Research Part of the listicle link building guide
Are listicle links dofollow? We checked 5,946 product links
We crawled 547 'best X' software articles. Only 53% of listed products got a followed, direct link, and 57% of product links were nofollow or sponsored.
Short answer
About half the time. In 332 'best X' software lists we crawled on 17 October 2026, 53% of listed products got at least one followed, direct link. The rest got only nofollow, sponsored or affiliate-tracked links. Vendor blogs gave a followed, direct link to 70% of products, publishers to 39%, review sites to 19%. And 39% of lists linked none of the products they named.
On this page
About half the time. We crawled 547 “best X” software articles on 17 October 2026 and found that only 53% of the products they list get at least one followed link that goes straight to the product’s site. The rest get links marked nofollow or sponsored, or links that pass through an affiliate redirect first. Who publishes the list matters more than anything else we measured.
This is the first study from our listicle panel, a fixed set of 584 “best X” articles that we will crawl again on a schedule. If you pitch lists for links, as our listicle link building guide suggests, this tells you what kind of link to expect before you send the first email.
Key findings
- 53% of listed products got at least one followed, direct link (1,347 of 2,522 products on 332 lists; 95% interval 39% to 68%). Link to this
- 57% of all product links carried
nofolloworsponsored(3,415 of 5,946). None usedugc. Link to this - 44% of product links were affiliate-tracked (2,645 of 5,946), and 96% of those went through the publisher’s own redirect, such as a /go/ link. Link to this
- 70% vs 19%: vendor blogs gave a followed, direct link to 70% of the products they listed; publishers to 39%; review sites to 19%. Link to this
- 63% of product links on publisher sites were marked
sponsored(673 of 1,067). Link to this - 39% of the lists we could read linked none of the products they named (215 of 547). Link to this
Are listicle links dofollow?
Sometimes, and you can usually tell in advance. Across the 332 lists where we could find product links, a listed product had a 53% chance of getting at least one followed link straight to its own domain. “Followed” here means the link had no nofollow, sponsored or ugc in its rel attribute. “Direct” means it did not go through an affiliate network or a redirect on the publisher’s site. If you want a refresher on what each attribute does, see dofollow vs nofollow backlinks.
The other 47% split three ways. 27% of products got only nofollow links, 15% only sponsored links, and 5% got a followed link that ran through an affiliate redirect. A followed redirect is a grey area: the link on the page carries no attribute, but it points at the publisher’s own /go/ URL, not at your site, so it is unlikely to pass much to your domain.
| Followed, direct | Followed, affiliate-tracked | nofollow | sponsored | n | |
|---|---|---|---|---|---|
| All lists | 53% (1,347) | 5% (120) | 27% (687) | 15% (368) | 2,522 |
| Vendor blogs | 70% (1,063) | 1% (15) | 29% (435) | 0% (0) | 1,513 |
| Publishers | 39% (183) | 1% (3) | 11% (52) | 49% (229) | 467 |
| Review sites | 19% (101) | 19% (102) | 37% (200) | 26% (139) | 542 |
The 95% interval on the 53% is wide, 39% to 68%, and that is a finding too. Pages on the same site almost always share one link policy, so the real unit of variation is the site, not the page. We have 54 sites with product links in the sample. A different set of sites could move the headline by more than ten points either way. The split by site type is far more stable than the overall number.
Which sites give followed links, and which don’t?
Vendor blogs are software companies writing lists in their own category: Zapier, HubSpot, ClickUp, Salesforce and 30 more in our sample. They gave a followed, direct link to 70% of the products they listed (1,063 of 1,513; interval 53% to 85%). None used sponsored, and only 6% of their product links were affiliate-tracked. When they did withhold, they used nofollow: 29% of their listed products got only that.
Publishers are media sites such as ZDNET, CNET, PCMag, Wired and Tom’s Guide. They gave a followed, direct link to 39% of listed products (183 of 467; interval 23% to 62%). Their product links mostly run through commerce redirects, and they label them: 63% of publisher product links carried sponsored, which is what Google asks for on paid and affiliate links.
