ChatGPT Ads are now live in India. As of late August 2026, OpenAI began showing sponsored cards to logged-in adults on the Free and Go tiers, and on September 4, it opened a self-serve Ads Manager with a minimum daily budget of ₹725. E-commerce is one of the categories best positioned to benefit — conversational discovery maps naturally onto how people already shop, asking for recommendations, comparing options, narrowing down a decision. But most online sellers in India don’t yet know how ChatGPT Ads for e-commerce actually work, what they cost, or what’s realistic to expect if they test the channel in month one. This piece by Mathew Thomas covers exactly that: what ChatGPT Ads are, how this new form of AI advertising for e-commerce differs from Shopping and catalog ads, what the early data suggests, and how to test the channel without wasting budget.
What ChatGPT Ads Mean for E-commerce Sellers
Understanding ChatGPT Ads for online sellers starts with recognising that this is a fundamentally different discovery mechanism from anything currently in your paid media mix.
How conversational discovery differs from search and social shopping
Search advertising works because a user has already typed intent into a box — “best running shoes under ₹5,000.” Social advertising works by interrupting a scroll with something visually compelling enough to stop for. ChatGPT Ads work through a third mechanism entirely: a user is mid-conversation, working through a decision — “I need running shoes for half marathon training, what should I look for?” — and a sponsored recommendation appears as part of that reasoning process, not alongside it.
This matters for e-commerce specifically because the user arriving at a ChatGPT ad has typically already articulated more context about their need than a typical search query captures. They’re not just searching a category — they’re mid-decision, often having described budget, use case, or preferences in the conversation leading up to the ad.
The context-hint model — no product catalog feed, no keyword bidding
Sellers coming from Google Shopping or Meta Catalog Ads will look for a familiar setup — upload a product feed, let the platform match products to queries automatically. ChatGPT Ads don’t work this way, at least not yet. There’s no catalog feed integration and no keyword bidding in the traditional sense.
Instead, advertisers set up campaigns using “context hints” — signals about what kinds of conversations and topics their ad should be eligible to appear in. It’s closer to defining a topic and audience context than bidding on exact-match keywords. For sellers used to granular keyword-level control, this is a real adjustment, and it’s worth setting that expectation early rather than discovering it mid-campaign.
How ChatGPT Ads Actually Work for Online Stores
The single ad format — sponsored recommendation cards, not display or shopping tiles
There’s currently one primary ad format: a labelled “Sponsored” card that appears beneath a ChatGPT response, distinct from the conversational content itself. In practice, this is what a ChatGPT shopping recommendation looks like today — not a display banner, not a shopping carousel, not a video ad. OpenAI has been explicit that ads stay visually separated from ChatGPT’s actual answers — the company has publicly stated that “answer independence is non-negotiable,” meaning the ad doesn’t influence or blend into the assistant’s actual response.
For e-commerce sellers, this means creative has to work within a fairly constrained format. There’s no room yet for the rich visual merchandising that Meta Catalog Ads or Google Shopping tiles offer. The card format rewards clear, specific product framing over visual spectacle.
What targeting looks like without a product catalog integration (an honest limitation, not spin)
This is worth stating plainly: if your business runs a large product catalog and is used to dynamic, automatically-matched product ads, ChatGPT Ads in their current form will feel limited. Campaigns are built around context hints and targeting parameters set manually, not automated catalog matching. For a seller with thousands of SKUs, that means choosing which products or product categories to actively advertise, rather than letting an algorithm surface the right SKU automatically for each user.
This isn’t a flaw to work around quietly — it’s a genuine constraint of where the platform is today, and any seller (or agency) telling you otherwise isn’t giving you an accurate picture.
ChatGPT Ads vs Google Shopping vs Meta Catalog Ads
Where each format wins
Each of the three channels solves a different part of the funnel, and none of them currently replaces the others for e-commerce:
- Google Shopping wins on high-intent, bottom-of-funnel capture. A user searching a specific product name or model is close to purchase, and Shopping’s catalog-driven, price-comparison format is built exactly for that moment.
