Blog

App Store Optimization is no longer just a metadata game. As assistants like ChatGPT become a front door to software, the ranking that matters is the one you cannot see: whether the assistant routes a user’s intent to your app at the moment of need.

Conversational ASO: the practice of optimizing how AI assistants understand, select, and fulfill an app’s capabilities from natural language requests.

From keywords to intents

Definition and schema

Your primary unit of optimization becomes the intent, not the query. Map top tasks in plain language, then express them as structured capabilities with parameters and outcomes. Think “book a same-day dog walker” rather than “pet services.”

Key insight: If the assistant cannot parse what your app does with precision, it will never route to you, no matter how strong your store keywords are.

From listings to capabilities

APIs beat aesthetics

Assistants prioritize reliable capabilities. Publish clear actions, input validation, rate limits, error semantics, and success states. Provide example prompts, sample payloads, and deterministic fallbacks so the assistant can execute confidently.

Key insight: Capability clarity is the new creative; the best screenshot is a well-specified action with predictable outcomes.

From installs to completions

Measure routing, not just downloads

Add metrics that reflect assistant-mediated demand: intent match rate, routing share by task, deep link landing accuracy, completion rate, and time-to-first-outcome. Treat installs as a byproduct of value, not the goal.

Key insight: What gets measured gets routed; assistants will learn from completion signals faster than from raw install counts.

From marketing to trust signals

Operational credibility

Assistants prefer trusted executors. Maintain verified developer profiles, transparent pricing, permission scopes, security attestations, and user-level controls. Surface guarantees and escalation paths inside your capability docs and deep links.

Key insight: Trust primitives decide tie-breaks when multiple apps can do the same task.

What to do next

Make the assistant your first user

Refactor your backlog around intents. Ship a public capability schema, harden deep links to drop users into the right flow, and align growth with completion metrics. Two quotable truths: “Capabilities route, not keywords.” “Completion rate is the new CTR.”

TL;DR

• Optimize intents, not keywords; publish precise capabilities the assistant can execute.
• APIs and reliability outrank screenshots; specify inputs, outputs, and fallbacks.
• Shift KPIs to routing share and completion rate over installs.
• Trust signals and verification determine tie-breaks among similar apps.
• Design deep links for task handoff so users land in the exact outcome flow.

OpenAI just put ChatGPT into its own browser. Atlas ships on macOS first with Windows, iOS, and Android on the way, and it brings an answer-first new tab, a persistent sidebar, memory, and an Agent Mode that can act on sites like Vrbo and Instacart for paid tiers as of late 2025 (Engadget, Shacknews).

That one move changes the ad conversation. Instead of bolting ads into chat transcripts, Atlas lets brands buy “actions” inside the flow of browsing. The assistant can see the page, summarize, compare, and complete tasks, which makes the sidebar and agent the new, high-intent ad inventory (MacRumors).

The new canvas: page-aware, answer-first, memory-enabled

From banners to “sponsored skills”

Atlas opens with a ChatGPT answer and quick tabs to traditional results, while the sidebar stays present as you work. Memory is optional but can personalize suggestions, and users can clear history or use Incognito. This creates natural “action slots” where the assistant can offer contextually relevant help that is clearly labeled and user-controlled (MacRumors, Mashable).

“Atlas is available on macOS now, with Windows, iOS, and Android coming soon,” which means this inventory will scale across devices faster than a standalone chat app could (Engadget).

Key insight: The premium placement is not a box on a page, it is a context-aware helper that proposes a next step at the moment of intent (MacRumors).

Agent Mode as the performance channel

Sponsored tasks, priced by completion

Agent Mode can navigate, fill forms, and even initiate purchases for users who opt in, though it remains constrained for safety: it cannot run code, download files, or install extensions, and it pauses on sensitive sites. Access is limited to Plus, Pro, and Business tiers today, with clear user controls and parental options (Mashable).

Definition: an “agentic ad” is a paid, opt-in task that executes a useful action, like booking a table or building a cart, while disclosing sponsorship and letting the user review steps. Atlas demos already show travel and commerce flows inside the browser, signaling a path from CPC toward CPA for verifiable actions (Shacknews, Digit.in).

“Agent Mode in Atlas is available only to Plus, Pro and Business users as of October 2025,” so early performance inventory will skew to high-intent, authenticated audiences (Mashable).

Key insight: Atlas turns performance media into sponsored workflows where completion, not clicks, is the billing primitive (Shacknews).

Governance, measurement, and competitive pressure

Consent-forward personalization beats surveillance

OpenAI states Atlas will not use the content you browse to train future models; users can manage memories and browse signed out or in Incognito. That shifts targeting toward consented memories and real-time context rather than passive tracking, with obvious implications for frequency, attribution, and brand safety review of agent steps (Engadget, Mashable).

