Here's a scenario that's already happening.

Someone says to their AI assistant: "I need new running shoes. Budget £150. I overpronate. Find me the best option."

The AI doesn't send them ten links to evaluate. It makes a decision. It evaluates hundreds of options based on criteria the human provided, cross-references reviews, checks availability, and recommends one or two products.

The human never visits your website. Never sees your brand. Never compares options manually.

The AI made the choice. And if your brand wasn't selected, you weren't just ranked lower, you were invisible.

This is the agentic era. AI agents making decisions on behalf of humans. And it's not coming, it's here.

The question isn't whether this will change commerce. It's whether your brand is ready when it does.

What AI Agents Actually Are

Let's define what we're talking about.

AI agents are autonomous systems that complete tasks on behalf of users. Not just answering questions (that's generative search). Actually taking actions. Booking. Purchasing. Scheduling. Managing.

Current examples:

• AI assistants that book travel based on preferences ("Find me a hotel in Dubai, under £200/night, near DIFC, with a gym")

• Shopping agents that compare products and make purchase recommendations

• Financial agents that optimise subscriptions and recommend changes

• Personal assistants that schedule meetings based on priorities and availability

These aren't hypothetical. They're operational. And they're getting more sophisticated rapidly.

Within two years, AI agents will handle a significant percentage of routine purchasing decisions. Within five, they might handle the majority of commodity purchases.

If your brand isn't designed to be selected by AI, you're designing for a shrinking market.

How AI Agents Make Decisions (and What That Means for Branding)

Let's talk about what factors AI agents consider when making recommendations.

Factor One: Structured, Machine-Readable Information

AI agents don't browse your website like humans. They parse structured data. Schema markup. Product feeds. API responses.

If your product information isn't structured and accessible, the AI can't evaluate it. You're not in consideration.

What this means:

Implement comprehensive schema markup. Product. Review. Organization. FAQ. Every structured data type relevant to your offering.

Maintain clean, updated product feeds. Accurate specs. Current pricing. Real inventory.

Build APIs that agents can query. Make your catalogue accessible programmatically.

The AI needs to "see" your products in a format it can process. Visual branding doesn't matter if the data isn't there.

Factor Two: Verifiable Quality Signals

AI agents are risk-averse. They're making decisions on behalf of humans who will blame them if things go wrong. So they heavily weight credibility signals.

What the AI looks for:

Reviews. Volume and recency. Not just average rating, but patterns in reviews. Are recent reviews better or worse than old ones?

Return rates. If available, low return rates signal satisfaction.

Certifications and compliance. Industry standards. Safety certifications. Regulatory approval.

Brand reputation. Press coverage. Awards. Third-party validation.

The AI isn't making emotional decisions. It's calculating risk-adjusted value.

Factor Three: Compatibility with User Profile

AI agents know the user. Their preferences. Purchase history. Values. Constraints.

When recommending products, the AI matches against this profile.

What this means:

Your brand needs to clearly communicate who you're for. Not vague "everyone." Specific use cases, values, and profiles.

If you're sustainable, that needs to be verifiable and structured. Not just marketing copy, actual certifications and data.

If you're premium, that needs to be justified with specifics. Materials. Craftsmanship. Warranty. Not just price signalling.

The AI is matching products to people. You need to make the match criteria explicit.

Factor Four: Transparent Pricing and Terms

AI agents compare total cost of ownership, not just list price. They factor in shipping, returns, warranties, subscriptions.

If your pricing is opaque or full of hidden costs, the AI will deprioritise you as risky or poor value.

What this means:

All-in pricing visible to automated systems. No surprises at checkout that the AI can't predict.

Clear return policies. Machine-readable terms.

Straightforward warranties and guarantees. No asterisks that require human interpretation.

The AI needs to confidently present total cost. Uncertainty makes you less recommendable.

Factor Five: Availability and Fulfilment

AI agents need to know if you can actually deliver. In stock? How long to ship? Can it arrive by the date the user needs it?

If this information isn't accessible, the AI moves to competitors who provide it.

What this means:

Real-time inventory data. Accessible via API or structured data.

