← All articles

Link Building 2027: What to Bury and What Still Wins

By VonixCo Team · · 14 min read

A digital marketing strategist reviewing a link building 2027 checklist at a modern office desk, with a laptop showing analytics charts

Before 2027, stop buying links by volume and stop paying for them by domain authority. Bury mass outreach, DA-priced link vending, low-tier guest-post farms, private blog networks, link swaps, forced commercial anchors and generic roundup requests. Keep funding editorial guest posts on real sites, digital PR and data-led content, brand mentions and citations — and judge every link by one test: it should benefit your brand, the publisher, and the publisher’s audience. That is the link building 2027 reset.

What is the link building graveyard?

The link building graveyard is a metaphor from a Search Engine Journal webinar held on October 21, 2026, “The Link Building Graveyard: What To Bury & What Still Wins.” Michael Johnson, CEO and founder of Resolve, explained which tactics he would leave behind and which he would keep funding into 2027. Loren Baker hosted the session, and Search Engine Journal published a recap on October 30, 2026. Source: Search Engine Journal, “The Link Building Graveyard: What To Bury & What Still Wins” (Oct 30, 2026), by Heather Campbell. A secondary summary is at Linkbaits, “The Link Building Graveyard: What to Bury” (Nov 5, 2026).

The graveyard is not a list of tactics Google has banned. It is a practitioner’s keep-or-cut list: the lists below are the analysts’ recommendations — Johnson’s and Gaetano DiNardi’s — not Google guidance. What has changed is the environment: AI assistants now answer buyer questions directly, and a link that only moves a ranking without ever being seen, cited or trusted by a real audience is worth less than it was. Johnson’s point is that link building in 2027 is measured not only on rankings but on whether the brand appears in the answers buyers ask AI assistants.

Why did domain authority stop being a useful price tag?

For years, link building ran on one model: the higher the domain authority, the higher the price of the link. Vendors priced placements by DA or Domain Rating, buyers compared offers by the same scores, and monthly reports counted links and DA values as if that were the result. DiNardi’s argument, in a BuzzStream podcast published on November 4, 2026, is that this model is broken — and AI search broke it. Source: BuzzStream, “Why Traditional Link Building Doesn’t Work Anymore (and What Does) with Gaetano DiNardi” (Nov 4, 2026). The episode discusses his Search Engine Land article on why the traditional link building model no longer works for AI search.

The problem is not that domain authority is a bad number, but what it is used for: a substitute for judgement. A site can carry a high DA score while being a content mill with no real business, no brand, and no audience that would ever buy anything. DiNardi’s phrase for the replacement measure is that “citation rate is the new domain authority” — how often a brand is actually cited in AI answers. He distinguishes citation rate from recommendation rate: an AI assistant can quote your page while recommending every competitor on the list. Johnson makes the same point from the publisher side, arguing against letting Domain Authority or Domain Rating be the primary measure of a placement. When DA is the price, vendors optimise for DA — high scores, low editorial standards — and sell the same inventory to everyone.

Two marketing professionals discussing a content and outreach strategy beside a whiteboard in a bright modern office

What link building tactics should you bury before 2027?

The bury list below combines Johnson’s cut list with DiNardi’s account of the old vendor model. None of these tactics is presented as a Google penalty trigger; the two analysts say they no longer repay their cost.

  • Mass outreach. DiNardi describes the industrial model directly: vendors with publications “in their back pocket” that they return to again and again, negotiating fees behind the scenes while the client pays an agency fee on top of a per-link price. Mass outreach is the engine of that model.
  • DA-priced link vending. When a link’s price rises with its DA score, the market rewards whoever manufactures the highest score most cheaply. DiNardi notes that in one marketplace database of roughly twenty thousand links, only a small fraction met a basic quality bar, and the prices for those ran into thousands of dollars.
  • Low-tier guest-post farms. These sites exist only to publish guest posts. They have no real product, no real brand and no readers who arrive for any reason other than search. A link from one is a receipt for a transaction, not an endorsement.
  • Private blog networks. Johnson puts private blog networks at the top of his cut list: they exist only to pass ranking signals and offer nothing to a reader, so they fail his three-way test.
  • Link-swap schemes. Reciprocal links arranged for their own sake are a transaction, not an endorsement — nobody involved is acting on editorial judgement.
  • Forced commercial anchors. Anchor text engineered to carry a commercial keyword is a signal of manipulation, not relevance. DiNardi points to exactly this pattern — “best solar panels” on one page, “best credit cards” on the next — as a red flag on vendor sites.
  • Generic requests to be added to roundups. A publisher asked to add a link for no reason has no reason to say yes — any pitch that cannot explain what the reader gains belongs in the graveyard.
  • Volume-for-volume’s-sake buying. Buying links to hit a monthly number keeps all the tactics above alive. DiNardi’s point is blunt: you no longer need batches of random links — you need to be part of the higher-influence sources for your category.

