawled or indexed. The list is the tool's own in-product help text, which cross-references Bing's why is my site not in the index (opens in a new tab) article for the longer version. Five are mechanical and one is not.

1. The site is new and Bing has not discovered and crawled it yet.
2. Bing is running into issues crawling the site.
3. The robots.txt file is preventing Bing from crawling the site.
4. Pages carry a robots meta tag set to NOINDEX.
5. URLs have been blocked inside Bing Webmaster Tools.
6. The site does not meet the quality threshold required by Bing.

The first five are answerable in an afternoon and are the ones every guide covers. The sixth is the one the inbound links recommendation is about, and Bing attaches no number to it anywhere in the interface.

Reason one and reason six compound on new sites, which is why a recently launched domain sees the widest Bing and Google gap. A site with no history has neither the links nor the impressions Bing weighs, and Bing's own framing says index coverage grows as those signals do rather than on a fixed timetable.

Bing's index coverage report says the same thing

Under Sitemaps, Bing groups unindexed URLs into named classes. Two of them, Crawled but not in index and Discovered but not in index, carry identical remedies in Bing's copy: consult the Webmaster Guidelines to help you boost ranking signals.

A third class, Content quality, tells the reader to improve content quality and signal quality for better index coverage, then wait a few days for popularity signals to be assessed. Bing uses one vocabulary across the recommendation, the not-indexed classes and the ranking-drop explanation, which is the reason to read them together.

Bing impressions are not a list of your indexed pages

A Search Performance report lists pages that earned an impression, so it cannot answer what Bing has indexed. An indexed page nobody searched for produces the same empty row as a page Bing never took, and the two are indistinguishable in that report.

Three places in Bing Webmaster Tools do answer the question: the Index Coverage report under Sitemaps, URL Inspection for a single address, and Site Explorer for a tree view. None of the three is exposed through the Bing Webmaster Tools API, so an index count has to be read in the interface by hand.

The same caution applies to the site: operator. A site: search on Bing returned a generic result count with no matching URLs in the response when fetched by script, which makes it unusable as evidence either way.

Fix the cheap causes before you chase links

Link building is the slowest available fix and the recommendation makes it sound like the only one. The checks below cost an hour between them, and each one rules out a cause that has nothing to do with links.

Check what the old URLs do now

A migrated site keeps its predecessor's addresses in Bing's index longer than in Google's. Pull Bing's page report, take every URL that does not match a current route, and request each one.

Trailing slashes hide this. On a framework that normalises /page/ to /page, a legacy URL answers 308 and looks handled, then resolves to a 404 one hop later. A 308 that lands on a 404 is worse than a bare 404, because it spends the redirect and consolidates nothing.

Read legacy slugs from the engine's own record rather than from memory of the old site. Google can hold a variant of the remembered slug, such as /data-processing-agreement-dpa for /data-processing-agreement, and a redirect written for the wrong one never matches.

Then check discovery, then check access

• Submit the XML sitemap inside Bing Webmaster Tools. A sitemap declared in robots.txt is discovered eventually; a submitted one gets the Index Coverage report, which is the only place Bing names its reasons per URL.
• Turn on IndexNow (opens in a new tab). It tells Bing a URL changed within seconds. Its own FAQ lists Amazon, Bing, Naver, Seznam.cz, Yandex and Yep as the participating endpoints, and Google is absent, so it is a Bing-side fix specifically.
• Confirm robots.txt allows Bingbot, including any named Bingbot block that overrides the wildcard, since a user-agent-specific group replaces rather than extends the * group.
• Run Bing's own URL Inspection in live mode on one page you expect to be indexed and are not sure about. It reports Bing's verdict rather than inferring one from traffic.

Submit URLs to Bing directly, and read the quota before you plan around it

URL Submission in Bing Webmaster Tools pushes addresses straight into the queue rather than waiting for Bingbot to crawl its way to them. Bing raised the allowance to up to 10,000 URLs a day with no monthly cap, announced on 31 January 2019, and the feature has carried a per-site quota ever since. Microsoft added a note to that same announcement in November 2025 recommending IndexNow as the primary method for real-time submission, so treat URL Submission as the manual fallback.