Review and comparison sites such as Cloudwards, Website Planet, Email Vendor Selection and TechnologyAdvice were the least likely to give a followed, direct link: 19% of listed products (101 of 542; interval 5% to 46%). 76% of their product links were affiliate-tracked, and most of those went through a redirect on their own domain marked nofollow or sponsored.
| Followed, direct | Followed, affiliate-tracked | nofollow | sponsored | n | |
|---|---|---|---|---|---|
| All lists | 36% (2,135) | 7% (396) | 37% (2,200) | 20% (1,215) | 5,946 |
| Vendor blogs | 68% (1,660) | 1% (30) | 30% (740) | 0% (0) | 2,430 |
| Publishers | 25% (265) | 0% (5) | 12% (124) | 63% (673) | 1,067 |
| Review sites | 9% (210) | 15% (361) | 55% (1,336) | 22% (542) | 2,449 |
Counting every link instead of every product makes the picture a little worse, because the “Visit site” buttons and logo links that sit next to a product’s name are the ones most often tagged. 36% of all product links were followed and direct, 37% were nofollow, 20% sponsored and 7% followed but affiliate-tracked.
How common are affiliate links in “best X” articles?
44% of product links in the panel were affiliate-tracked (2,645 of 5,946; interval 24% to 58%), and this is likely an undercount (see the limits below). Almost all of them, 2,537, went through a redirect on the publisher’s own domain: a /go/ or /out/ path, a goto.php script, or a commerce subdomain such as cc.zdnet.com. Only 90 used an affiliate network’s domain in the link itself and 18 an affiliate parameter on a direct link.
Of the affiliate links, 43% were marked sponsored, 42% nofollow and 15% had no attribute at all. A further 108 product links (2%) carried a light referral tag like ?ref= or utm_source naming the publisher. We did not count those as affiliate, since many are plain attribution.
How many lists link to the products at all?
Fewer than you would think. Of the 547 pages we could read, 215 (39%) had no link to any product they named. Vendor blogs were the most likely to skip product links (146 of 334, 44%), often because the list compares the vendor’s own tool with rivals it does not want to send readers to. Some review sites, Capterra and GetApp among them, link products only to their own profile pages. Publishers were the least likely to skip them (34 of 119, 29%).
That matters for anyone pitching lists. A mention with no link is still worth having for buyers and for AI answers that read these pages, but if the link itself is what you need, check the page before you pitch.
Check before you pitch
Open the list, find any product link, and look at two things: the rel attribute, and whether the link goes to the product’s domain or to a /go/-style URL on the publisher’s site. Our nofollow link checker does the first part. If every product on the page is linked through the publisher’s own redirect, expect yours to be too.
Does language or detection method change the answer?
These two cuts were planned before we looked at the results, mainly to check the method rather than to find a story.
- English vs other languages. English pages (310 lists on 50 sites) gave a followed, direct link to 55% of products. Pages in other languages (22 lists on 11 sites, mostly locale editions of vendor blogs) gave one to 36%, but with only 22 lists the interval runs from 19% to 81%, so we would not read anything into the gap.
- How we found the products. Lists with numbered headings (182 lists, Method A below) gave a followed, direct link to 57% of products. Lists without them (150 lists, Method B) gave one to 48%. The intervals overlap heavily (39% to 75% and 30% to 67%). The difference mostly reflects which sites use numbered headings, not the method.
See which lists you're missing from
The free report finds 'best X', alternatives and comparison posts in your niche that leave you out, with the editor for each.
Get your free prospect reportHow we collected and counted this
The panel. We picked 88 hosts that publish “best X software, tools or apps” articles: tech and business publishers, review sites and software vendors’ blogs. From their sitemaps we kept URLs whose last path segment starts with best- and names software, tools, apps or similar, dropped hardware and finance topics, and drew up to 12 per host by a fixed hash. That gave 584 URLs on 67 hosts. This is a convenience panel of well-known sites, not a random sample of the web.