Atlas also lands in a hotly competitive moment as Chrome deepens Gemini and Opera and Perplexity ship agentic browsers. Answer-first UX plus quick access to web results will pressure traditional search ads to prove utility, not just visibility (Engadget, MacRumors).

“OpenAI says it will not use the content users browse to train future models,” a policy that will shape how brands approach measurement and creative testing inside Atlas (Engadget).

Key insight: The winning Atlas ad stack will trade data exhaust for declared intent and auditable, user-reviewed actions (Mashable).

What marketers should build first

Design utility, label sponsorship, prove outcomes

Prioritize three formats: sponsored skills in the sidebar that improve a task, agentic ads that execute discrete steps with explicit consent, and answer-page recommendations that declare sponsorship and link to open web results for verification. Build measurement around action receipts and user-confirmed completions, not page views (MacRumors, Mashable).

Key insight: In Atlas, advertising that does work will outperform advertising that seeks attention (Engadget).

Conclusion

Atlas changes the unit of value. When the browser can act, the best ad is a useful action that is safe, consented, and measurable. Start small with sponsored skills and agentic pilots, label everything, and back every spend with a verifiable completion trail (Mashable).

The battle for the browser is now a battle to be the most helpful at the moment of intent. Plan media the way Atlas behaves: context first, action second, proof third (Engadget).

TL;DR

• Atlas moves ad value from impressions to actions via a page-aware sidebar and agentic tasks (MacRumors).
• Early formats: sponsored skills, answer-page recommendations, and paid agent routines priced by completion (Shacknews).
• Safety and consent rules shape targeting toward memory opt-ins and real-time context, not tracking (Mashable).
• OpenAI says browsing content is not used to train future models, changing measurement and experimentation norms (Engadget).

As of late 2025, ChatGPT’s Instant Checkout and agentic shopping capabilities are beginning to move buying moments off traditional websites and into dialogue. This isn’t a gimmick; it’s a reordering of the conversion playbook. When customers can complete a purchase without visiting your site, the surface-level CRO tests (buttons, forms, banners) no longer capture the driver of decision-making. The real leverage shifts to data feeds, recommendation logic, and the quality of conversational UX that guides intent directly in chat. [Source](https://americanbazaaronline.com/2025/10/13/openai-partners-with-indias-payment-authority-to-bring-upi-to-chatgpt-468659/)

This is not just about checkout friction; it’s about how brands stay visible, trusted, and relevant when the answer is delivered inside a chat. A corresponding perspective from the legal and governance lens reminds us that AI-driven experiences must balance privacy, control of data, and user trust as commerce moves into conversational spaces. As Law.com notes, navigating privacy myths and legal realities remains essential as AI-enabled workflows become part of everyday commerce. [Source](https://www.law.com/2025/10/13/between-privacy-myths-and-legal-realities-using-ai-without-losing-privilege-trade-secrets-or-control-of-your-notes/)

Thesis: CRO as a discipline survives, but it must become Conversational Optimization, where outcomes hinge on prompts, trust signals, and the relevance models that power chat-based commerce. In late 2024–2025 experiments, brands that excel in conversational UX design, data quality, and prompt engineering started outperforming traditional funnel optimizations even before a sale. [Source](https://americanbazaaronline.com/2025/10/13/openai-partners-with-indias-payment-authority-to-bring-upi-to-chatgpt-468659/)

Quote: “Agentic AI payments transform AI assistants from simple discovery tools into full-fledged shopping agents.” — Razorpay/OpenAI pilot remarks cited in the American Bazaar piece. [Source](https://americanbazaaronline.com/2025/10/13/openai-partners-with-indias-payment-authority-to-bring-upi-to-chatgpt-468659/)

Key insight: The future of conversion lives inside conversation—prompts, signals, and the UX that makes a chat feel like a trusted salesperson, not a page-on-click experiment.

Strategic shift: From funnel optimization to conversational optimization

The funnel isn’t vanishing; it’s evolving. When a chat-based checkout becomes the primary path, the strategic emphasis moves from button A/B tests to prompt design, trust cues, and contextual relevance within the dialogue. Brands must architect prompts that surface intent accurately, while ensuring the chat response includes clear trust signals (social proof, verified payment, privacy assurances) and concise, decision-ready options. The shift is corroborated by real-world AI commerce pilots that embed payments and product discovery inside chat, signaling a new optimization surface beyond the website. [Source](https://americanbazaaronline.com/2025/10/13/openai-partners-with-indias-payment-authority-to-bring-upi-to-chatgpt-468659/)

Promo cycles, product launches, and recommendations increasingly hinge on how well a brand’s data feeds align with conversational intents. If the chat replies with inaccurate product details or slow responses, trust erodes faster than a subpar landing page test ever could. This is why governance around data quality and prompt reliability matters just as much as conversion metrics. [Source](https://www.law.com/2025/10/13/between-privacy-myths-and-legal-realities-using-ai-without-losing-privilege-trade-secrets-or-control-of-your-notes/)

Key insight: Prompts, signals, and prompt governance are the new CRO stack—data quality and trust become the levers that move buyers through a chat, not a form.

