Link building outreach, explained

By Bill Hartzer · Hartzer Consulting · 25+ years in search, domains and expert witness work

What outreach is, and what it is not

Link building outreach is the practice of asking a specific person to add a link to a specific page, and giving them a reason to say yes. That is the entire discipline. The tooling, the sending infrastructure, the follow-up cadence and the tracking spreadsheet all exist to put one request in front of one human at a moment when they are able to act on it.

It is worth being precise about what outreach is not, because the adjacent disciplines are measured differently and fail differently.

• It is not digital PR. Digital PR pitches a story to a journalist who is looking for stories. Outreach pitches a resource to a site owner who is not looking for anything.
• It is not content marketing. Content marketing publishes something and waits. Outreach publishes something and then goes and asks.
• It is not paid placement. The moment money changes hands for the link itself, you are in a different risk category with a different set of rules.

Most people who say "we do outreach" mean they send a large number of near-identical emails and accept a very low reply rate as the cost of doing business. That is one way to run it. It is also why the average recipient's inbox has trained them to delete your message before the second sentence.

The honest framing: outreach is a favor economy. You are asking someone to spend two minutes of unpaid attention editing their own site for your benefit. Every design decision follows from taking that seriously.

The ask, and why it usually has no answer

Every outreach email contains an implicit answer to the question why would you do this. Most of the time that answer is missing, and the recipient supplies their own: because someone wants a link.

There are only a few answers that hold up.

• You are fixing something broken. A dead link on their page, a misattributed image, an outdated statistic. The favor runs in your direction because they get a corrected page.
• You are supplying something genuinely missing. A resource list that does not include the obvious tool. A guide that stops short of the thing readers ask about next.
• You already exist in their world. They have mentioned you, quoted you, used your data, or you have a relationship that predates the email.
• You are offering something they were going to have to make anyway. Original data, a quote from a named expert, an image they can license.

If your pitch does not fit one of those, you do not have an outreach campaign. You have a mailing list. I have watched teams send tens of thousands of emails on the theory that volume substitutes for a reason, and the arithmetic never works out — the reply rate collapses faster than the send volume rises, and the sending domain burns.

Module 5 works through pitch structure line by line, including a full annotated example and the specific sentences worth cutting from every draft. The reference version is simply this: name the page, name the reason, make the ask once, and stop.

Sequencing and follow-up

A single email is not a campaign. Most positive replies to outreach arrive after the first message, for the mundane reason that people are busy and email is a queue. But there is a point past which persistence becomes harassment, and it arrives sooner than most sequence templates assume.

The reference shape:

• First contact. The full pitch, the reason, the specific page.
• One follow-up. Spaced by several days, not hours. It should say something new — a second angle, a clarification, an offer to send the asset rather than link to it. A follow-up that only says "just bumping this" is a reminder that you want something, not a reason to give it.
• Stop. Two unanswered emails is a no. Three is an annoyance. Beyond that you are damaging a relationship you may want later.

Automated sequences make the third step optional, which is exactly the problem with them. A tool that will happily send seven messages to someone who never opened the first is a tool that converts prospects into people who will never work with you.

Sequencing also interacts with the calendar in ways worth planning around. Editorial sites publish on cycles. A resource page maintained once a quarter will not be edited for you on a Tuesday in between. Timing your ask to a moment when the page is being touched anyway raises the hit rate more than any subject line experiment I have run.

Deliverability, at a summary level

None of the above matters if the message lands in a spam folder. Deliverability is the least interesting part of outreach and the part that most often silently kills a campaign, because a message filtered before delivery produces the same result as a message ignored: nothing.

The mechanics, at reference depth:

• Authentication. SPF, DKIM and DMARC are three DNS records that let a receiving server verify that mail claiming to come from your domain actually did. Without them, cold email from a new domain has a poor chance of inbox placement, and major providers have been tightening the requirement rather than relaxing it.
• A separate sending domain. Outreach should not be sent from the domain that carries your customer email, your invoices and your password resets. If the sending reputation goes bad — and on a cold campaign it eventually will — you want the damage contained.
• Warmup. A brand new domain that starts sending hundreds of messages a day looks exactly like a spam operation, because that is what spam operations do. Volume has to ramp.
• Engagement signals. Replies, opens and the absence of spam complaints feed back into placement. This is the deep reason relevance matters technically and not just ethically: irrelevant mail gets marked as spam, and spam complaints degrade delivery for every message that follows.

Marketing automation platforms are the wrong tool here. They are built for opt-in lists, they add tracking that trips filters, and their shared sending infrastructure means someone else's behavior affects your placement. Module 5 covers the setup properly, including what warmup actually involves.

Why most outreach fails

Failure in outreach is rarely a single mistake. It is a chain, and it breaks at the earliest weak link.

1. The prospect list is wrong. Everything downstream is wasted if the sites on the list were never going to link to anything. This is the most common failure and the least often diagnosed, because a bad list produces the same symptom as bad copy: silence.
2. There is no asset. Outreach for a commercial page with nothing to recommend it is a request for charity.
3. The email is obviously templated. Recipients detect merge fields instantly. A first line that could be pasted into any of four hundred other emails tells the reader precisely how much thought went into the request.
4. Nobody follows up, or everybody follows up forever. Both extremes cost links.
5. The wrong metric is watched. Emails sent is an activity number. Links live after ninety days is an outcome number.

There is a structural point underneath all five. Outreach does not scale gracefully. Every increase in volume reduces the per-message quality unless headcount rises with it, and the quality is the thing that produces replies. Teams that treat outreach as a volume channel discover this the expensive way.

Judging whether outreach is working

Reply rate is the first honest signal and the fastest to read. If a campaign is producing almost no replies at all — including negative ones — the problem is upstream of the copy: the list, the deliverability, or the absence of a reason.

Beyond that, three measures are worth keeping.

• Referring domains gained, not links gained. One site that links to you eleven times is one endorsement, not eleven. This distinction dominates real link profiles; on one twenty-year profile I analyzed, the top 100 referring domains accounted for 84.3% of all recorded links, and a single sitewide link accounted for 16.8% on its own.
• Survival at ninety days and one year. Links are not permanent. Across 1,301,839 links from 22,260 referring domains, the median referring domain stopped linking after 1,080 days and 49.6% of all links ever recorded were eventually lost. A campaign report that counts placements without checking whether they are still there is reporting a snapshot as a balance.
• Relationships opened. The reply that says "not this one, but send me the next thing" is worth more than the link you did not get, and no standard outreach tool tracks it.

The uncomfortable conclusion is that good outreach is slow, personal and hard to delegate. That is also why the links it earns are worth having: the same friction that limits your throughput limits everyone else's.