The crawl. One GET per URL on 17 October 2026 between 20:53 and 20:57 UTC, with an identifying user agent, robots.txt checked first, one request per host every 5 seconds, no JavaScript, no cookies and no logins. 547 pages returned 200.
Site types. Each host was labelled publisher, review or comparison site, or vendor blog by hand before any results were computed. Backlinko counts as a vendor blog because Semrush owns it.
Finding the products. A product link is a link to a product the list names. We found the names two ways:
- Method A, numbered headings. Every H2 or H3 that starts with a number gives a product name (the text after the number, cut at the first colon, dash or bracket).
- Method B, for lists without numbered headings. Any H2 or H3 whose name matches the domain of a link on the page, or the slug of a redirect link.
A link counts as a product link if its domain matches a listed name (asana.com for “Asana”, hubspot.com for “HubSpot CRM”), or if it is an affiliate or redirect link whose anchor text, path or embedded destination names the product. An affiliate or redirect link that names no product is assigned to the product whose section it sits in. Links to app stores, share buttons and the publisher’s own pages are never product links. The publisher’s own product on a vendor blog is an internal link, so it is not counted.
Affiliate links. A link is affiliate-tracked if it uses a known affiliate network or link-cloaking domain (Impact, Partnerize, Awin, CJ, ShareASale, Rakuten, PartnerStack, Skimlinks, Sovrn, Amazon short links, Geniuslink and others), a redirect path or subdomain on the publisher’s own site (/go/, /out/, /visit/, /recommends/, goto.php, cc. or link. subdomains), or a strong affiliate parameter (irclickid, aff_id, tap_a, fpr, an Amazon tag and similar).
Rel classes. sponsored if the rel contains sponsored; otherwise nofollow if it contains nofollow; otherwise ugc; otherwise followed. Each product takes the best class among its links on the page, in this order: followed direct, followed affiliate, ugc, nofollow, sponsored.
Intervals. Pages on one site are not independent, so the intervals above come from a bootstrap that resamples whole sites (2,000 draws). We report counts and shares with intervals and run no significance tests.
Quality check. We checked a random 100 product-link decisions against the page by hand: 95 were right (95%, Wilson interval 89% to 98%). Four of the five errors were Google pages matched to a Google product name, such as a Search Central help page counted as a link to Google Keyword Planner. The fifth was a Tableau link on a Salesforce page, which owns Tableau. We also checked a random 50 external links the method had left out on the same pages: 7 (14%) should have counted as product links. Most of those were affiliate links to shops on consumer-product pages, or products whose domain differs from their name (Dragon by Nuance).
The rules, the cuts and the site labels were written down before we tabulated any rel or affiliate results. While checking how pages are built, we added rules for redirect subdomains, goto.php scripts, destination URLs inside redirects and the section rule, and excluded app-store and share links. All of that happened before the outcome counts were run.
What this study can’t tell you
- It describes this panel, not every listicle. We chose the hosts, and we capped each at 12 pages so no single site dominates.
- 36 pages blocked us. Forbes (11 of 12 pages), TechRepublic (12), Fit Small Business (12) and one Fast Company page returned 403. A retry the same day with the same user agent got the same answer, so they are left out. We did not try to get around the blocks. All four are publishers or review sites, which link more conservatively than vendor blogs, so leaving them out probably flatters the followed shares a little. If all 36 had been readable, the share of lists where every product link is followed and direct would sit between 38% and 48% (we measured 42%). One more page returned 404.
- We read the HTML as served, without JavaScript. Product links added by scripts, such as some price widgets, are missing.
- The detector misses some product links. Our check suggests about one product link in seven was missed, mostly affiliate links. So the true affiliate share is probably higher than 44%, and the true followed, direct share a little lower than measured.
