La recherche par IA tue-t-elle le marketing d'affiliation ? Je vois mon trafic affilié chuter alors que mes positions restent stables

Let me share data from across my client base (50+ affiliate sites).

The aggregate picture:

Traffic impact:

• Average traffic decline: 28% YoY
• Worst hit: Generic review sites (40-50% decline)
• Best performing: Niche specialists (10-15% decline)

Revenue impact varies more:

• Conversion rates UP for traffic that does come
• Users who click through are more qualified
• Average order value UP 12%

Net revenue:

• Generic sites: Down 30-40%
• Niche authority sites: Down 10-20%
• Some specialists: Actually UP (focused + authority)

What differentiates winners:

1. Original content AI can’t replicate (hands-on testing, proprietary data)
2. Brand recognition (users seek out specifically)
3. Email lists (direct audience, not dependent on search)
4. AI visibility (getting cited in AI responses WITH affiliate links sometimes preserved)

This isn’t the death of affiliate. It’s the death of commodity affiliate content.

I’m one of those specialists still doing okay. Here’s what’s actually working.

My niche: High-end coffee equipment

My approach:

Original testing:

• I buy every product I review
• 30-day real-use testing minimum
• Comparison tests with controlled methodology
• Actual photos and videos (not stock)

Expert positioning:

• SCA certified barista
• 15 years in specialty coffee
• Known in coffee communities
• Guest on coffee podcasts

Direct audience:

• 45,000 email subscribers
• Active YouTube with tutorials
• Private community for enthusiasts

AI impact:

Traffic from Google: Down 22% Direct traffic: Up 18% Email referrals: Up 35% Revenue: Up 8% overall

The calculation:

Less dependent on Google. When AI cites product recommendations in my category, my site often appears. The brand recognition matters.

My advice:

If your only distribution is Google, you’re vulnerable. Build multiple channels.

The ugly reality for commodity affiliates.

What’s dying:

• “Best X 2026” roundup posts with regurgitated specs
• Multi-niche sites covering everything
• Review content based on manufacturer specs only
• Sites with zero brand recognition

Why:

AI can do this work. It can synthesize product specs, summarize Amazon reviews, and recommend products. It doesn’t need to send users to a middleman affiliate site.

What’s surviving:

• Content AI literally cannot create (real testing)
• Brand trust that influences purchase decisions
• Value beyond “which product to buy”
• Direct audience relationships

The hard truth:

If a prompt to ChatGPT could replace your content, AI search will replace your traffic. You need to create value AI can’t replicate.

For many affiliates, that means a complete strategic pivot.

SEO perspective on what’s still working.

Query types still sending affiliate traffic:

Comparison queries: “[Product A] vs [Product B]” - AI gives answers, but users often want more detail

Specific use case: “Best [product] for [very specific need]” - Narrow queries still click through

Problem-solution: “[Product type] for [specific problem]” - Users want validation

Brand queries: “[Your site name] review of [product]” - If you have brand recognition

What’s NOT sending traffic:

“Best [product] 2026” - AI answers directly “Top [product category]” - AI summarizes “[Product] review” - AI synthesizes from multiple sources

The strategy shift:

Focus on queries where users WANT to click through, not just find an answer. Comparison deep-dives, specific use cases, expert analysis.

The AI visibility angle for affiliates.

A different approach:

Instead of fighting AI, optimize to be cited BY AI.

How it works:

When someone asks ChatGPT “What’s the best budget espresso machine?”, it pulls information from authoritative sources. If your content is structured well and has authority signals, you might be cited.

The affiliate link question:

Sometimes AI includes source links. Users who want deeper info click through. Your affiliate content is accessed via AI citation rather than organic ranking.

Monitoring this:

Use Am I Cited to track when your affiliate content is mentioned in AI responses. Understand which content gets cited and for which queries.

The data we’re seeing:

Affiliates with strong authority signals and well-structured content maintain ~30-40% of their “visibility” through AI citations.

It’s not replacing lost traffic 1:1, but it’s a meaningful channel.