- Meta Catalog Ads win on scale and visual discovery. Dynamic retargeting, broad reach, and rich visual creative make Meta the strongest channel for driving volume and re-engaging past visitors.
- ChatGPT Ads currently win on context-rich, consideration-stage discovery. A user working through “what should I look for,” “what’s the difference between X and Y,” or “what would work for my situation” is earlier in the decision but arriving with more explicit context than a typical search query captures.
None of these replaces the others. They intercept the buyer at different points with different strengths, and the honest read right now is that ChatGPT Ads add a new discovery layer rather than displacing an existing one.
Why this is a test-and-learn channel right now, not a replacement
OpenAI’s India rollout launched with a notably cautious footprint: just 50 brands initially, delivered through agency partnerships with WPP and Omnicom, with self-serve access opening only in early September. That’s a deliberately controlled rollout, not a mature, fully-scaled ad platform. Pricing is still going through active discovery — early U.S. CPMs reportedly launched around 3x Meta’s typical CPM before compressing meaningfully within weeks as the market found its level. India, with its price-sensitive advertiser base, is likely to go through a similar compression period.
The practical implication: treat ChatGPT Ads as a test-and-learn allocation this year, not a channel you shift meaningful budget into based on Google Shopping or Meta benchmarks. The economics are still being established in real time.
What’s Working So Far — Early Signals from Live Campaigns
The early performance data on ChatGPT Ads for e-commerce tells a more interesting story than raw click-through rate suggests.
What the early conversion data shows
The clearest signal so far isn’t about volume — it’s about the tradeoff between click-through rate and conversion quality. Early advertiser data puts ChatGPT ad click-through rates meaningfully lower than Google Search — in the range of 0.3% to 1.5%, against Google Search’s typical benchmark north of 6%. On the surface, that looks like a weak channel.
But conversion behaviour tells a different story. Criteo’s early data on large language model referral traffic found that visitors arriving via LLM platforms convert at roughly 1.5 times the rate of other digital channels, and spend 60-80% more time on-site than visitors from social platforms. Separately, a 2026 industry study across 19 verticals found ChatGPT ad conversion rates ranging from 0.2% to 5.8% depending on category, with commercial and considered-purchase verticals performing toward the higher end.
What this suggests about buyer intent inside conversations
Put together, the pattern is consistent: fewer people click, but the ones who do have effectively pre-qualified themselves through the conversation that preceded the ad. A user who’s spent several exchanges narrowing down what they want, then clicks a sponsored recommendation that matches that context, is arriving at your site with meaningfully higher intent than a typical cold click.
This is the honest case for testing ChatGPT Ads as an e-commerce seller in 2026: not high volume, but potentially high-quality, pre-qualified traffic — provided your product and landing experience actually match what the conversation promised.
The Real Constraints E-commerce Sellers Should Know Before Testing
No pixel-level attribution — aggregated impressions/clicks only
ChatGPT Ads Manager currently reports impressions, clicks, spend, CTR, average CPC, average CPM, and conversions at an aggregated campaign level — not the granular, user-level pixel tracking that Meta and Google advertisers are used to. Advertisers can add UTM parameters to landing page URLs to track referred traffic in their own analytics tools, but the attribution model is meaningfully less precise than what e-commerce sellers typically rely on for channel-level ROAS decisions.
Practically, this means ChatGPT-referred traffic may also show up in your analytics as unassigned, direct, or generic referral traffic if UTM tracking isn’t configured carefully — the same “dark traffic” problem that already affects organic ChatGPT citations.
No retargeting, no dynamic product ads yet
There is currently no retargeting capability and no dynamic, catalog-driven product ad format — the features that make Meta and Google Shopping powerful for scaling e-commerce spend efficiently. Every campaign right now is a fresh-context, top-of-conversation placement, not part of a broader always-on retargeting funnel.