Clear delivery timeframes. By region if you ship internationally.

Fulfilment reliability. If you frequently miss delivery dates, the AI learns and stops recommending you.

The AI is evaluating your operational reliability, not just your brand appeal.

The Decline of Visual Branding (in Agent-Mediated Commerce)

Here's the uncomfortable truth.

When an AI agent is making the purchase decision, your beautiful logo doesn't matter. Your stunning photography doesn't matter. Your carefully crafted brand voice doesn't matter.

Because the human never sees them.

The AI evaluates based on data, reviews, specs, and price. It makes a recommendation. The human approves or adjusts parameters. The purchase happens.

Your brand becomes invisible if it's only visual.

This doesn't mean branding is dead. It means branding has to work at two levels:

Level One: Human-facing brand for when people discover you directly. This is traditional branding. Visual. Emotional. Narrative-driven.

Level Two: Machine-facing brand for when AI evaluates you. This is data. Structure. Verifiable claims. Clear signals.

The brands that win will master both.

How to Design a Brand AI Agents Recommend

Let's get tactical. Here's how you position your brand for AI recommendation.

Strategy One: Become a Structured Data Powerhouse

Every piece of information about your products needs to be machine-readable.

Implementation:

Full schema markup across your site. Not just basic Product schema. Reviews. FAQs. How-to guides. Videos. Everything.

Comprehensive product data feeds. Google Shopping. Amazon. Any platform AI agents might query.

Open APIs. If you sell B2B or high-value products, AI agents might query your systems directly. Make the data accessible.

Example: A furniture retailer that provides detailed dimensions, materials, weight, assembly requirements, and environmental certifications in structured format will be recommended over a competitor with the same products but poor data structure.

Strategy Two: Optimise for Review Quality and Volume

Reviews are the social proof AI agents trust most. But not all reviews are equal.

Implementation:

Encourage detailed reviews. Prompt customers to mention specific use cases, problems solved, and experiences.

Respond to reviews. AI agents see engagement as a signal you care about customer satisfaction.

Address negative reviews. AI agents look at how you handle problems, not just whether problems exist.

Verify purchases. Verified reviews carry more weight.

Example: A skincare brand with 500 detailed, verified reviews will be recommended over a brand with 2,000 generic "great product!" reviews.

Strategy Three: Make Value Propositions Explicit and Verifiable

AI agents don't do brand storytelling. They evaluate claims. If you say you're sustainable, the AI needs proof.

Implementation:

List certifications prominently and machine-readably. B Corp. Fair Trade. ISO standards. Whatever's relevant.

Publish data. Carbon footprint. Supply chain transparency. Ingredients sourcing. Make it verifiable.

Third-party validation. Awards. Press from credible sources. Independent testing results.

Example: An eco-friendly clothing brand that provides carbon footprint data per garment and lists textile certifications will be recommended to environmentally conscious buyers over brands making vague "eco" claims.

Strategy Four: Simplify Purchase Decisions

AI agents prefer straightforward offerings. Too many options create complexity the AI has to evaluate.

Implementation:

Clear product tiers. Good/Better/Best. Not 47 variants that differ in unclear ways.

Decision-making help. "Most popular." "Best for X use case." Guide the AI's evaluation.

Transparent differentiation. What actually changes between variants? Make it explicit.

Example: A laptop manufacturer with three clear models (Budget, Performance, Premium) with obvious differentiation will be easier for AI to recommend than one with 15 models with overlapping specs.

Strategy Five: Build Authority in Your Domain

AI agents weight authoritative sources. If you're known as the expert in your category, you're recommended more.

Implementation:

Publish educational content. Comprehensive guides. How-tos. Industry reports.

Get cited by credible sources. Press. Academic papers. Industry publications.

Build thought leadership. Your founders and experts should be visible, credentialed, and published.

Example: A cybersecurity company whose CTO is regularly quoted in TechCrunch and whose blog is cited in industry reports will be recommended over competitors with no public authority.**

How This Plays Differently by Product Category

The impact of AI agents varies by what you're selling.

Commodity Products (Basic, Undifferentiated)

High AI agent impact. These are perfect for automated purchasing. "Buy the best toilet paper in my price range." AI makes the call.