The hybrid approach:

1. Optimize for AI citations
2. Build direct audience (email, social)
3. Create un-replicable content
4. Accept some traffic loss, optimize for revenue per visitor

Video affiliate perspective.

YouTube has been a savior:

Written affiliate content is being replaced by AI. But:

• AI can’t create hands-on video reviews
• YouTube is a search engine itself
• Video builds trust for high-consideration purchases
• AI often cites video content in responses

My split:

2023: 80% Google traffic, 20% YouTube 2025: 45% Google, 35% YouTube, 20% direct/social

Revenue:

Actually UP 15% because video converts better. Users who watch a 10-minute review are more likely to buy than those who skim a text review.

The effort reality:

Video is harder. Equipment, editing, on-camera presence. But it’s defensible.

My advice:

If you’re in a category where visuals matter (electronics, home goods, equipment), video is the pivot. AI can’t replace demonstration.

Historical perspective for the panicking affiliates.

This isn’t the first disruption:

• 2011: Panda killed thin content sites
• 2012: Penguin killed link schemes
• 2015: Mobile-first hurt desktop-focused sites
• 2018: Medic update devastated YMYL affiliates
• 2020: Core Web Vitals added technical barriers

Each time, many affiliates failed. Those who adapted survived and often thrived.

What’s different this time:

The fundamental discovery mechanism is changing. Not just algorithm updates - the entire way users find products is shifting.

What’s the same:

Quality, authority, and genuine value still win. The affiliates succeeding now are those who would have succeeded in any era: real expertise, genuine helpfulness, brand trust.

My take:

Commodity affiliate marketing is dying. Expert affiliate marketing is evolving. Figure out which you’re doing and adapt accordingly.

The numbers on what’s working.

Revenue per visitor trends:

Sites seeing traffic decline but improving RPV:

• Fewer visitors but more qualified
• Higher conversion rates
• Larger average orders

The calculation:

Old: 100,000 visitors x 0.5% conversion x $50 AOV = $25,000 New: 65,000 visitors x 0.9% conversion x $55 AOV = $32,175

For some, revenue is actually UP despite traffic decline.

Who this works for:

• Sites with high expertise signals
• Content that pre-qualifies visitors
• Categories with complex purchase decisions

Who it doesn’t work for:

• Low-consideration product categories
• Sites without differentiation
• Generic recommendation content

The optimization:

Focus less on traffic volume. Focus more on visitor quality and conversion optimization. The traffic that comes is more valuable than before.

Where this is heading.

The 2027-2028 affiliate landscape:

1. AI-integrated affiliates - Sites cited by AI, traffic flows through AI
2. Brand affiliates - Direct audience, email lists, community
3. Video/media affiliates - YouTube, podcasts, un-replicable formats
4. Experience affiliates - Real testing, original data, expertise

What won’t exist:

• Generic comparison sites
• Multi-niche aggregators
• Review sites using manufacturer specs

The opportunity:

Less competition. Survivors will have larger share of a smaller pie, but the pie is higher-value traffic.

Investment implications:

• Build brand and audience now
• Create defensible content assets
• Diversify traffic sources
• Accept transition pain for long-term positioning

The affiliates building for 2027 now will dominate. Those waiting for “normal to return” will struggle.

Sobering but clarifying thread.

My honest assessment:

My sites are largely commodity affiliate content. I’ve been good at SEO but haven’t built brand, audience, or truly differentiated content.

What I’m taking away:

1. The model is changing, not dying - But MY current model may be dying
2. AI visibility matters - Need to monitor and optimize for citations
3. Original value is required - Can’t just aggregate anymore
4. Direct audience is essential - Email, community, YouTube
5. Niche down and go deep - Specialists are surviving

My plan:

• Pick one niche where I have genuine interest/expertise
• Start building email list immediately
• Create hands-on testing content
• Set up AI monitoring with Am I Cited
• Accept 12-18 month transition period

The reality check:

10 years of affiliate experience doesn’t entitle me to continued success. The game changed. I need to change too.

Thanks for the honest perspectives.