What syndication is and is not
Syndication is republishing the same article, in full or very nearly in full, on another organisation's site with your permission and your name on it. The partner gets a finished piece for its readers; you get reach, attribution and a route back to your site. The content distribution channels guide places it among the earned channels, beside partners and press, and warns in one sentence about the main risk: a verbatim copy on a larger site can be the version search shows. This guide expands that sentence into a working method.
Several arrangements look similar and are not. The table separates them, because each has a different search question and a different page on this site that covers it.
| Arrangement | What is copied | Who it serves | The search question | Where it is covered |
|---|---|---|---|---|
| Article syndication to a partner | Your article, in full, on the partner's site | The partner's readers, and you through attribution | Which copy search shows for the article's queries | This guide |
| Republishing on a platform you post to | Your article on a publishing platform or professional network | Your followers on that platform | Whether the platform lets the copy point at the original | This guide |
| Repurposing | Adapted derivatives in other formats | Each channel's own audience | None, because each derivative is a different page | Repurposing content |
| Wire distribution of a press release | The release on many partner sites at once | Newsdesks and the public record | Copies are treated as duplicates and add little authority | Press releases and the online newsroom |
| Lead-generation syndication | A gated asset placed by a vendor to collect contact details | The sales pipeline | Not a search arrangement at all | MQL, SQL and lead stages |
| Scraping | Someone else's content, without permission or added value | Nobody but the copier | A practice Google's spam policies name | Google's spam policies |
The last row matters in both directions. Google's spam policies, as the spam policies guide sets them out, describe scraping as content taken from other sites and republished without original content or added value, including copies that have been altered slightly or reworded by automation. Syndication done properly is a different arrangement: the copy exists by agreement, names its source and is set up so that search can tell which version is the original, which is the case Google's documentation on duplicate URLs plans for.
When a partner copy is worth it
A copy of an article is only worth arranging when the partner's readers are people the brief was written for and your own channels do not already reach them. A gardening retailer whose how-to articles appear on a gardening magazine's site meets readers who garden; the same articles on a general news site meet people who scroll past. Start from the audience named in the brief and ask whether this partner's readers match it better than anyone the newsletter, search and social accounts already reach.
Four further questions decide most cases:
- Will the partner credit you where readers will see it? A byline with your organisation's name and a link to the original at the top or the foot of the copy is the minimum. A credit hidden in a footer does nothing.
- Can the partner protect the original in search? Ask before agreeing whether its system can add a canonical to the copy, or a noindex. If it can do neither, decide whether reach alone justifies the risk.
- Is the piece the right kind? Evergreen how-to articles and explainers syndicate well. Pieces whose job is to convert on your site lose their next step on someone else's page, and original research you want cited at your own address is better offered as a summary with a link, which is the caution in the distribution guide's decision table.
- Will the copy stay a faithful duplicate? A partner that wants to cut, rewrite or add sections is asking for a different piece, which changes the search arrangement described below.
If the answers are mostly no, a guest article written for that publication, or a short adapted piece pointing to the original, usually serves the same readers with none of the risk.
Protecting the original: canonical, noindex or a link
There are three ways to keep the original in front of search, and the partner's system often decides which one is available.
A cross-domain canonical is a Canonical TagTells search engines the preferred URL. on the partner's copy that names your original as the preferred URL. The canonical tag versus permanent redirect comparison records that Google's documentation on consolidating duplicate URLs describes syndication and cross-domain canonicals among the cases it supports. It also records the limit: the canonical is a hint that Google weighs against other signals. The crawling and indexing guide adds that cross-domain canonicals are honoured only when the pages really are duplicates, so a copy that the partner has edited heavily, extended with new sections or retitled may be treated as its own page whatever the tag says.
A noindex on the copy asks search engines not to show it in results, while the partner's readers can still find it through the partner's own pages and newsletter. The noindex versus robots.txt comparison explains the condition that makes it work: the crawler has to be able to fetch the page to read the directive, and a robots.txt block used instead does not reliably keep a linked copy out of the index. Use a noindexA robots directive telling search engines not to show a page in their results. where the partner cannot add a canonical, or where it wants to edit the piece enough that it is no longer a duplicate.
