Skyscraper technique

The skyscraper technique, popularized by Brian Dean: find content that has attracted a lot of backlinks, create something demonstrably better, then reach out to everyone linking to the original and pitch your upgrade. It's a classic tactic that still works in 2026, with caveats. This page walks through the 3-step workflow, why most attempts fail, the "upgrade test" you should apply before starting, and when to pick an alternative tactic instead.

The three steps

1. Find content that has many links

Tools: Ahrefs Content Explorer, SEMrush, BuzzSumo. Search for your topic, sort by referring domains. Identify pieces with 50+ linking domains.

2. Make something genuinely better

"Better" requires real effort. Common dimensions:

• More comprehensive (covers more angles, more depth)
• More current (data, examples, screenshots from 2026)
• Better structured (easier to navigate and consume)
• Better visuals (original graphics, diagrams, videos)
• Better UX (faster, mobile-optimized, interactive elements)
• More credible (original data, authoritative authors)

3. Reach out

Export everyone linking to the original. Email them individually with the same structure: "You linked to X. I just published Y, which is Z better in these ways. Thought you'd want to see it."

Why skyscraper often fails

The upgrade test

Before starting: could a sophisticated reader land on your version and the original, read both, and say "wow, the new one is way better"? If not, skip. Length and adjectives aren't the test. Actual value is.

Conversion rates

• Cold skyscraper outreach: 1 to 3%
• Genuinely superior content plus personalized outreach: 5 to 10%
• Same plus an existing relationship: 15 to 30%

When to use skyscraper

• The topic has at least one authoritative existing resource
• You have unique value to add (data, experience, better examples)
• You have time to do real outreach (100+ personalized emails)
• The topic is evergreen, so you want durable link value

When to skip it

• Topic is already covered by 10 amazing resources; hard to be "the" upgrade
• You don't actually have more to say than the incumbent
• You can't commit to the outreach phase
• Topic is trend-dependent (today's upgrade is tomorrow's stale content)

Alternatives to consider

Often a skyscraper is less effective than:

• Skyscraper + digital PR. Your "better" version is the data study. Earn press coverage, then backfill with outreach.
• Topic expansion. Instead of competing on the same topic, cover an adjacent one nobody has written about well.
• Format change. Top content is a text article. You publish an interactive tool on the same topic. Links come from novelty.

The honest take

Skyscraper works, but it's not magic. It's content marketing plus outreach, packaged. The content bar is high. The outreach effort is high. Most attempts fail because they skip the "genuinely better" step. Don't skip it.

What to do with this

Pick one topic you want to own. Find the single most-linked page on that topic today. Read it carefully. Ask honestly: can I produce something meaningfully better? If yes, commit to the full 3 steps. If no, pick another tactic.

Next: anchor text, the detail that quietly signals to Google whether your links are natural.

The three steps

1. Find content that has many links

Tools: Ahrefs Content Explorer, SEMrush, BuzzSumo. Search for your topic, sort by referring domains. Identify pieces with 50+ linking domains.

2. Make something better

"Better" requires real effort. Common dimensions:

• More comprehensive (covers more angles, more depth)
• More current (data, examples, screenshots from 2026)
• Better structured (easier to navigate + consume)
• Better visuals (original graphics, diagrams, videos)
• Better UX (faster, mobile-optimized, interactive elements)
• More credible (original data, authoritative authors)

3. Reach out

Export everyone linking to the original. Email them one by one with the same structure: "You linked to X. I just published Y, which is Z (better in these ways). Thought you'd want to see it."

Why the technique "works"

• Sites that linked once to this topic will link again to better content on it
• The original proved the topic is link-worthy
• Outreach is high-intent: everyone on your list already linked to a related page

Why it often fails

• "Better" is subjective. Your upgrade is only "better" to people who would actually swap.
• Inertia. Most people won't edit old posts to add new links.
• Saturated technique. Skyscraper is popular, which means recipients are jaded.
• Generic outreach. "I wrote something longer than X" templates get ignored.

The upgrade test

Before you start: could a sophisticated reader land on your version and the original, read both, and say "wow, the new one is way better"? If not, skip. Length and adjectives aren't the test.

Conversion rates

• Cold skyscraper outreach: 1-3%
• Genuinely superior content + personalized outreach: 5-10%
• With a prior relationship: 15-30%

When to use skyscraper

• The topic has at least one authoritative existing resource
• You have unique value to add (data, experience, better examples)
• You have time to do real outreach (100+ personalized emails)
• The topic is evergreen, you want durable link value

When NOT to use it

• Topic is already covered by 10 amazing resources, hard to be "the" upgrade
• You don't actually have more to say than the incumbent
• You can't commit to the outreach phase
• Topic is trend-dependent (today's upgrade is tomorrow's stale content)

Alternatives to consider

Often a skyscraper is less effective than:

• A skyscraper combined with digital PR. Your "better" version is the data study. Earn press links, then run the outreach to backfill.
• Topic expansion. Instead of competing on the same topic, cover an adjacent one no one has written about well.
• Format change. The top content is a text article. You publish an interactive tool on the same topic. Links come from the novelty.

The honest take

Skyscraper works, but it's not magic. It's content marketing + outreach, packaged. The content quality bar is high; the outreach effort is high. Most attempts fail because they skip the "genuinely better" step. Don't skip it.