What actually works
in link building
outreach.

In this dataset, highly personalized emails (referencing specific articles, recent publications, or named editors) showed higher response rates — but primarily when targeting higher-DR domains. For lower-DR targets, the response rate difference between personalized and templated emails was minimal. The ROI of personalization effort scales with target DR. Spending 20 minutes personalizing an outreach to a DR 15 blog is a poor use of time. Spending it on a DR 55 publication is not.

This one is more nuanced than the conventional wisdom suggests. Across this dataset, email length had a weaker correlation with response rate than specificity did. A 300-word email that clearly articulates why the link is relevant to the target site's audience outperformed a 60-word generic pitch at comparable DR levels. Specificity matters more than brevity. The ideal length is "as short as it can be while still being specific" — which is longer for high-DR targets and shorter for lower-DR ones.

This holds clearly in the data. Teams that send exactly one follow-up (typically 5–7 days after the initial email) show meaningfully higher placement rates than teams that don't follow up at all. The effect of a second follow-up is smaller and more variable. A third follow-up shows no consistent benefit and correlates with being blocked/marked as spam by editorial contacts. The optimal follow-up strategy across this dataset: one, at 5–7 days, with new information or a slightly different angle — not a repeat of the original email.

This is one of the most widely cited pieces of outreach advice. In this dataset, send-day variation showed no consistent pattern in response rates across the teams that tracked it. The day of week effect — if it exists — appears to be swamped by factors like email quality, relevance of the pitch, and the individual preferences of the recipient. Optimizing send time is a low-leverage activity. Improving pitch specificity and target relevance scoring are higher-leverage uses of the same time.

Strongly confirmed. Teams that systematically identify the specific editor, content manager, or writer responsible for the relevant section of a target site show substantially higher response rates than teams that pitch to generic editorial or info@ addresses. At high-DR publications, sending to the wrong person within the organization has roughly the same outcome as not sending at all. Investment in contact identification — not just domain identification — pays off consistently in this dataset.

At low DR targets, yes — to a point. At high-DR targets, no. The data shows a clear ceiling on volume-compensated outreach: beyond a certain scale, response rates decline as the same contacts receive repeated generic pitches from the same or similar domains. High-volume, low-personalization outreach is self-defeating at DR 45+ targets because it degrades the domain's reputation with exactly the editorial contacts it most needs to reach. Teams that burned these contacts with volume campaigns reported subsequent difficulty placing even genuinely relevant pitches.

Confirmed across multiple dimensions. Pitches that offer a concrete value to the target site — a data point they can use, an expert quote, a correction to outdated content, a gap in their existing coverage — outperform pitches that simply request a link. The framing shift from "please link to us" to "here's something useful for your readers" is associated with the largest single response rate improvement in this dataset — larger than personalization, email length, or timing.

Response rate and placement rate are related but not the same thing — and in this dataset, teams optimizing for response rate sometimes made choices that improved responses while reducing placements (e.g. more casual, lower-commitment initial asks that generated replies but rarely converted). The metric that matters is DR-weighted placements per 100 pitches, not response rate alone. A 4% response rate that converts to 3% placements at average DR 48 outperforms a 12% response rate that converts to 2% placements at average DR 22.