Free and Go tier only — audience size context
Ads currently appear only to logged-in adults on ChatGPT’s Free and Go tiers (the latter priced at ₹399/month in India) — paid Plus and Pro subscribers remain ad-free. Given ChatGPT’s reported user base of over 100 million weekly active users in India, a large share on free or low-cost tiers, the addressable audience is substantial. But it’s worth knowing precisely which segment you’re reaching, and that it currently excludes the platform’s highest-paying subscriber tiers.
Which Products and Categories Suit ChatGPT Ads Best
Considered-purchase items vs impulse/commodity products
Given the context-hint targeting model and the consideration-stage user behaviour the early data suggests, ChatGPT Ads appear best suited to products people genuinely deliberate over — items where a buyer would naturally ask a series of clarifying questions before deciding. Electronics, home appliances, fitness and wellness equipment, furniture, and considered fashion or lifestyle purchases fit this pattern well.
Commodity or impulse products — where the purchase decision doesn’t really involve deliberation — are a weaker fit right now. A user isn’t likely to have an extended conversation about buying a phone charger or a basic grocery item, which limits the natural opportunities for a sponsored card to appear organically in that kind of exchange.
A simple self-check: “would someone ask ChatGPT for this recommendation?”
Before allocating budget, it’s worth running your product line through one honest question: would a real customer plausibly ask ChatGPT something like “what’s the best [category] for [use case]” before buying this? If the answer is clearly yes, ChatGPT Ads are worth testing. If the honest answer is no — because the purchase is too low-consideration, too routine, or too price-driven to prompt that kind of conversational research — your budget is very likely better spent on Google Shopping or Meta for now.
How to Test ChatGPT Ads Without Wasting Budget
Starting budget logic from the ₹725/day floor
₹725/day is the platform’s technical minimum, translating to roughly ₹21,750 over a 30-day month if run continuously. For a genuine test — enough data to see a directional signal without overcommitting to an unproven channel — most e-commerce sellers should plan for at least a 2-3 week test window at or slightly above the minimum, rather than the single-day minimum spend, which won’t generate meaningful data on its own.
What a first two-week test should look like
A sensible first test keeps scope narrow and deliberately limited:
- Pick one considered-purchase product line — not your entire catalog — and run the self-check above before committing
- Set up UTM tracking on every landing page URL before the campaign goes live, since native attribution is aggregated, not granular
- Choose context hints deliberately, based on the kinds of questions and conversations a real buyer of that product would plausibly have
- Keep expectations calibrated to the CTR-vs-conversion tradeoff — a lower click-through rate than Google Search is not a failure signal on its own; what matters is whether the clicks you do get convert well
- Compare the resulting CAC against your existing Google Shopping and Meta benchmarks after the test window closes, rather than judging the channel against volume expectations set by more mature platforms
If a seller would rather have this tested and managed by a ChatGPT ads agency in India already tracking the format closely, that’s exactly the kind of early-stage, test-and-learn work we’re set up to handle at Mathew Digital — reach out if you’d rather have ChatGPT Ads management handled than run it in-house while the platform is still this new.
This kind of paid experimentation also connects to a broader shift worth keeping in view: as AI platforms become a genuine discovery layer for e-commerce buyers, the same signals that earn organic visibility in AI Overviews and LLM citations — structured content, clear entity signals, genuine expertise — increasingly matter alongside paid placement. Testing ChatGPT Ads and building AI-search-ready organic content aren’t competing priorities; they’re two sides of the same emerging channel.
Conclusion
ChatGPT advertising for online stores in India is a genuinely new discovery layer for e-commerce, not a replacement for Google Shopping or Meta Catalog Ads. The format is early, the targeting model is unfamiliar, and several features e-commerce sellers rely on elsewhere — retargeting, dynamic catalog ads, granular attribution — aren’t available yet. But the early conversion data suggests a real pattern worth testing: fewer clicks, but potentially higher-intent ones, from users who’ve already worked through part of their decision in conversation before the ad ever appears.
