AI Search vs E‑Commerce Revenue in 2025: A Reality Check

Let’s talk about the elephant in the boardroom: AI search.

You’ve probably heard the terms getting thrown around, ChatGPT traffic, Generative Engine Optimisation (GEO), “AI visibility.”

And sooner or later, someone pipes up:

“Are we missing something?”

“Should we move budget from SEO into AI optimisation?”

Fair questions.

So, we ran the data across real e‑commerce brands, millions in sales, multiple countries, and a mix of categories.

We looked at where traffic comes from, how it performs, and more importantly: what actually drives revenue.

Short version?

AI search traffic is real. It’s growing.

But it’s nowhere near replacing Google or Bing.

The perception gap is wide, wider than most CMOs realise.


1. The Hype vs The Reality

At a recent webinar, we asked marketers how much revenue they thought ChatGPT was driving for big e‑commerce players. Guesses landed in the six-figure range.

One retailer we looked at, millions in annual revenue, well-established brand, made just $2,000 from ChatGPT across six months.

Total. That’s not even one decent day of Google Ads.

Across the whole dataset, AI traffic contributed less than 1% of revenue in nearly every case.

And yet, a LinkedIn poll showed 30-40% of marketers think ChatGPT will overtake Google organic and paid search in six months.

That’s not just optimistic. It’s wrong, by a lot.


2. What We Measured

We looked at brands selling fashion, homewares, lifestyle gear, and niche goods. Some big, some mid-market, some small but high-performing.

Agv Vud6Pt Jd Uzqfq1G0Nwgr8Znjo193Iq5Imyuxnhgehd9 Oh20Crecp6Jm92Pe5Fid5Fbmky Fve2R8 7O4Tf568Bugj33Qmpoptsukyyfzh3M1Fmdmlq3Lul4Tgpq81W4 Kodoow=S2048?Key=Yna3Nnkimw2S Mxzi6Ssw

Metrics we tracked:

  • Traffic: sessions, users, new users
  • Engagement: bounce rate, dwell time, pages per visit
  • Conversions: transactions, conversion rate, revenue per visit, AOV
  • Brand signals: branded search volume, niche size, domain strength

And we tracked sources including:

  • Google Organic
  • Google Ads
  • Bing Organic
  • Bing Ads
  • ChatGPT
  • Perplexity
  • Gemini

Not just what gets traffic, what converts. What makes money.


3. Where the Money Actually Comes From

Here’s what the revenue breakdown looked like:

ChannelRevenueShare of Total
Google Paid$247.5M64.3%
Google Organic$122.9M31.9%
Bing Organic$11.3M2.9%
Bing Paid$3.4M0.9%
ChatGPT (Direct)$44.6K0.012%
Perplexity (Direct)$6.4K0.002%
Gemini (Direct)$3840.0001%

Let’s be clear: 96% of all revenue from search is still coming from Google. Add Bing, and it’s more like 99.9%.

Weirdly, Bing Ads outperformed ChatGPT traffic by 76×, and most brands haven’t even maxed out Bing yet.


4. Why Execs Get Stuck on This

There’s a logic trap that’s easy to fall into. I’ve seen it firsthand in CMO strategy calls:

“If ChatGPT is the future, shouldn’t we invest early?”
“Why scale Bing or Google Ads when AI is the trend?”

Because, right now, one is driving millions in revenue. The other isn’t. It’s that simple. You wouldn’t move your sales team off email because Slack’s having a moment.


5. Conversion Performance: Still No Contest

We broke down conversion rate by channel:

  • Bing Organic: 3.55%
  • Bing Paid: 2.62%
  • Google Paid: 2.59%
  • Google Organic: 2.20%
  • ChatGPT Direct: 1.05%
  • ChatGPT Referral: 1.22%
  • Perplexity Direct: 1.75%
Agv Vucmpwqs N8Bcefwx7T7Rhxu8Xeepfiolqoymi5Ptuwwvaepzmgby3Fsfbt0D0Uyptucmzkwp4Yv9Ebd9Rh9Hjwmoaiazq8B0Zxuvhsdmtqxjmjachm3Arwyzc7Ysiatdbbxiyja W=S2048?Key=Yna3Nnkimw2S Mxzi6Ssw

And revenue per visit:

  • Bing Organic: $9.05
  • Google Paid: $4.99
  • Google Organic: $4.28
  • ChatGPT Direct: $1.93
  • ChatGPT Referral: $2.42
Agv Vufsbx61Adosedaecw9Pmkpjw5N1Qkelemwlstorlpaz5Unknh0Gmszcula4Qhdvdv4Oeuldn3Tqjk2Zspvjvvtfm3Rupw0E9Fchonm Mhyeseo0Olxj0Kz9Gs6Zdt20Myjngmlika=S2048?Key=Yna3Nnkimw2S Mxzi6Ssw

The differences aren’t subtle. Traffic from search engines is more qualified and more likely to convert.

One client quipped that ChatGPT traffic was like “a distracted window shopper who leaves before asking a single question.” Brutal. But fair.


