The New Media Map: Where Smart Ad Budgets Go Next

Part 1: Search is fragmenting

For two decades, “search” meant typing keywords into Google. Today the same intent is spread across LLMs, maps apps and online communities. People ask ChatGPT for advice, open Apple Maps when they’re ready to visit, and add “reddit” to a query when they want real opinions. Each of those moments now has a paid layer, an organic layer, or both. A modern digital marketing strategy has to cover all of them.

ChatGPT: advertising inside the answer

ChatGPT is the fastest-growing new ad channel of 2026, and also the least proven. OpenAI published its ads principles on January 16, 2026 and began running paid ads in the US on February 9. By August 31, the company said ChatGPT Ads had passed a $1 billion annualized revenue run rate, was used by tens of thousands of advertisers, and was available in more than 40 countries.

How the ads work

ChatGPT ads appear as a clearly labeled sponsored card that sits beneath the AI’s answer, not inside it. OpenAI matches ads to the topic of the current conversation, and it has publicly committed that advertising does not influence the answers ChatGPT gives and that conversations are not shared with advertisers (and, as advertisers on the platform, we can confirm). Ads run only for users on the Free tier and the low-cost Go plan. Plus, Pro, Business, Enterprise and Edu users don’t see them, which requires deeper thought and strategy around audience alignment.

Targeting is contextual rather than keyword-based. Instead of bidding on “running shoes,” you’re effectively bidding on conversations about training for a first marathon. The intent signal is richer than a keyword, but it’s also harder to plan against, because there’s no keyword report showing exactly what people typed. A newer format lets brands share product feeds with OpenAI so relevant products can surface in shopping conversations.

What it costs

The price of entry has fallen fast. At launch, OpenAI priced ads at roughly a $60 CPM with commitments around $200,000. In April it launched a self-serve Ads Manager and cut the minimum to $50,000. In May, it added cost-per-click bidding, a measurement pixel and a Conversions API. Ad-tech partners have pushed entry costs lower still.

Why marketers are paying attention

ChatGPT is used by nearly three-quarters of the roughly 153 million US adults who use generative AI, and it accounts for about 70% of AI assistant traffic, according to figures EMARKETER compiled in June. Conversations also reveal intent in a way keywords rarely do. Someone who writes “I have a 12-year-old Honda with 150,000 miles, should I fix the transmission or buy something new?” has told you their budget pressure, timeline and decision stage in one message.

The caveats

Measurement is the weak point. Early advertisers got little beyond impressions. Click pricing, the pixel and a LiveRamp partnership are meant to fix that, but the tooling is months old. Trust is the other risk: a January 2026 Ipsos survey found 63% of US adults said ads in AI search results would make them trust the output less. And EMARKETER expects ads in AI chatbots to stay under $1 billion in total 2026 revenue, a rounding error next to search.

How to approach it

• Buy on clicks where you can. CPC pricing gives you something to compare against your search benchmarks.
• Craft careful copy. An ad under an answer competes with the answer. Lead with a specific, useful message, not a brand slogan.
• Invest in organic visibility. What ChatGPT says about your brand in its unpaid answers probably matters more than any ad today. Earned media clips feed LLMs as credible sources about your brand, making public relations efforts more important than ever.
• Treat it as learning budget. The goal of a test is to understand the audience and general performance, not to hit a scaled ROAS target. Learn from the campaign before deciding if the channel stays in the mix.

Google AI Mode: Google’s answer to the answer engine

ChatGPT isn’t the only answer engine selling ads. Google has been showing ads in AI Overviews for some time, and at Google Marketing Live in May 2026 it introduced new ad formats for AI Mode, its conversational search experience.

• Conversational Discovery ads use Gemini to write ad copy tailored to the user’s prompt, drawing on a brand’s own site and assets.
• Highlighted Answers place offers inside AI-generated recommendation lists.
• AI-powered Shopping ads add short explanations of why a product fits what the shopper asked for.

A tool called AI Brief lets advertisers emphasize or remove phrases in the AI-written copy. Google is also testing a Business Agent for Leads, which lets people ask a brand’s AI agent questions directly in search.

