A domain rating over 100 doesn't exist on Ahrefs' own scale. What the real 2026 data on expired domains says about redirects, vetting, and inflated metrics.
Start with the number that should stop you before you spend anything: Ahrefs' Domain Rating scale runs from 0 to 100. There's no such thing as a domain with a DR of 250 on the metric everyone means when they say "DR." If you've seen that figure attached to an expired-domain listing, it isn't a typo you can shrug off, it's the first data point in evaluating the seller, not the domain. The rest of the expired-domain market rewards exactly that kind of scrutiny right now, because the easy version of this strategy stopped working a while ago and most of the people still selling it haven't updated their pitch.
What actually changed, and what didn't
Google's spam policy for this is not new. Per Google's own documentation, expired domain abuse is "where an expired domain name is purchased and repurposed primarily to manipulate search rankings by hosting content that provides little to no value to users," and it's been codified since March 2024, alongside site reputation abuse and scaled content abuse. What changed in the March 2026 update wasn't the rule. Google's detection got more precise at enforcing it, rolling out in under 20 hours, the fastest confirmed spam-update rollout on record. If your expired-domain setup was clean before that update, nothing about your risk profile shifted. If it was already borderline, the margin got thinner.
The operative word in Google's own definition is "primarily." The signal Google is actually weighing isn't domain age, and it isn't backlink count. It's thematic coherence, whether what's hosted on the domain now is meaningfully related to what the domain built its authority doing. A decade-old regional legal-resource domain doesn't transfer trust to a crypto-arbitrage blog no matter how well the new content is written. A decade-old gardening site rebuilt as another gardening site, with original content, carries forward a real chunk of its history. Same domain, opposite outcomes, and the difference is topical match, not metrics.
The vetting math nobody selling domains wants to show you
Here's the part that should recalibrate how confident you feel walking into an expired-domain auction. An analysis of 915 expired domains with a Domain Rating of 20 or higher, all acquired on GoDaddy auctions over a 14-day window in early 2025, found that only 3.82% of them ended up actually used in a private blog network. Worth noting where that number comes from: it's published by a PBN hosting company, so treat it as an insider's estimate rather than neutral third-party research, but it's also a number that runs against their own sales interest, since the honest reading of it makes the strategy look harder, not easier. The rest of that pool got flipped back to auction after failing vetting, redirected directly with no rebuild, or picked up by gambling-adjacent operators targeting markets with a different risk tolerance.
That 96%+ failure rate isn't evidence the strategy is dead. It's evidence that the supply of genuinely clean expired domains has been thinning for years, because the good ones got bought early by people who are still using them. What's left on auction platforms skews toward domains that already cycled through a PBN once, domains carrying inflated metrics from link farms pointing at them specifically to look attractive at auction, and domains in niches narrow enough that finding relevant content to rebuild them with erodes whatever cost advantage you thought you were getting. The same source that produced the 3.82% figure is candid that a growing number of experienced operators have quietly shifted toward registering fresh domains and building real topical authority over six to twelve months instead, precisely because the expired-domain vetting math stopped paying off reliably. That's a meaningful admission coming from a company that sells PBN hosting for a living.
Reading "DR250" as a seller signal, not a domain signal
Which brings the inflated-metric problem back around. A domain rating can't exceed 100 on the scale that term normally refers to, so a listing advertising DR250, or any figure past the ceiling, is doing one of three things: describing a different metric and mislabeling it, combining multiple scores into a number that looks like one, or just inventing a figure because inflated numbers move faster in a marketplace where most buyers don't check.
None of those are disqualifying on their own if the seller clarifies what the number actually means when asked. What's disqualifying is a seller who can't or won't clarify it, because that's the same seller unlikely to disclose that the domain spent three years in a PBN network, or that its clean-looking Trust Flow is sitting on top of a backlink profile built specifically to inflate that score before the auction listing went up. Treat any domain metric you can't independently verify with your own tools as marketing copy, not data, until proven otherwise.
The vetting framework that actually holds up
Real evaluation runs on three checks, and no single one is disqualifying alone, they have to be read together.
Wayback Machine history. Pull archived snapshots across the domain's entire active life, not just its most recent version. Topical consistency across years is a genuine positive signal. Abrupt subject pivots, multi-year content gaps, or any visible prior life as a link farm are red flags regardless of what the current backlink metrics show.
Referring domain topicality. Don't just check the authority score of sites linking in, check what those sites are actually about. Referring domains clustered in adjacent, coherent verticals suggest a defensible topical identity. Referring domains scattered across unrelated niches suggest manufactured authority, the exact pattern detection systems are built to discount.
Anchor text distribution. A backlink profile heavy on exact-match or money-keyword anchors was built for manipulation, not earned through organic mentions. Natural anchor text, brand names, bare URLs, generic phrases like "this article" or "read more," indicates links that accumulated the way real editorial links do.
If all three line up with your intended use for the domain, it's a reasonable acquisition. If any one of them contradicts it, you're buying a liability with a price tag, not an asset.
What still works with redirects, and where they break
A 301 redirect from an expired domain to an existing site is still one of the more defensible uses of this strategy, but the same topical-alignment rule governs it directly. If the old domain's history and the destination site are in the same niche, the redirect is easy to justify and Google treats it as a normal consolidation. If the domain is from an unrelated niche and the redirect exists purely to pass authority, that's the specific pattern named in Google's own expired-domain-abuse examples, and in practice it can get ignored entirely or, per real-world reports from SEOs working this space, actively hurt the destination site's rankings.
The other operational risk is concentration. Pointing more than two or three expired domains at the same destination URL creates a footprint that's easy to flag as manufactured, even when each individual domain would pass vetting on its own. Spreading acquired authority across multiple relevant pages on your site, rather than funneling everything at one URL, reduces that signature.
What this actually means for a Tuesday
If you're evaluating an expired domain this week, the checklist is short: verify every metric independently rather than trusting the listing, run the Wayback history check before you check the price, confirm referring-domain topicality and anchor profile match your intended use, and if you're doing a redirect, stay inside the same niche and don't stack more than two or three domains on one target. None of that is complicated. It's also not what most auction listings are optimized to survive, which is exactly why the pass rate on real vetting sits under 4%.
The broader shift underneath all of this is the same one showing up across other parts of link building right now: authority that's inherited from a domain's past and authority that's earned through genuine, verifiable, topically consistent activity are being scored differently by systems that have gotten better at telling them apart. A backlink or a citation that comes from a real, current, topically matched source doesn't carry any of this risk, because there's no history to misrepresent and no metric to inflate. That's a slower way to build authority than buying a domain with a big number attached to it. It's also the version that doesn't require you to bet on whether this week's DR250 listing means what the seller says it means.
