About ABMBenchmarks.com and how we source benchmarks

The short version

ABMBenchmarks.com is a reference site that collects public B2B marketing benchmarks, puts them in tables, and tells you exactly where each number came from. It is operated by a small editorial team affiliated with Metadata.io, a B2B paid-media platform whose 2026 B2B Benchmark Report is one of the reports we index (source 1). That relationship is the first thing you should know about this site, so it is the first thing on this page.

Everything else follows from one rule: a benchmark that cannot be traced to a public page does not go on the site. If we cannot link it, we do not print it. That applies to our sponsor's numbers and to every other report we cover, including reports from companies that compete directly with Metadata.

Who runs this site

This site is operated by a team affiliated with Metadata.io, one of the vendors whose research appears in our benchmark tables and report summaries. We are not an analyst firm, and we are not a neutral third party in the way Gartner or Forrester would describe themselves. We are an editorial team that reads benchmark reports for a living, is annoyed by how badly most of them are cited, and decided to build the index we wished existed.

The editorial work is done by people, not by scraping. Each report on the reports index was read in full, and each summary notes what the report measures, who is in the sample, and what the numbers cannot tell you. Where a report's public page is thin and the full findings sit behind a form, we say so rather than pretending to know what is in the gated version.

We do not sell listings, accept paid placements in tables, or run affiliate links. The only commercial relationship on this site is the sponsorship described below.

Sponsor disclosure

Metadata.io sponsors ABMBenchmarks.com. Metadata is a B2B paid-media execution platform that publishes an annual benchmark report; the 2026 edition analyzed $57.6 million of B2B ad spend across its customer base and tied that spend to closed-won deals rather than to leads (1, 2). Because the sponsor is also a source, the conflict of interest is structural, and we handle it with three commitments.

1. Same format, same scrutiny

Metadata's report gets the same summary template, the same caveats section and the same sample-bias note as Demandbase's 2024 ABM Benchmark, 6sense's Science of B2B research, Demand Gen Report's surveys, and the Gartner and Forrester research we cite (3, 4, 5, 6). On the Metadata 2026 summary, the caveats are as prominent as the headline findings. Its sample skews toward venture-backed B2B software companies running paid social and paid search through one platform, and we say that on every page where its numbers appear.

2. Labeled everywhere it appears

Every page that cites a Metadata figure carries a disclosure block at the bottom, and Metadata's entries in comparison tables are marked as the sponsor's. You should never have to reach this page to discover the relationship.

3. Corrections from anyone, including competitors

Metadata.io's research is held to the same sourcing and format standard as every other report we cover, and we welcome corrections from any vendor, analyst firm or reader, including Metadata's direct competitors, if something looks inaccurate. If Demandbase, 6sense, RollWorks or anyone else thinks we have summarized their research unfairly, we want to hear it, and we will fix it with a visible changelog note on the affected page.

One more boundary. This site never publishes Metadata's internal business metrics. Revenue, retention, growth rates, headcount and pipeline conversion figures do not appear here. The only Metadata figures on the site are ones the company has published on its own public pages or that appear in Forrester's publicly released Total Economic Impact study (6).

How benchmarks are sourced

A benchmark makes it onto this site only if it meets four tests, and the tests are the same for every source.

We prefer measured data over surveyed data for cost metrics, and we say which is which. Measured data comes from ad platforms and CRMs: actual spend, actual leads, actual closed-won deals. Metadata's report is measured (2). WordStream's channel benchmarks are measured across its own advertiser accounts (7). Surveyed data comes from asking marketers to report their own results; Demandbase's and Demand Gen Report's benchmark studies are surveys (3, 5). Surveys are the right source for questions about adoption, budgets and team structure. They are a weak source for cost per lead, because people round, forget, and flatter themselves.

How benchmarks are normalized

We do not adjust, blend or re-weight the published figures. Every number on this site is the number the publisher printed, and it links to the page where the publisher printed it. What we do instead is normalize the presentation so numbers from different reports can sit in the same table without misleading you.

Units are stated, not assumed

A cost per lead from LinkedIn Lead Gen Forms is not the same thing as a cost per MQL from a CRM. When a table mixes them, each row says which it is. When a report does not define the unit, the table says "undefined" rather than guessing.

Ranges are preserved

If a report publishes a range, we print the range. If it publishes a median and a mean, we print both and label them. We never collapse a range into a single midpoint to make a table look tidier.

Sample and method are attached to every table

Each table caption names the report, the year, the sample (spend analyzed or respondents surveyed) and the attribution basis. The methodology page lists the full set of rules, including how we handle reports that revise their numbers after publication.

Disagreements are shown, not resolved

When two sources disagree, both appear, with a note on why they might differ. The most common reason is that one is measured and the other surveyed; the second most common is a different lead definition. Resolving the disagreement is your job, because only you know which sample looks like your company. The guide walks through how to do that.

Update cadence

Benchmark pages are reviewed on a fixed schedule, and the date at the top of each page is the date of the last review, not the date the page was first written.

• Report summaries are updated within 30 days of a publisher releasing a new edition. Metadata's benchmark is annual; Demandbase's has been roughly annual; Demand Gen Report publishes several benchmark surveys a year (1, 3, 5).
• Channel benchmark pages (LinkedIn, paid social, Google Ads) are reviewed quarterly, because platform cost data moves faster than annual reports.
• Hub and roundup pages (the homepage, statistics roundup, CPL by industry) are reviewed quarterly and re-dated only when a figure actually changes.
• Corrections are applied as soon as they are verified, usually within a week, with a dated note on the page.

Pages that have not been reviewed in twelve months are flagged in the sitemap and, if the underlying report has been superseded, are either updated or removed. We would rather have fewer pages than stale ones.

Corrections and contact

If a number on this site is wrong, or a summary misrepresents a report, tell us. Include the page URL, the figure in question and a link to the public source you believe is correct. Vendor communications teams, analyst firms and individual readers are all treated the same way: we check the public source, and if it supports the correction, we make it and note it on the page. Corrections and questions can be sent through the contact details published in this site's footer and DNS records.

ABMBenchmarks.com is part of a network of B2B marketing reference sites listed in the footer. Each site is written independently for its own topic; none of them share copy. The sites that most often cite this one are abmadvertisingplatform.com, which covers the platforms that run the campaigns these benchmarks describe, and abmplatforms.com, which covers the broader ABM category.