6. Bigger Baskets, Better Buyers

Average Order Value told a similar story:

  • Bing Paid: $463
  • Bing Organic: $382
  • Google Organic: $345
  • ChatGPT Direct: $179
  • ChatGPT Referral: $328
  • Perplexity Referral: $554 (tiny sample though, not scalable)

Yes, Perplexity looks high here, but it’s from a handful of sessions.

You wouldn’t bet your forecast on that.


7. Who’s Actually Engaging?

Search users behave like shoppers. AI users…not so much.

Dwell time: Search channels (especially Bing) show consistent, longer sessions. Gemini stood out too, possibly people playing with it rather than buying.

Pages per session: Higher from search. Deeper browsing, more exploration.

Bounce rate: Lower from search. Higher from AI, people click, skim, leave.

AI tools are still more curiosity engines than checkout drivers.


8. The AI Growth Curve

Yes, there’s growth. A few brands saw ChatGPT referrals go from nothing in late 2024 to thousands per month by mid‑2025.

In some cases, revenue scaled 10× in six months.

Sounds impressive until you realise it’s scaling from $100 to $1,000.

One apparel brand saw a mini spike after being referenced in a Reddit thread and picked up by ChatGPT a week later. Cool moment - but hardly a pattern.

Forecasts suggest sessions might hit 5,000+ monthly in early 2026. Still nowhere near Google.


9. Brand Power ≠ AI Revenue

Across all niches, branded search volume totalled 10.6 million. AI traffic? Just 55,000 sessions.

We ran the numbers. The correlation between branded demand and AI revenue per session was, 0.13, not only weak, but negative.

In other words, being a household name doesn’t help much if the household isn’t asking ChatGPT about you.


10. What AI Might Influence Later

This bit is trickier, and honestly more interesting.

We saw examples where ChatGPT or Perplexity mentions lifted direct and branded traffic within 48-72 hours. Users didn’t click the link but searched the brand later or typed the URL in cold.

This was clearest in high-ticket or considered purchases, luxury goods, specialty items.

So yes, AI referrals might influence buying decisions. Just not in the same session, and not in a way that’s easy to track in your GA4 reports.


11. How Big Is the Gap?

Sessions volume by channel:

  • Google Paid: 63.5M
  • Direct: 48.8M
  • Google Organic: 44.0M
  • AI combined: ~55K

Even in strong niches, AI traffic barely registers. Branded search in fashion? 4.84M. AI search? Basically irrelevant.

That doesn’t mean AI won’t grow. It just means it’s nowhere close, yet.


12. The Forecast: Gradual, Not Sudden

Will AI search overtake traditional search?

Maybe one day. But not this year, and probably not next year either. Our forecasting shows slow, steady growth, not a takeover.

We’re not looking at a flip switch. We’re looking at a long dimmer dial.


13. What It Means for SEO

SEO still means Google.

If you’re chasing visibility and revenue, that’s where your effort should go. Organic and paid together are still the backbone of e‑commerce traffic.

Optimising for AI is worth doing, but it’s a side quest, not the main campaign.


14. What to Tell Your CEO

If your CEO or CMO asks, “Are we falling behind on AI search?” you can say confidently:

  • Not in revenue terms.
  • The brands making real money are still making it from Google and Bing.
  • The shift to AI visibility is real, but slow.
  • Layer it in, don’t flip the strategy.
  • Review every six months with hard data. Not hot takes.

15. Final Word

Look, clicks are fun. Rankings are nice. But revenue is what pays the bills.

Right now, AI search isn’t paying much. Google and Bing are. The smart brands are doubling down on what works, while quietly prepping for what’s next.

If AI traffic explodes in two years, great, you’ll be ready. But don’t sink the boat trying to chase a ripple.


16. StudioHawk’s Real‑World Roadmap

If you’re running SEO in-house, here’s what we recommend:

  1. Track AI sources. GA4 filters for “gpt,” “perplexity,” etc.
  2. Dashboards matter. Keep AI traffic visible alongside Google/Bing.
  3. Product pages need answers. Top-of-page clarity. Use headings and structured data.
  4. Build topical clusters. AI often breaks questions down. Have content ready for the follow-ups.
  5. Schema, merchant feeds, clean product info. Feed the Google Shopping graph.
  6. Mentions matter. Get cited in blogs, Reddit, YouTube, wherever AI models look.
  7. Test AI agents. Try ChatGPT Shopping and Google’s AI mode. Spot friction before AI tools do.

Wrap Up

Stay calm. Keep SEO tight. Don’t waste money chasing shadows.

The brands that win will:

  • Defend their Google and Bing real estate
  • Experiment with AI safely
  • Be ready when the volume arrives

We’ll be watching the data, and updating the strategy, as it does.

Don’t guess what AI search needs, build for it.

Grab the StudioHawk AI SEO Toolkit and get ahead while everyone else is still talking about it.

About the Author

Lawrence Hitches is an AI SEO consultant based in Melbourne and General Manager of StudioHawk, Australia's largest dedicated SEO agency. He specialises in AI search visibility, technical SEO, and organic growth strategy - leading a team of 115+ across Melbourne, Sydney, London, and the US. Book a free consultation →