An attribution link to the original is the minimum in every arrangement, and the only protection when neither of the others is possible. It gives readers a route to the source and makes the relationship visible, but it does not tell search which version to prefer. When it is the only protection, accept that the copy may be the version shown, or do not syndicate that piece.
| Option | What the partner does | What it tells search | Use it when | Watch for |
|---|---|---|---|---|
| Cross-domain canonical | Adds a canonical on the copy naming your URL | Prefer the original among duplicates | The copy stays a faithful duplicate | Heavy edits, extra sections or a new title that make the pages differ |
| Noindex | Adds a noindex to the copy | Leave the copy out of results | No canonical is possible, or the copy will be edited | A robots.txt block used instead, which stops the noindex being read |
| Attribution link only | Credits you and links to the original | Nothing about which copy to prefer | Reach matters more than search for this piece | The copy appearing for your queries instead of the original |
Two combinations cause trouble. The crawling and indexing guide notes that Google documents a canonical and a noindex on the same page as contradictory, so the copy carries one or the other, never both. And the attribution link should never be a condition that removes the partner's choice: the spam policies guide lists requiring a link as part of a contract, without allowing it to be qualified, among its examples of link spam. Ask for credit and a link; leave the partner free to mark it with nofollowA link attribute asking search engines not to treat the link as an editorial endorsement. if its policy is to do so. If money changes hands in either direction, the copy is advertising, and the links in it are qualified as sponsored, as the outbound links guide explains from Google's documentation on qualifying outbound links.
Agreeing the terms in writing
The distribution channels guide's advice on syndication ends with agreeing the terms in writing, and most disputes come from terms that were only agreed in an email thread. A short written agreement covers:
- Which articles, and whether in full or as an excerpt. Name each one; a blanket licence to "any article" turns into copies nobody tracks.
- What the partner may change. Formatting and headings to fit its template, yes; anything that changes the meaning, no. Corrections to the original are passed on and applied to the copy.
- How the original is protected. Canonical or noindex, who adds it, and how you will check it is there.
- Credit. The byline, the author's name, the attribution wording and the link back, placed where readers see it.
- Timing. Publish on your own site first, so that the original exists before any copy does.
- Duration and removal. How long the copy stays up, and what happens when an article is withdrawn or replaced.
- Rights in the piece. Images, quotes and charts cleared for the partner's use as well as yours; the repurposing guide makes the same point about derivatives.
- Money, if any. A paid arrangement is advertising, with the labelling and link rules that follow.
This is general practice, not legal advice. Joint pieces written together with another business, rather than copies of existing articles, are a co-marketing arrangement, and the co-marketing partnerships guide covers dividing that work.
Republishing on platforms you do not control
Publishing platforms, professional networks with article tools and newsletter platforms with public archives all invite you to post articles you have already published. Whether a copy posted there can carry a canonical pointing at the original is a feature of each platform, documented in its own help pages rather than in the sources this site cites, so check before posting. Where the platform allows it, use it. Where it does not, you have three choices: publish an adapted version written for that platform's readers, publish an excerpt that links to the full article, or publish the full piece and accept that the copy may be what search shows.
The adapted version is usually the best of the three. The repurposing guide's rule, adapting rather than copying, applies here exactly: a post that makes one point from the article in the platform's own style reaches the platform's readers and sends the interested ones to the original, without creating a second full copy.
Republishing other people's work on your site
The same questions arise in reverse when a supplier, a partner or an industry body offers an article for your blog, or when someone on the team wants to republish a piece they admired. Google's spam policies matter more here, because it is your site that carries the copy. Scraping covers content republished without original content or added value, and site reputation abuse covers third-party pages published to take advantage of the host's ranking signals, judged by purpose: is the content there to serve your readers, or to rank on your site's standing?