The practical difference from ChatGPT is access. You don’t buy AI Mode as a separate channel. Eligibility runs through Performance Max, AI Max for Search, Shopping and broad match campaigns you may already be running. That makes it easy to start but hard to isolate. If you want to know how AI Mode performs, push for placement-level reporting and watch your search term data closely. And because Gemini writes the ad from your materials, clean product data and clear site content matter more than ever.

Apple Maps: paid placement at the moment of the visit

Apple Maps is the most local channel on this list, and for businesses with physical locations it may be the easiest win. After thirteen years without ads, Apple Maps began showing sponsored listings to iPhone and iPad users in the US and Canada in late August 2026, with the rollout ramping through the following weeks.

The move had been coming for a while. Apple rebranded “Apple Search Ads” to simply “Apple Ads” in April 2025, a clear sign it planned to sell inventory beyond the App Store. In March 2026 it announced Apple Business, a single platform that replaced Business Connect, Business Manager and Business Essentials, and confirmed Maps ads would be part of it. Adweek reported that Apple’s ad business was on pace for about $8.5 billion in 2026.

How the ads work

There are two placements. Before someone types anything, a sponsored business can appear at the top of Suggested Places. After they search, a relevant sponsored listing can appear at the top of the results, with only one ad per search. The ads look like ordinary listings with a small blue “Ad” badge, and users can’t turn them off.

Advertisers bid in an auction and pay for outcomes such as views or taps. Any business with a physical location and a claimed Apple Maps listing is eligible. Campaigns can be built through Apple Business with largely automated creative, or through Apple Ads for more control over scheduling and geo-targeting. Apple also offered an introductory 15% credit toward future spend, on up to $1,000 a month during the first year.

Privacy is the defining difference from Google Maps. Apple says a user’s location and ad interactions aren’t tied to their Apple Account, and no data is shared with third parties. Targeting relies on the search, approximate location and map view, not a behavioral profile. Some categories are off limits: home services, bail bonds and crypto ATMs can’t advertise in Maps, and medical services are reviewed case by case.

Why it matters

Every iPhone ships with Apple Maps as the default, and Siri and CarPlay route through it. Local searches are some of the most commercially valuable there are. Someone searching “tire shop” or “sushi” is often deciding on their final destination in minutes. For restaurants, retailers, auto services, gyms, banks and hotels, Maps ads add a second local-search channel while auctions are still thin.

The caveats and how to approach it

The ad load is light by design, and reporting is limited by Apple’s privacy model.

• Audit your listings first. Claim every location in Apple Business and fix hours, categories, photos and phone numbers. The ad pulls from the listing.
• Start with visit-driven categories such as food, auto, retail and fitness.
• Measure with visits and calls. Use store-visit studies and call tracking rather than click attribution.

Reddit: where people search for real opinions

Reddit belongs in a conversation about search, not just social. It has become the place people go when they don’t trust polished marketing content, and that behavior shows up in two ways.

People search Reddit directly. Many people add “reddit” to Google queries such as “best running shoes reddit” to find candid opinions from real users. Reddit has also built its own AI search, Reddit Answers, which summarizes community discussions in response to questions.

AI answers draw on Reddit. Both Google and OpenAI have struck data deals with Reddit, and AI tools frequently cite Reddit threads when recommending products or services — on Google, Reddit content accounts for over 21% of all AI Overview citations. A thread comparing three brands of mattresses, or describing a bad experience with a local contractor, can shape what an answer engine tells the next person who asks.

That makes Reddit an organic visibility priority first. Monitor the communities where your category is discussed, answer questions honestly from a clearly identified brand account, and fix real problems people raise. Heavy-handed promotion backfires quickly — Reddit users are quick to call it out.

Paid ads come second. Reddit’s ad products work more like social ads than search ads, but they let you reach people mid-research by targeting specific communities and conversation topics. Promoted posts that read like genuinely useful contributions tend to perform best. For considered purchases, such as tech, finance, travel, healthcare and home services, Reddit ads are a sensible way to show up where people are already comparing options.