That figure is a ceiling and not an entitlement, which the same announcement states plainly. Bing said the daily quota per site would be determined by the site's verified age in Bing Webmaster Tools, site impressions and other signals available to Bing, and that verified age is one signal among several rather than the deciding one.

The consequence is worth sitting with. A new site with few impressions is given a smaller submission allowance, so the weak signals that keep pages out of the index also limit how hard you can push against that. Read the quota the tool reports rather than planning around the headline number.

None of these earns a single link, and none of them is what the recommendation asked for. They remove the reasons Bing would decline a page for something other than its signals, which is what makes the remaining gap interpretable.

What earns inbound links from high-quality domains

Bing's guidance treats manipulation of inbound links as a violation rather than a shortcut. Buying links, link schemes, link farms and private link networks are named as things that count against a site, which leaves the slow route as the only durable one.

Bing frames the comparison against all other sites on the web, which gives nothing concrete to aim at. A workable substitute is the referring domain profile of the sites already ranking for the queries you want, and that substitution is ours rather than Bing's. The slower route of earning those links through coverage depth is set out in topical authority without backlinks.

Bing weighs traffic alongside links, which makes the two awkward to separate in practice: a page nobody visits and nobody links to is short of both signals at once. That is an argument for publishing something worth visiting rather than for running link acquisition as a separate exercise.

One practical note on reading Bing's own guidance

Bing's help articles are rendered in the browser rather than served as HTML. Fetching the Webmaster Guidelines and the not-in-the-index article on 18 September 2026 returned byte-identical documents, the same 125,562 bytes with the same checksum, because both URLs serve one application shell.

The practical consequence is that scripted checks, archive captures and any crawler that does not execute JavaScript see none of the rendered guidance. Read those pages in a browser rather than expecting a command-line fetch to confirm a quotation from them.

What the shell does carry is the interface's own string bundle, so the passages quoted here are reproducible without a browser. Request that same URL and search the response for as part of our selection criteria, and Bing's selection sentence is there; search it for quality threshold required by Bing and the sixth reason is there twice.

One warning if you try it. Several strings in the bundle separate words with narrow no-break spaces rather than ordinary ones, so a longer phrase fails to match while a shorter one succeeds. Searching for selected into the index returns nothing on a document that plainly contains it, which reads as the quotation being wrong when it is the whitespace that differs.

Why Bing's index decides more than Bing's search results

Bing's index is the retrieval layer behind Microsoft's grounding, so index membership now gates more surfaces than Bing's own result pages. Microsoft describes grounding as the system connecting AI to current information, and its February 2026 statement on grounding (opens in a new tab) says that Microsoft grounding powers nearly every major AI assistant in the market.

That second claim is Microsoft's own characterisation of its reach, no one outside Microsoft can check it, and at least two major assistants sit outside it. The verifiable part is narrower and still substantial: Bing Webmaster Tools reports citations across Microsoft Copilot, AI-generated summaries in Bing, and what Microsoft calls select partner integrations.

That reporting arrived as the AI Performance report in public preview on 10 February 2026, and gained Intents, Topics, Citation Share and Compare on 16 June 2026. Its guidance for improving citation rates addresses pages that are indexed and less frequently cited, which places indexing before citation in Microsoft's own framing.

Being absent from Bing's index therefore costs more than a share of search traffic. Whether index position transfers into assistant citation at any predictable rate has not been measured here, and nothing in this article should be read as claiming it does.

Do not extend the argument to every assistant. ChatGPT and Perplexity operate their own crawlers and their own indexes, with inclusion decisions that report into neither Bing Webmaster Tools nor Search Console, and those are covered separately in how AI crawlers read your site. Bing's index earns its own check because Microsoft's surfaces read from it.

How RankX AI tracks this

RankX AI connects Bing Webmaster Tools alongside Google Search Console and syncs both nightly. The Bing side reports on a weekly grain because that is the grain Microsoft's API publishes, and the two engines' figures sit in one view, so a gap between Bing and Google is visible without opening two consoles.