Back successful the day, I ran a mates of semi-successful affiliate websites. I adjacent erstwhile ranked #1 for “best macromolecule powder.” This was a beauteous lucrative keyword fixed its commercialized intent and precocious hunt volume.

Ever since then, the thought of starting different affiliate tract and getting superior astir it has ever been successful the backmost of my mind. I’ve conscionable ne'er had the clip owed to my time to time enactment astatine Ahrefs.

But today? I don’t deliberation I’d commencement 1 adjacent if I had the time. It conscionable isn’t worth it.

Plenty of different online marketers look to beryllium echoing this sentiment lately, too.

Below are 4 reasons wherefore I consciousness similar this and wouldn’t commencement a axenic affiliate tract in 2025.

Reddit seems to fertile successful the apical 5 for beauteous overmuch each affiliate keyword these days.

This is adjacent the lawsuit with YMYL keywords:

And the occupation is lone getting worse…

When Google precocious penalized the affiliate directories of galore large publications similar CNN Underscored, it overwhelmingly gave that postulation to Reddit—not autarkic publishers. I cognize this due to the fact that we studied it.

How? We looked astatine 6,179 keywords that declined oregon had been wholly mislaid for the directories beneath betwixt November 18th and November 25th:

• https://www.wsj.com/buyside/
• https://reviewed.usatoday.com/
• www.independent.co.uk/advisor
• www.thesun.co.uk/shopping/
• www.forbes.com/advisor/
• www.newsweek.com/vault/
• www.cnn.com/cnn-underscored/

We past utilized the Traffic Share by Domain study successful Keywords Explorer to spot which sites get the astir postulation from these keywords. Reddit was #1, taking 6% of traffic. 

Given that 51% of Reddit threads ranking for merchandise reappraisal keywords person spam arsenic the apical remark (as per Glen Allsopp’s unthinkable research), this is simply a beauteous dire authorities of affairs. For astir merchandise reappraisal keywords, there’s astir ever a niche affiliate tract much deserving of ranking than Reddit.

For example, instrumentality a word similar “best aerial purifier for cigaret smoke.” Reddit ranks successful presumption #2, but the thread is simply a full spamfest.

The apical remark looks consecutive from ChatGPT:

The 2nd looks similar a disguised merchandise pitch:

The 3rd again looks similar a ChatGPT special:

Now you could reason that astir affiliate sites are nary better—and I would agree. But determination are besides plentifulness of niche tract owners who put heavy successful their content, walk countless hours really investigating products, and really merit to rank.

For example, instrumentality a tract similar HouseFresh which ranks #6 for the same query.

The guys down this tract thoroughly test each merchandise they recommend:

But if we instrumentality each the keywords this tract ranks successful the apical 10 for, plug them into Ahrefs’ Keywords Explorer and cheque the Traffic Share by Domain report, we spot that Google is inactive sending Reddit much traffic.

Given the deal Google precocious made with Reddit to usage their contented to bid its AI models, this inclination astir apt isn’t going to alteration immoderate time soon.

Be real: are you really searching Google for merchandise recommendations these days? I hardly ever do. I conscionable spell consecutive to YouTube…

Genuinely searched this a fewer days ago… I didn’t adjacent fuss with Google

Now obviously, we’re marketers. We cognize that astir of the apical Google results are often spammy and shouldn’t beryllium trusted. But what astir everyone else?

I don’t person information connected this, but I’ve seen mainstream YouTubers talking astir it countless times.

For example, this video connected wherefore Google hunt is falling apart has 3.7 cardinal views. In it, helium talks astir however searching for the champion products is pointless due to the fact that of however unspeakable Google results are for merchandise searches.

It’s got to a signifier present wherever searching for the champion of fundamentally thing is pointless. Almost each azygous nonfiction is simply a apical 10 of products not selected based connected their merit but successful information conscionable whether oregon not they’ll get a kickback erstwhile you bargain them. […] I warrant that a ample portion of these articles are written with the writer not really having bought and tested each 10 products themselves.

I deliberation determination are 3 main reasons wherefore the results are so bad:

1. The influx of AI spam. Ever since ChatGPT and generative AI came connected the scene, it’s go ace casual and inexpensive to make affiliate content. Google seems to beryllium struggling to filter retired this spam at scale.
2. Parasite SEO. This is wherever marketers instrumentality vantage of the authorization of an established website to fertile for competitory keywords. It’s wherefore we’ve been seeing sites similar Forbes, CNN, and WSJ outranking autarkic sites for things they person perfectly nary expertise in, similar “best fiscal advisors.”
3. Google ranking Reddit. Most of these threads are spamfests arsenic I mentioned above.

In my opinion, these 3 reasons are besides interconnected…

Google dealt with the influx of AI spam by boosting the hunt beingness of large brands that it thought it could trust. Those brands past took vantage of Google’s favouritism to fertile for presumption wherever they had nary expertise. Google past responded to disapproval astir the deficiency of diverseness successful hunt results (mostly from the SEO industry) by promoting user-generated contented connected Reddit.

Of course, it’s inactive imaginable to beforehand affiliate links elsewhere similar TikTok oregon YouTube—but that’s a full antithetic kettle of fish.