- A followed link is not a promise of ranking value. We read the
relattribute only. Google treatsnofollowandsponsoredas hints, and a followed link through a publisher’s redirect may pass nothing at all. We did not test what search engines do with these links. - One snapshot. Lists change. We will re-run the count on every crawl of the panel and update this page.
Disclosure: BacklinkGenie sells link outreach, so we have an interest in lists being worth pitching. The counts above come from the crawl whatever they showed, and the data is public so you can check them.
Get the data
- Per-URL counts (CSV): 584 rows, one per panel URL, with site type, detection method, product and link counts by class. Version 1.0, 85 KB, SHA-256
412c7baaaecea828e97b57b5662b1d9f654e6a6427a83cfcf47ff8febd2dcf1b. - Data dictionary (CSV): what each column means. SHA-256
5f8fed99ce7946c15f7d8a02bfc53ebfcbc5de95399b0a982ccdc3dd13b3a4ca.
Both files are released under CC BY 4.0. They hold counts and URLs only, never publisher HTML or anchor text. To report an error, email support@backlinkgenie.com and we will log the fix in the changelog below.
Cite this study
APA:
Thakkar, N. (2026, November 11). Are listicle links dofollow? We checked 5,946 product links. BacklinkGenie. https://backlinkgenie.com/blog/are-listicle-links-dofollow/
BibTeX:
@misc{backlinkgenie_r1_2026,
author = {Thakkar, Nikunj},
title = {Are listicle links dofollow? We checked 5,946 product links},
year = {2026},
month = nov,
publisher = {BacklinkGenie},
note = {Listicle panel T0, crawled 17 October 2026, dataset v1.0, CC BY 4.0},
url = {https://backlinkgenie.com/blog/are-listicle-links-dofollow/}
}
HTML link to a single finding:
<a href="https://backlinkgenie.com/blog/are-listicle-links-dofollow/#finding-products-followed">53% of products in "best X" lists get a followed, direct link (BacklinkGenie, 2026)</a>
What to do with this
If a followed link is the goal, start with vendor blogs in your category and with smaller, independent lists, and treat big publishers and review sites as visibility, not link equity. Before you pitch, look at how the page already links products, since your link will almost certainly be treated the same way. To build a list of pages worth pitching, start with finding the listicles in your niche.
Your next step
Find the lists where you'd fit
Enter your product's URL and get a free report of 'best X' and alternatives posts in your niche, each with the editor to contact. No card.
Free, no card. You approve every email before it goes out.
What changed
- 11 November 2026: first version, from the panel’s first crawl (17 October 2026). Dataset v1.0.
Questions people ask
Are links in listicles dofollow?
About half the time. In our crawl of 332 'best X' software lists, 53% of listed products got at least one followed link that went straight to the product's site. Vendor blogs were the most likely to give one (70% of products) and review sites the least (19%).
Why do publishers mark product links as sponsored?
Because most of their product links earn a commission. In our sample, publisher product links mostly went through affiliate or commerce redirects, and Google asks sites to mark paid and affiliate links with rel=sponsored. 63% of publisher product links carried sponsored, so a pitch to a big publisher rarely ends in a followed link.
Is a nofollow link from a listicle worthless?
No. It still sends readers who are comparing tools, and a mention in a list that ranks or gets cited by AI assistants has value on its own. It just passes less, or no, ranking signal than a followed link.
How can I check if a listicle link is followed?
Open the page source or use a nofollow checker on the article and look at the rel attribute of the link to your site. Also check whether the link goes straight to your domain or through a redirect such as /go/ or /out/.
Can I download the data?
Yes. The per-URL counts are a free CSV under CC BY 4.0, with a data dictionary that explains every column. Each of the 584 panel URLs has its site type, detection method and link counts by class. The file holds counts and URLs only, with no publisher HTML or anchor text.
Written by
Nikunj Thakkar · Founder, BacklinkGenie
Nikunj builds BacklinkGenie, an outreach agent that pitches editors for links you earn rather than buy. He writes about link outreach, email deliverability and how products end up in the posts people read.
Published · Last updated