One warning: burying a tactic does not mean deleting the links you already have. Audit first, and treat the bury list as a buying decision for next year. If you are unsure whether an existing link is harmful, read our guide to how to vet a guest posting site before acting on it.

What still wins in link building 2027?

The keep list is shorter than the bury list, and that is the point. In link building 2027, the winning tactics survive both tests in this article — Johnson’s three-way benefit test and DiNardi’s common-sense rubric.

  • Editorial guest posts on real sites. A guest post on a site with a real business, a real brand and a real audience is still the most reliable way to earn a link that reaches readers, carries an editorial endorsement, and can be cited by AI systems. The difference from the farm model is the site, not the format. See also our analysis of whether traditional link building is dead.
  • Digital PR and data-led content. Johnson’s keep list centres on assets audiences actually need — original data, research, tools and resources publishers want to reference. Digital PR reverses the pitch: instead of asking a publisher to add your link, you give them something their readers want and the link follows. DiNardi’s example is the executive interview: getting your team interviewed on relevant podcasts, with links to your content pages in the show notes, is a tactic he has seen make the difference for AI visibility.
  • Brand mentions and citations. Unlinked mentions of your brand on real sites are the raw material of AI visibility. AI assistants cite brands they have seen mentioned in credible context, so a mention on a genuine publication can matter even before it becomes a link. Our guide to link building for 2027 covers the same shift in more detail.
  • Resource pages and listicles that earn their place. Johnson does not bury listicles or resource pages — he buries the generic request to be added to one. The keep version: identify the worthwhile publishers and resource pages in your niche, create the asset their audience needs, and pitch it. The link is earned by the asset, not by the ask.
  • Links from sources of influence in your category. Categories are shaped by a few influential sources — review platforms, analyst firms, major publications — and presence in them matters more than the raw count of backlinks. DiNardi’s example is a Gartner category page: for insider risk management platforms, it influenced the majority of AI outcomes, and the right move was to fix the brand’s category alignment and review presence — not to buy more links to a listicle.
Bury before 2027 Keep funding in 2027
Mass outreach to hundreds of sites Targeted pitches to real publishers
Links priced by DA or DR Links judged by audience fit and relevance
Low-tier guest-post farms Editorial guest posts on real sites
Private blog networks Digital PR and data-led content
Link-swap schemes Brand mentions and citations
Forced commercial anchor text Resource pages earned with a real asset
Generic roundup requests Presence in your category’s influential sources
Buying links to hit a monthly number Fewer, better links that pass the three-way test

The table summarises the two analysts’ recommendations, not a Google policy.

How do you test whether a link is worth buying?

Johnson’s test runs first: it takes ten seconds and kills most bad placements on the spot. A worthwhile link should benefit three parties at once: your brand, the publisher, and the publisher’s audience. If it benefits only your brand, only the publisher as a fee, or nobody in the publisher’s audience, it fails.

DiNardi’s rubric is the second test, run on the site itself. He calls it common sense rather than an SEO metric, and it has four questions. Is this a real business? If content is the product, the site is generally not a good source. Does it have a brand at all? Most vendor-supplied sites have none. Does the traffic geography align with your client? Vendor sites often draw most of their traffic from countries unrelated to the client’s market. And can a user complete a task on the site? His example is an ROI calculator page that, instead of a working calculator, offers steps for calculating ROI by hand.

  1. Apply Johnson’s three-way test. Name the benefit to your brand, to the publisher, and to the publisher’s audience. If you cannot name all three, do not buy the link.
  2. Check the site is a real business with a real brand. Look for a product, a service, a team — anything that exists independently of selling links.
  3. Check the traffic geography. If the site’s audience sits in countries your client does not serve, the placement reaches nobody who matters, whatever its DA score says.
  4. Check that a user can complete a task. Open the site as a reader: can you do the thing the page promises?
  5. Read the page’s other outbound links. A page linking to “best solar panels”, “best credit cards” and “best LMS software” in the same breath is selling anchors, not making editorial choices. Walk away.
  6. Ask what happens if the ranking value disappears. Would the link still be worth having for the readers it reaches? If not, you are buying a ranking signal.

The two tests also answer how many links you need: a list of placements that pass both tests — and the discipline to stop when the list runs out.

Hands typing on a laptop keyboard beside a notebook and pen, planning an editorial guest posting campaign

What is the “dark SEO” era, and why does it change how you measure links?