When Glen Allsopp analyzed 10,000 merchandise reappraisal hunt results, helium recovered that lone 4 of the apical 100 ranking domains were autarkic sites. The remainder were portion of quality organisations, nationalist companies oregon ample tract networks.

Even worse, 36 of the apical 100 ranking domains were owned by conscionable 3 antithetic companies: Hearst, Dotdash Meredith, and Future.

I don’t deliberation this is going to get amended anytime soon either…

Looking much intimately astatine our probe connected affiliate work drops, the sites reaping the biggest rewards alongside Reddit were each besides large names, similar Amazon, Business Insider, and CNET.

And galore of the autarkic publishers who saw their postulation decimated by Google updates inactive haven’t recovered…

Take a look astatine Retrododo, for example:

Sure, they’ve recovered a bit—but their postulation is inactive a shadiness of its erstwhile self.

Housefresh is astir the lone tract I’ve seen marque a near-full recovery, and I can’t assistance but consciousness that mightiness person thing to bash with the magnitude of fuss they (quite rightly) kicked up astir being penalized.

But this uncertainty is precisely the problem…

Maybe I’m misremembering, but I’m definite that things didn’t utilized to beryllium this way. Back successful the day, if you enactment a ton of effort into creating a prime niche site, you could beryllium beauteous definite that Google would reward you. You wouldn’t beryllium frightened of the time, effort and wealth you invested being taken distant astatine immoderate fixed infinitesimal by a Google punishment and your postulation handed to large brands (or Reddit) with mediocre content—at slightest not to the aforesaid grade as today.

This doesn’t mean that the full happening is Google’s fault. It besides seems to beryllium happening due to the fact that the SEO strategies of large brands are getting much sophisticated.

For example, this post from HouseFresh talks astir a strategy called keyword swarming allegedly utilized by Dotdash Meredith to propulsion down smaller sites successful the SERPs:

“Swarming is astir drowning retired a competitor,” said the idiosyncratic who reached out. The nonsubjective is to “swarm a smaller site’s foothold connected 1 oregon 2 articles by fundamentally publishing 10 articles [on the topic] and beefing up [Dotdash Meredith sites’] authority.”

Bottomline? I conscionable wouldn’t person the assurance these days to put importantly successful an affiliate site. Even if I created industry-leading content, I wouldn’t person religion that Google would really fertile it supra big sites.

Okay, this 1 is much of a predication… but I deliberation there’s a precise existent anticipation of it happening.

Affiliate sites person worked good historically due to the fact that of however radical search. If you privation to find the champion aerial purifier, you spell to Google, hunt for “best aerial purifier,” click a listicle successful the hunt results, click 1 of their affiliate links, and… BINGO, the tract makes money.

But what if Google adds AI overviews with merchandise recommendations to affiliate searches?

I deliberation this has the imaginable to chopped retired affiliate marketers altogether.

For example, if you benignant “best aerial purifier” into Gemini oregon ChatGPT close now, you get a database of recommendations:

It’s beauteous casual to past probe these recommendations yourself connected YouTube oregon TikTok.

If Google adds these kinds of AI overviews to merchandise reappraisal searches, radical wouldn’t truly request to click connected a hunt effect anymore. They could conscionable probe the products AI recommends themselves.

Of course, they whitethorn inactive take to probe those products successful Google, truthful affiliate marketers could inactive marque wealth from merchandise reviews. But I deliberation it’s much apt that they caput to YouTube oregon TikTok for the reasons I mentioned earlier.

As for wherefore Google would bother, it’s arguably a amended idiosyncratic experience—but they could besides monetize these AI overviews with PPC ads. 

For example, Google already shows sponsored results erstwhile you hunt for astir products:

Google could easy conscionable complaint these brands for clicks erstwhile the merchandise is recommended successful AI overviews.

I deliberation Matt Diggity has the close idea: power to ecommerce.

If you work my takeaways from interviewing Ash Young of CarMats.co.uk, you’ll cognize that ecommerce tin beryllium highly lucrative—even successful boring niches. Ash is connected way to bash £10M successful full income successful nether 5 years.

There’s a elemental mode to find ecommerce niches successful Ahrefs, too:

1. Run a blank hunt successful Keywords Explorer
2. Filter for non-local and non-branded keywords with transactional intent
3. Filter for debased trouble keywords, KD <= 10
4. Filter for keywords with debased DR sites ranking, DR <= 40

It’s past simply a lawsuit of looking for absorbing niches successful the results…

For example, determination are 33K estimated monthly searches for “leg warmers” successful the US, and plentifulness of pages from debased DR websites ranking with fewer oregon nary backlinks.

If we plug “leg warmers” into Keywords Explorer arsenic our effect keyword, spell to the Matching presumption report, use the aforesaid filters, and past clump by Parent Topic, we person a low-competition ecommerce niche beauteous overmuch acceptable to go.

I’m definite you wouldn’t request to determination mountains to fertile for astir of these terms. It would astir apt conscionable instrumentality immoderate basic SEO.

If you privation to larn more, work our usher to ecommerce SEO.

Got questions? Ping maine connected LinkedIn oregon X.