Strategy: Compete on data. Reviews. Price. Availability. Fulfilment speed. There's no emotional brand differentiation here.

Considered Purchases (High-Value, Complex)

Medium AI agent impact. AI will shortlist options, but humans will make final decisions. "Find me three laptops for video editing under £2000." AI narrows, human chooses.

Strategy: Make shortlisting easy. Clear specs. Strong reviews. But also maintain human-facing brand appeal for the final decision.

Luxury and Aspirational (Brand-Driven, Emotional)

Lower AI agent impact (for now). These purchases are about identity and emotion. Humans want control.

Strategy: Focus on human-facing brand. But still implement structured data because AI agents will eventually influence luxury purchases too ("Which luxury watch holds value best?").

B2B and Enterprise

Growing AI agent impact. Procurement is being automated. AI agents evaluate vendors based on compliance, pricing, reviews, and case studies.

Strategy: Build machine-readable credibility. Client lists. Case studies with data. Certifications. Make due diligence easy for AI.

The Ethics of Agent-Mediated Commerce

Let's address the uncomfortable questions.

Who's responsible if the AI makes a bad recommendation?

This is legally unclear. Is it the AI platform? The brand? The user who set the parameters?

Expect litigation and regulation in the next few years. Brands should document that their product information is accurate and not misleading.

Can brands pay to be recommended?

This is already happening. Some AI platforms accept advertising. But users will abandon platforms that prioritise paid recommendations over quality ones.

Transparency will be key. "This recommendation is sponsored" disclosure will become standard.

How do smaller brands compete?

AI agents favour established brands with lots of reviews and authority signals. This could entrench dominant players.

The counterbalance: AI agents also optimise for value. If a smaller brand offers better specs at a better price, the AI will recommend it. Quality data and competitive pricing matter more than brand size.

The London vs. Dubai Context

Adoption of AI agents will vary by market.

UK Market

Early adoption. British consumers are already using AI for research and recommendations. Trust in AI decision-making is relatively high.

Implication: UK brands need to prioritise agent-readiness now. Your competitors likely are.

UAE Market

Tech-forward but relationship-driven. Gulf consumers value personal service and human interaction. AI agents will be adopted, but perhaps more slowly in high-touch categories.

Implication: UAE brands have slightly more time, but should prepare. Start with commodity categories where agents will be adopted first.

In both markets, the trajectory is clear: AI agents are coming. The only variable is pace.

What This Means for Brand Strategy

Here's the strategic shift brands need to make.

Old model: Build emotional connection with humans. Create desire. Drive traffic. Convert.

New model: Build machine-readable credibility. Make selection easy for AI. Ensure humans validate the AI's choice when they're involved.

This doesn't replace traditional branding. It adds a layer.

You still need human appeal for categories where humans make final decisions. But you also need machine legibility for the growing number of decisions AI makes autonomously.

How to Prepare Now

Here's what to do today.

Audit your structured data. Is every product properly marked up? Are reviews, specs, and availability machine-readable?

Test AI recommendations. Ask ChatGPT, Perplexity, and other AI tools to recommend products in your category. Are you mentioned? If not, why?

Build review volume and quality. This is the single biggest factor in AI recommendations. Prioritise collecting detailed, verified reviews.

Simplify your offerings. If your product range is confusing to humans, it's confusing to AI. Make decision-making clearer.

Document your authority. Get your expertise visible and verifiable. Publish. Get cited. Win recognition.

The DARB Edge

We're helping brands prepare for agent-mediated commerce whilst maintaining human appeal.

That means:

• Implementing comprehensive structured data

• Building machine-readable credibility signals

• Simplifying decision architecture

• Maintaining emotional brand appeal for human touchpoints

Whether you're in London, Dubai, or global markets, we make sure your brand works for both the humans who love you and the AI agents who recommend you.

Because the future of commerce isn't human or machine. It's both, simultaneously. And the brands that master both will dominate.

Ready to design a brand that AI agents choose? Let's make sure you're not invisible to the algorithms that will mediate commerce. Get in touch with DARB.