Ask, Compare & Choose: The New Rules of Channel Mix
There was a time when asking for a recommendation meant asking a friend who somehow has an opinion on everything. Now? It has competition.
There was a time when asking for a recommendation meant asking a friend who somehow has an opinion on everything.
Now? It has competition.
People are increasingly turning to AI to plan trips, compare products, shortlist brands, research financial products, pick outfits, and make decisions faster than ever before. What used to take five tabs, three review sites, one Reddit spiral, and a mild existential crisis can now happen inside one AI conversation.
That is the real shift marketers need to pay attention to. AI is changing how people discover, compare, decide, and buy. And when the consumer journey changes, the channel mix cannot stay the same.
In our recent Performance+ webinar, Rethinking Channel Mix in the AI Era, Nachiket Desai, our Client Partner and Eliz Ng, Senior Growth Manager from Moloco unpacked what this means for marketers, and where brands need to rethink their approach across discovery, media, consideration, and conversion.
AI is rewriting the customer journey
“The biggest AI disruption is not just technological. It is behavioral.”
Eliz Ng, Senior Growth Manager at Moloco
The traditional “search, click, browse, convert” journey is weakening. If people are getting their answers directly inside AI tools, brands may have fewer opportunities to influence the journey through traditional search and website visits alone.
And this is where channel mix needs a rethink.
Because if fewer people are clicking, if discovery is happening inside AI answers, and if decisions are being influenced before someone even lands on your site, then brands need to start asking a new set of questions.
- Are we showing up in AI answers?
- Are we being recommended?
- Are we present in the sources AI tools trust?
- Are we building enough proof beyond our own website?
- Are we still reaching consumers in the places where they actually spend time?
GEO: from being searchable to being citable
SEO is still important. Nobody needs to throw their keyword strategy into the sea.
But SEO alone is no longer enough.
With the rise of AI-led discovery, brands now need to think about Generative Engine Optimization. In simple terms, this means optimizing your brand so AI tools can understand, extract, cite, and recommend you.
Traditional search was largely about ranking. GEO is about being part of the answer.
That is a huge difference. AI tools look for consistency, credibility, context, and validation across multiple sources. Your website matters, but so do reviews, directories, third-party mentions, product pages, articles, FAQs, social profiles, comparison sites, and customer experiences.
The goal today is to be useful enough, credible enough, and clear enough to be included in the answer.
That means brands need to create content that is easy for both humans and machines to understand. Short answers. Clear explanations. Well-structured pages. FAQs. Product information. Comparison-friendly content. Consistent descriptions across platforms.
It also means brands need to pay attention to what is being said about them outside their own walls.
Because if your website says one thing, reviews say another, and third-party sources say something else entirely, AI tools will see the contradiction. And so will your potential customers. Basically, your brand needs to become the kind of source AI would happily bring to a dinner party.
In-app advertising: growth is hiding in everyday moments
As AI puts pressure on traditional discovery channels, marketers need to look at where consumer attention is still strong.
One of the biggest opportunities? In-app advertising.
People spend most of their mobile time inside apps, not browsers. Yet many brands still over-index on the usual suspects: search, social, and a handful of walled gardens.
The problem is that consumers are not living their digital lives inside one or two neat categories. A fintech user is not only hanging out in finance apps. A high-value shopper is not only browsing shopping apps. A travel customer is not only in travel apps.
People move across weather apps, gaming apps, fitness apps, education apps, recipe apps, utility apps, entertainment apps, and dozens of tiny digital moments throughout the day.
This is where traditional audience assumptions start to break.
A banking customer might check the weather before a flight. A frequent traveller might use a currency converter. A fitness enthusiast might browse meal-planning apps. A high-value ecommerce user might come from a gaming app.
Humans are messy like that. Very inconvenient for media plans. Very useful if you know how to read the signals. This is why the independent app ecosystem matters. It gives marketers access to high-value users beyond the biggest platforms, across the many apps people actually use every day.