Four rules keep this safe. Republish only with permission, recorded like any other agreement. Either add something of your own, such as commentary, a local example or a worked case, or put a canonical naming the original, or a noindex, on your copy. Hold the piece to your own editorial standards and subject area, as the spam policies guide recommends for third-party and commercial content. And if the supplier is paying for the placement, label it for readers and qualify its links as sponsored.
Often the better answer is not a copy at all. A short piece in your own words that summarises the original, cites it where the claim is made and links to it, as the outbound links guide recommends for citations, serves your readers and credits the author. A commissioned piece written for your readers serves them better still.
Measuring a syndication arrangement
Judge each arrangement on what it brought, against what the same effort would have produced in your own channels.
| What to measure | Where to find it | What it tells you |
|---|---|---|
| Visits referred by the copy | Analytics referral reports, and tagged links where the partner allows them | Whether the partner's readers followed the credit to your site |
| What those visitors did next | Key events and conversions for the referred sessions | Whether the readers were the right ones |
| Whether the original still appears for its queries | The search performance report for the original's URL | Whether the canonical or noindex is doing its job |
| Which copy appears for a distinctive sentence | A search for an exact phrase from the article | A quick check on which version search prefers |
| Links and mentions the copy earned | Your link and mention monitoring | Whether the arrangement built any lasting BacklinkA link from another site to yours. value |
| Time spent arranging and checking | Your own records | The cost to set against all of the above |
Tag the links back to your site with campaign parameters, following the naming conventions and UTMs guide, so that referred visits from each partner are reported separately; a UTMURL tags identifying traffic source. parameter on a link the partner controls should be agreed with it. Review each arrangement at the end of its term and renew, change or end it on the evidence.
A garden retailer syndicates its how-to articles (illustrative)
Fernlattice Allotment Supplies lets a magazine republish four articles
Fernlattice Allotment Supplies, an invented allotment and garden supplies retailer, agrees to let a gardening magazine's website republish four of its seasonal how-to articles. The written agreement names the four pieces, rules out edits that change their meaning, and asks for a byline, an attribution line with a link back and tagged links. The magazine's system can add a canonical to its standard article template but not to its seasonal hub, so three copies carry a cross-domain canonical to the originals and the fourth, which sits in the hub, carries a noindex instead. Over ten weeks the tagged links bring 1,240 visits and 86 newsletter sign-ups, and all four originals keep appearing for their main queries. In week four one copy starts appearing in place of its original: the magazine has added a sidebar of reader tips that makes its page differ from the original. Fernlattice asks for the tips to move to a separate article, and the original returns. The same month a seed supplier asks Fernlattice to post the supplier's own article on soil testing, unedited, on the Fernlattice blog. Fernlattice declines, and offers instead to interview the supplier's grower for a piece written for its own customers, with the supplier's article cited and linked.
When a partner's copy already appears in place of the original, the diagnosis for a syndicated copy outranking the original confirms which version search prefers and traces it to timing, an edited copy, an unreadable directive or weak internal links.
Common mistakes
- Syndicating before the original is live. The copy becomes the first version anyone sees. Publish on your own site first.
- A canonical on a copy that is no longer a duplicate. Heavy edits, extra sections or a new title can make the canonical a hint that is not followed; use a noindex for edited copies.
- Canonical and noindex on the same copy. Google documents the combination as contradictory; choose one.
- Blocking the copy in robots.txt instead of using noindex. A crawler that cannot fetch the page cannot read its noindex, and the block alone does not reliably keep a linked copy out of the index.
- Making a followed link a condition of the deal. Requiring a link that may not be qualified is one of the link spam examples; ask for credit and leave the attribute to the partner.
- Republishing other people's articles unchanged. A copy with nothing added and nothing pointing at the original is the pattern the scraping policy describes.
- Counting copies as coverage. Copies on many sites measure nothing; referred readers and what they did next do.
- No end date and no correction route. Copies outlive the facts in them; agree how corrections reach the copy and when it comes down.
