6 AI Search Moves for Marketers Who Aren't Mega Brands

Summary: 

Google recently published a roundtable with marketing leaders from Ikea, Debenhams Group, and DoorDash on adapting to AI Search. The advice is sound but built for enterprise budgets and enterprise data sets. This piece translates the same six themes, longer conversational queries, answer engine optimization, structured data, AI-powered ad testing, first-party data, and next-quarter action steps, into what's actually achievable for a $5M-$50M company.

Why this matters: 

If you're a mid-market marketer, you're going to see this same roundtable summarized on a dozen other blogs this month, most of them repeating the stats without translating them. The risk isn't ignoring the trend, longer and more conversational search queries are real and already showing up in your own analytics. The risk is copying enterprise tactics at the wrong scale: chasing an AI Max budget shift before you've proven incrementality, or assuming you need loyalty-program-scale data to personalize anything. Reading this before you brief your team or your client saves you from building a Q4 plan around someone else's balance sheet.

Global Retail Adapting to AI Search

Google recently ran a roundtable with a handful of global retail and marketplace leaders about how they're adapting to AI Search. It's a good conversation, and it's already making the rounds. It's also a conversation between organizations with resources most of us will never have: enterprise loyalty programs with hundreds of millions of members, tens of thousands of brand partners, and ad budgets big enough to absorb a testing quarter that returns messy results.

That doesn't make the panel's insights wrong. It means your response has to look different when you don't have a nine-figure budget behind you. Here's what actually applies at the mid-market level, and where the read needs to change.

1. Longer queries are real, but you don't need enterprise-scale data to answer them

The panel's core observation was consistent: search queries are getting longer and more specific, and that shift is showing up outside AI-native surfaces too, not just inside them.

You don't have a loyalty program with hundreds of millions of members feeding your personalization. You do have the ability to write content that actually answers a specific, long-tail question better than a generic page does. That's an editorial advantage, not a budget one. A mid-market company with a genuinely useful answer to a niche query can out-rank a mega brand still publishing generic category pages.

2. AEO is your existing SEO, not a new function

It's the most useful line from the whole discussion, and it applies just as much at your scale as at theirs. AEO is the same site architecture, structured data, and crawlability fundamentals that have always mattered, now serving a different reader.

For a mid-market team, this is good news. You don't need to build a new function. You need to make sure the SEO fundamentals you may have deprioritized over the last few years are actually in place, because AI systems are less forgiving of a messy site than a human skimming a results page.

3. Structured data matters more when nobody's fixing your feed for you

Part of the panel's advice centered on making sure AI systems can actually read and recommend a product catalog correctly, which at enterprise scale means integrating hundreds of thousands of SKUs. At that size, it's an engineering team's job.

At a smaller company, this usually comes down to whether your product feed, schema markup, and Merchant Center setup are clean and current. It's one of the highest-leverage, lowest-glamour fixes available, well behind content quality and site authority, but nobody on your team is going to sort it out on its own if it's broken.

4. AI-powered ad budgets need proof, not adoption

One panelist cited a 127% jump in store visitation from an AI-powered ad product. Worth being honest about what's behind numbers like that before you repeat them to a client or a CFO. Big brands can absorb a testing budget that returns unclear results for a quarter while they figure out attribution. Most mid-market companies can't.

The panel's own advice supports caution here. One speaker was explicit that incrementality testing, not click-based attribution, is what makes AI ad spend defensible. Copy that part: test in a way that isolates what the automation is actually adding, before you shift a third of your budget the way one of these companies did.

5. Full-funnel personalization looks different without a loyalty program

The panel's full-funnel story leans heavily on first-party loyalty data collected at massive scale. Most mid-market companies don't have anything close to that.

That doesn't mean full-funnel thinking is out of reach. It means your first-party data has to come from somewhere else: CRM history, email engagement, past purchase behavior, even sales conversations. The principle holds, connect what you know about existing customers to what AI search is telling you about new ones, but the data source is smaller and needs to be used more deliberately.

6. What to actually do this quarter

The panel's closing advice was to audit your search architecture for AI-readiness and lean into AI-powered campaigns. Fine advice, sized for a company that can run that audit across dozens of markets at once.

Sized down, it looks like this:

• Pull your current site structure and structured data and check it against basic AEO fundamentals, not a new framework, the same technical SEO checklist that's always mattered

• Pick one AI-powered campaign type and run it against a real incrementality test, not a spend increase

• Identify the one piece of first-party data you have that's underused, and use it to inform one piece of search or content strategy this quarter

The AI Search shift the panel describes is real. The response to it doesn't have to look like theirs. It has to look like something you can actually execute and defend to whoever signs off on your budget.

Frequently Asked Questions:

Sources

• Growth lessons from Ikea, Debenhams, and DoorDash for the AI Search era, Think with Google, August 2026