WHAT HAPPENED?
Google is redesigning advertising for a version of Search where people ask longer, more complex questions.
At Google Marketing Live 2026, the company introduced new Gemini-powered ad experiences including Conversational Discovery ads and Highlighted Answers, while expanding its Direct Offers pilot.
1. WHAT GOOGLE ACTUALLY CHANGED
Conversational Discovery ads use Gemini to create an ad response around the specific question someone is asking.
Highlighted Answers can surface a sponsored product inside an AI-generated recommendation and explain why that product may fit the shopper’s needs. The placement remains clearly labeled as sponsored.
Google is also expanding Direct Offers.
Conversational Discovery Ads
Highlighted Answers
Expanded Direct Offers
Automated Campaign Foundation
Google says advertisers should build the foundation for these experiences through AI Max for Search, AI Max for Shopping and Performance Max.
That detail matters.
The route into conversational advertising is increasingly connected to automation, product data and broader campaign signals.
2. WHY IT MATTERS FOR ADVERTISERS
Paid search traditionally worked around a relatively simple relationship:
Query → Keyword → Ad
AI Mode makes that relationship more complicated.
Someone may no longer search:
“Carry-on suitcase.”
They could ask:
“What lightweight suitcase would work for a two-week Europe trip, fit strict cabin limits and still have space for my laptop?”
There is far more information inside that request than a single keyword can represent.
Google now has to understand what the shopper is trying to accomplish and which products genuinely fit those conditions.
That makes Merchant Center data, landing-page information, product attributes, brand information and conversion signals increasingly important inputs.
Google has already introduced new Merchant Center attributes specifically designed to help products appear in conversational discovery experiences.
Keywords still matter.
They just have more company..
3. Here’s an Explainer!
Imagine two luggage brands both bidding around -
“carry-on suitcase.”
Brand A has a basic product feed:
Cabin Suitcase — $199
Brand B provides dimensions, weight, laptop compartment size, materials, airline compatibility and detailed landing-page information.
When someone asks AI Mode for a lightweight suitcase that meets a particular airline limit and carries a laptop, Google has much more information available to understand why Brand B may be relevant.
The bidding strategy still matters.
But the quality of the information feeding Google's AI may increasingly determine whether the advertiser is even a strong candidate for the answer.
4. WHAT ADVERTISERS SHOULD DO
Ask whether Google can understand, who each product is for, what problem it solves and what differentiates it.
Second, review landing pages through the same lens.
Thin pages give AI systems fewer useful signals. Detailed specifications, use cases and clearly structured product information become more valuable as searches become more conversational.
Third, do not abandon keywords because AI Mode exists.
Instead, start treating keywords as one layer of the system alongside product data, website content and conversion signals.
The new question is:
Does Google have enough information to understand why our product is the right answer?
ORBIT. TODAY’s TAKE
AI Search is moving toward understanding what someone actually wants.
That does not make the keyword obsolete.
It means advertisers increasingly need to optimize the information behind the keyword.
