The 2026 guide to buying guest posts that actually rank

About 70% of guest post placements sold online today are technically real links that won't move your rankings. They're on real domains, they're indexed, they don't violate any rule a Google representative has ever stated publicly. They just don't work, because the publisher selling them has been doing the same thing for so long that the page's trust signal has decayed to zero.

The other 30% are the ones you want. Telling them apart is the job. This is the checklist we run before any publisher gets approved on Bazsy.

Check 1: Organic traffic, not just traffic

The first filter is non-negotiable. The publisher must show measurable organic search traffic, ideally above 2,000 monthly visits per a tool like Ahrefs or Semrush. Sites with 50,000 "traffic" that turns out to be branded or direct or referral don't count. Branded traffic means the audience already knows the domain — those visitors don't transfer trust to your link. Organic traffic means Google sends real people there for real queries, which is what the algorithm is rewarding when you place a link on the same domain.

We've seen publishers with DR 65 and 80 monthly organic sessions. They got to DR 65 by buying their own backlinks. The page where your link sits would rank for nothing.

Check 2: Traffic-to-DR ratio

This one catches the inflated DR sites. Take their organic traffic, divide by their DR. A healthy ratio is somewhere above 30 — meaning a DR 50 site should have at least 1,500 monthly organic visits. Below 10 is a warning. Below 5 means the DR is almost certainly purchased and the link will not transfer authority.

The ratio isn't a hard rule. Newsrooms with high DR and traffic that fluctuates with news cycles can look bad in a quiet month. But across the marketplace, the correlation between this ratio and downstream ranking impact is the strongest single predictor we've found.

Check 3: Anchor pattern of their last 50 outbound links

Open Ahrefs or Semrush. Look at the publisher's outbound links across recent placements. If 40 of the 50 last anchors are commercial money terms ("best CRM software", "online casino bonuses", "personal injury lawyer"), the publisher is operating a placement farm. Google's link evaluators have been classifying outbound-link patterns since 2018, and a pattern that obvious produces a domain-level discount.

A clean publisher's outbound anchor mix looks like editorial work: brand names, naked URLs, generic phrases, occasional partial-match terms, rare exact-match. The ratio matters more than any individual anchor.

Check 4: Domain age and registration history

Domain age alone is overrated. What matters is registration continuity. A domain registered in 2008, dropped in 2019, and re-registered in 2023 is a 17-year-old domain on Whois and a two-year-old domain to Google. The Wayback Machine catches these every time. We won't approve a publisher whose archived content type changed dramatically between registrations.

Check 5: Indexation health

Take five recent articles from the publisher. Search Google with site:domain.com "article title". If three of the five don't return the article on page one of site: results, the domain has indexation problems. Your link will be live but living on a page Google barely crawls.

This check takes four minutes and eliminates more publishers than any other on the list.

Check 6: Editorial process exists

You should be able to identify, from public information on the site, at least two of the following: a named editor or editorial team, contributor guidelines, a corrections policy, an author bio system that links to author archives. Sites that don't have these aren't disqualified — small blogs operate without them — but the absence raises the threshold on every other check.

Check 7: Outbound link velocity

A site adding 40 outbound dofollow links per month has either gone viral (rare) or is selling at scale (common). Compare their outbound link count today versus six months ago. A 5x increase without a corresponding traffic increase is a red flag.

Check 8: Author byline coherence

Pull the author profile of whoever will publish your piece. Have they written about a coherent topic set? Do they have any signal of being a real person (LinkedIn, Twitter, a portfolio)? Or are they "Sarah Williams" with 340 articles across 28 unrelated topics and a stock-photo headshot? Editors who don't exist aren't editing.

Check 9: Niche pricing alignment

Sensitive niches (casino, crypto, CBD, adult, forex, financial trading) cost more on real publishers. If a site offers casino placements at $120, the placement is either on a page that's going to be deindexed or surrounded by enough other casino outbound links that yours becomes a needle in a stack of identical needles. Bazsy applies a 2.5x multiplier on casino, 2x on crypto/CBD/adult/forex specifically because pricing below those levels signals a problem.

Check 10: Real placement examples on file

The publisher should be able to share two or three recent placements they've sold, with URLs. You're checking three things: are the placements still live, are they still indexed, did the surrounding content remain natural after publication. Many publishers will pull links 60 to 90 days after payment. Sampling three recent placements catches this pattern fast.

Check 11: Topical fit, not just niche fit

The final check is judgment. A high-DR publisher with strong traffic in fintech is the wrong placement for a workflow automation tool if their last 30 fintech pieces are all about consumer banking. Topical fit operates at the page level, not the domain level. The page where your link sits has its own topical neighborhood, and that neighborhood is what Google reads.

This is the check that automation can't fully replace, which is why every Bazsy listing gets a final human pass before approval.

What the eleven checks actually filter

Of every 100 publishers that apply to Bazsy, our approval rate sits between 12 and 18 depending on the niche. The bottleneck checks are Check 2 (traffic-to-DR), Check 3 (anchor pattern), and Check 5 (indexation health). Together they eliminate 60-some percent of applicants. Domain age and editorial process screen out another tier. The final 12–18% are publishers worth recommending.

If you're buying guest posts from sources that don't run an equivalent screen, you're paying for the 70%. The math doesn't work even when the price looks attractive.