DiNardi calls the current moment the “dark SEO” era: more organic performance now shows up as branded and direct traffic, because AI assistants answer questions without sending a click. The click you can see in analytics is a shrinking share of the influence a link actually had.

That has two consequences. The first is measurement: judged only by referral traffic and ranking movement, a campaign undervalues the links that matter most.

The second consequence is strategy. In the dark SEO era, you cannot brute-force your way into a category you do not belong in. DiNardi’s example: a brand ranks number one for a “best X” search by pounding backlinks at a list page, while the AI assistant takes the page’s content, recommends every competitor, and gives the brand only a citation. The fix is not more links to the same page; it is the work of belonging in the category.

How do you turn the keep-or-cut list into a 90-day plan?

Johnson’s webinar ended with a five-step sequence, each step depending on the one before it.

  1. Audit the existing programme. List every link you pay for or build, run the two tests on each, and mark every failure “do not renew”.
  2. Establish a baseline for relevant AI buying prompts. Write down the questions buyers ask AI assistants about your category, and record where your brand appears in the answers today.
  3. Identify worthwhile publishers and resource pages. Build a short list of real sites in your niche — publications with audiences, resource pages buyers use, and the influential sources for your category.
  4. Create the assets those audiences need. Original data, a useful guide, a working tool, a study with numbers you collected yourself. The asset makes the pitch possible; without it, you are back to the generic roundup request.
  5. Pitch them. Targeted outreach to the short list, one publisher at a time, with a reason the publisher’s readers benefit. Then measure against the baseline: did the brand appear in more of the answers buyers ask?

Ninety days is enough to run this sequence once, properly — not while also buying the old way in parallel. The honest version is a switch: the money that bought volume now funds the assets and the outreach.

When should you hire a service like VonixCo?

Hire help when the 90-day plan is more work than your team can do consistently — auditing the links you already pay for, building the short list of real publishers, creating the assets they want, and pitching them while tracking AI answers. One audit is easy; doing it every quarter takes a team and a process.

VonixCo is a niche-relevant guest posting and SaaS link building service based in Lahore, Pakistan. Its services are SaaS guest posts, niche guest posts, link building and guest post content writing, with plans at $500, $950 and $1,800 per month, content included and a live link report with every placement. Contact: info@vonixco.com and WhatsApp +92 340 0096621.

If your needs are simple — a few placements on sites you already know, a short audit — do that work yourself, using the two tests in this article. Bring in a service when the volume or the outreach grows beyond what your team can sustain. No honest provider can promise that a particular link will lift your rankings, or that an AI assistant will recommend your brand.

Frequently asked questions

What is the link building graveyard?

It is a Search Engine Journal webinar from October 21, 2026, in which Resolve CEO Michael Johnson set out which tactics to bury and which to keep funding into 2027. It is a keep-or-cut list, not a Google policy: tactics that fail his test — a link should benefit the brand, the publisher and the publisher’s audience — belong in the ground.

What link building tactics should you bury before 2027?

Mass outreach, DA-priced link vending, low-tier guest-post farms, private blog networks, link-swap schemes, forced commercial anchors, generic roundup requests and volume-for-volume’s-sake buying. Johnson’s cut list covers the manipulative tactics; DiNardi adds the vendor model behind them.

What still wins in link building 2027?

Editorial guest posts on real sites, digital PR and data-led content, brand mentions and citations, resource-page placements earned with a genuine asset, and presence in your category’s influential sources.

What is citation rate, and how is it different from recommendation rate?

Citation rate is how often a brand is cited in AI-generated answers; DiNardi calls it the new domain authority. Recommendation rate measures whether the AI assistant actually recommends your brand, rather than merely quoting your page while recommending competitors. A rising citation rate with a flat recommendation rate means your content is used but your brand is not chosen.

How do you test whether a link is worth buying?

Run Johnson’s three-way test first: the link should benefit your brand, the publisher, and the publisher’s audience — if you cannot name all three benefits, do not buy it. Then run DiNardi’s rubric: real business, real brand, aligned traffic geography, and a site where a user can complete a task. Check the page’s other outbound links too.

Is link building dead in 2027?

No. Both analysts say links still matter — a brand needs a certain level of authority to clear competitive thresholds, and editorial links from real sites remain an asset. What is dead is the volume model: links bought by the dozen, priced by DA, from sites with no audience. Link building in 2027 means fewer, better links, judged by audience fit and by whether the brand appears in the answers buyers ask AI assistants.

Want niche-relevant guest post placements?

Tell VonixCo about your website and niche, and we will reply with suitable publishing options and pricing.

Keep reading