But reaching people across thousands of apps is not easy. It is a needle-in-a-haystack problem. This is where AI becomes genuinely useful. Not in the vague “we use AI” way that appears on every marketing deck now, but in the specific, performance-driven way: identifying which user, in which context, at which moment, is worth bidding for.
First-party data is a training model
First-party data gets talked about a lot.
But in the AI era, first-party data becomes much more valuable when brands stop treating it as a static audience list and start treating it as training data for AI models.
Every meaningful customer action tells you something.
Installs. App opens. Registrations. Purchases. Deposits. Repeat purchases. Booking frequency. Drop-off points. Revenue events. Loyalty behavior.
These signals help AI models understand what valuable users actually look like for your business. Not in theory. In behavior.
The key is to map the journey properly. The easiest event to track is not always the most important one. A registration may matter, but a repeat purchase may matter more. An app open may be useful, but a deposit or booking may be a stronger signal of business value.
When brands feed the right signals into their media and AI systems, acquisition becomes more intelligent. Campaigns can optimize toward users who are more likely to convert, stay, spend, and return.
Product comparisons: persuasion is giving way to proof
Earlier, a brand could control a lot more of the journey. Now, the moment someone becomes interested, they can ask AI to compare every option in the category.
Best app for this. Best credit card for that. Best skincare product under this budget. Best travel platform for families. Best food delivery app with the lowest fees.
Suddenly, your brand is not being judged by a full body of evidence. Reviews. Ratings. Third-party comparisons. Customer testimonials. Product details. Pricing. Availability. Expert mentions. Social proof. Complaints. Reddit threads. App store reviews. The lot.
This means brands need to stop thinking only in terms of persuasion and start thinking in terms of proof.
- What evidence exists about your brand?
- Is it easy to find?
- Is it consistent?
- Is it credible?
- Is it structured in a way that AI tools can extract and summarize?
If your strongest claims live only on your website, that is a problem. AI-led comparison pulls from a broader ecosystem. So your proof needs to live across the places where consumers, and AI tools, are looking. This also means product content needs to become more machine-readable. Messy, thin, vague, or inconsistent information can get ignored, misread, or misrepresented.
Agentic commerce: when AI does more than recommend
The next shift is even bigger. A growing number of consumers are becoming comfortable with AI making purchases on their behalf.
That is agentic commerce changing the conversion moment.
Instead of a simple click-to-buy journey, we move closer to a data-to-decision journey. An AI agent may compare products based on price, stock, availability, reviews, delivery time, product attributes, return policies, and past customer satisfaction before deciding what to recommend or purchase.
That means retail media also needs to evolve. Paid placement will still matter, but it will not be the whole game. Brands will need to optimize for the signals AI agents care about: clean product data, competitive pricing, availability, ratings, reviews, fulfillment, and customer experience.
ROAS will still matter too, of course. Marketers are not about to stop caring about performance metrics. But the brands that win will also look at repeat behavior, lifetime value, retention, and whether the product actually delivers on its promise. Because in a world where AI can compare everything instantly, weak experiences become very easy to spot.
In Summary
The AI era does not mean throwing away the old playbook. It means updating it before the customer journey moves too far ahead.
Brands need to keep their SEO foundations strong, but start building for GEO. They need to understand how they appear inside AI answers. They need to diversify their channel mix beyond the usual platforms and follow consumer attention into the app ecosystem. They need to turn first-party data into a strategic advantage, not just a CRM asset. They need to build proof across the web, because product comparisons are becoming more powerful. And they need to prepare for agentic commerce, where recommendability may matter as much as visibility.
The funnel in short is getting compressed, summarized, automated, and occasionally bossed around by a chatbot with excellent patience.
The brands that win will be the ones that are easy to find, easy to understand, easy to trust, easy to recommend, and present in the moments where customers are actually spending their time.
In other words, the channel mix needs a rethink. And the sooner brands start, the better.
Talk to Us