For sellers with considered-purchase products, a modest, well-scoped test — one product line, UTM tracking in place, realistic expectations about CTR versus conversion quality — is a reasonable way to find out early what this channel can actually do for your business, without betting meaningful budget on a platform still finding its price.
Frequently Asked Questions
How is ChatGPT Ads different from other AI advertising for e-commerce?
ChatGPT Ads is currently the most prominent form of AI advertising for e-commerce built directly into a conversational AI platform, but it sits alongside a broader shift where AI systems — Google AI Overviews, Perplexity, Gemini — are all becoming discovery surfaces that influence buyer decisions before a single click happens. What sets ChatGPT Ads apart from this wider AI advertising landscape is that it’s a paid, advertiser-controlled placement inside an active conversation, rather than an organic citation earned through content quality and structured data. E-commerce sellers exploring AI advertising broadly should treat paid channels like ChatGPT Ads and organic AI-search visibility work (AEO and GEO) as complementary parts of the same strategy, not separate initiatives — the buyer behaviour driving both is the same shift toward AI-assisted research and recommendation.
Does ChatGPT Ads support product catalogs like Google Shopping?
Not currently. ChatGPT ads for e-commerce use a context-hint targeting model rather than a catalog feed integration, which means there’s no automated, dynamic matching of individual SKUs to user queries the way Google Shopping or Meta Catalog Ads work. Advertisers manually select which products or product categories to promote and define the conversational context their ads should be eligible to appear in. For sellers with large catalogs, this means choosing a focused set of products to advertise rather than letting the platform surface the right SKU automatically. This is expected to evolve as the platform matures, but as of the September 2026 self-serve launch in India, catalog integration isn’t part of the offering.
Can I retarget users who saw my ChatGPT ad?
No, not at this time. ChatGPT Ads currently don’t offer a retargeting capability — every placement is a fresh-context exposure rather than part of an ongoing remarketing funnel. This is a meaningful difference from Meta and Google, where retargeting is often a core part of an e-commerce advertiser’s strategy for improving overall return on ad spend. Sellers testing ChatGPT Ads should treat it purely as a discovery and consideration-stage channel for now, and continue to rely on Meta or Google for retargeting previously engaged visitors.
How much should an e-commerce brand budget to start testing ChatGPT Ads?
The platform’s technical minimum is ₹725 per day, but a single day at that spend won’t generate meaningful data. A more realistic starting test runs at or slightly above the minimum for two to three weeks — roughly ₹15,000 to ₹25,000 total — focused on a single, considered-purchase product line rather than a full catalog. This gives enough volume to observe a directional signal on click-through rate, conversion behaviour, and cost per acquisition without overcommitting budget to a channel that’s still going through active price discovery in the Indian market.
Is ChatGPT Ads worth it for small online stores?
It depends heavily on what you sell. For small stores selling considered-purchase products — items customers would plausibly research or ask questions about before buying — a modest, well-scoped test is reasonable and the entry cost is low relative to Google or Meta minimums. For stores selling commodity, low-consideration, or highly price-driven products, the current format is a weaker fit, since ChatGPT Ads work best where a real conversational decision-making process precedes the purchase. Small stores should run the simple self-check — would a real customer plausibly ask ChatGPT for a recommendation before buying this? — before allocating any budget here.
How is ChatGPT Ads attribution different from Google Ads or Meta Ads?
ChatGPT Ads Manager currently reports performance at an aggregated level — impressions, clicks, spend, CTR, average CPC, average CPM, and conversions — rather than the granular, user-level pixel tracking that Google Ads and Meta provide. Advertisers can add UTM parameters to landing page URLs to track referred traffic through their own analytics tools, but without careful setup, ChatGPT-referred visits can show up as direct or unassigned traffic rather than being clearly attributed to the campaign. E-commerce sellers testing the channel should set up UTM tracking on every landing page before launching, and treat the reported numbers as directional rather than as precise as what they’re used to from more mature ad platforms.