Implementation framework: Define terms, metrics, and governance for Conversational Optimization

Definition first: Conversational Optimization refers to designing marketing experiences that guide customer intent through interactive dialogue across chat interfaces, with a focus on prompt construction, real-time relevance modeling, and trust-signaling within the response. Implementation requires three bets: 1) high-quality product and pricing data feeds for chat accuracy, 2) robust relevance models that rank responses by expected conversion probability, and 3) transparent trust signals (payment methods, data privacy, seller credibility) embedded in chat prompts. The AI-enabled checkout pilots (UPI-enabled ChatGPT shopping, agentic payments) illustrate how these elements come together in practice. [Source](https://americanbazaaronline.com/2025/10/13/openai-partners-with-indias-payment-authority-to-bring-upi-to-chatgpt-468659/)

Specifically, as of late 2025, partnerships with NPCI, Razorpay, Axis Bank, and Airtel Payments Bank showcase how embedded payments within chat can streamline the purchase flow while preserving user control and security. This is a concrete signal that Conversational Optimization will increasingly define ROI in AI-powered commerce. [Source](https://americanbazaaronline.com/2025/10/13/openai-partners-with-indias-payment-authority-to-bring-upi-to-chatgpt-468659/)

Key insight: If you can’t design prompts that surface intent and embed credible trust signals within a chat, you’ll be outperformed by brands that can.

TL;DR

• Conversion is migrating from pages to chats via prompt-driven conversations and agentic commerce.
• Trust signals and data quality in the conversation are as critical as price and product data on a landing page.
• AI-enabled payments inside chat (UPI, PayPal, cards) redefine what ‘checkout’ even means.

Conclusion

CRO in the conversational era isn’t about abandoning optimization; it’s about retooling it for dialogue. The surface changes, but the core discipline—understanding user intent and building trustworthy paths to purchase—remains constant. Start by codifying prompts, curating data feeds for chat accuracy, and surfacing explicit trust signals in every reply. Actionable takeaway: map your funnel to a conversation map, then optimize prompts and signals for each step of the chat journey.

As of late 2025, the most successful brands will treat Conversational Optimization as the default CRO playbook, not an experiment in a corner of the website. [Source](https://www.law.com/2025/10/13/between-privacy-myths-and-legal-realities-using-ai-without-losing-privilege-trade-secrets-or-control-of-your-notes/)

Perplexity stands at the intersection of AI powered answers and monetization, a moment that matters as AI search becomes a primary way people discover and compare products. The company has signaled that the ads layer is being reconsidered in real time, with an emphasis on preserving the integrity of AI answers and user trust. This is not a typical ad-sell cycle; it’s a test of how to monetize a knowledge engine without corrupting the quality of its responses. As Perplexity charts its path, the emphasis appears to be on consent, attribution, and long term ecosystem value, not quick wins.

[Perplexity ads pause] Perplexity is reportedly pausing new advertisers and reassessing the ad roadmap, a move described by AdExchanger and related outlets as a deliberate recalibration rather than a collapse of intent to monetize. This caution aligns with the broader industry note that AI search and answer experiences require careful measurement and governance to avoid misalignment with user needs. Source

In parallel, Perplexity has leaned into licensing relationships to anchor value in the AI knowledge stream. A Detroit Free Press piece highlights content licensing to Perplexity’s Comet-enabled browser, underscoring a path where publisher content fuels trusted AI answers and creates revenue without compromising answer quality. For advertisers, that signals a potential shift toward partner-led, attribution-safe integrations rather than blunt paid placements. Source

Key insight: Monetization in the AI answer era will come from trusted, attribution-forward partnerships that preserve answer integrity rather than aggressive banner play. Source

Platform Positioning and What It Signals for Advertisers

Perplexity’s current stance positions ads as a secondary, governance-heavy layer rather than a near-term revenue engine. The company appears to favor an architecture where paid placements feel native to AI answers and where measurement tools align with user intent, not just impression metrics. The AdExchanger analysis notes a broader industry trend toward more cautious deployment of AI-driven ad products, warning about “rage bait” and the risk of misalignment with brand safety. This suggests Perplexity may pursue sanitised, contextually relevant ad experiences that rely on rigorous source credibility and publisher partnerships. Source

Key insight: The next wave of AI-driven ads will be built on credibility and context, not volume and velocity. Source

Paths Forward: Monetization, Governance, and Measurement

Experts point to multiple potential monetization routes beyond traditional ad formats. A model anchored in licensed content, transparent attribution, and embedded, non-disruptive placements could align Perplexity with publishers and brands alike. Forbes’ 2026 AI trends emphasize expansionary AI use cases and nuanced governance — signals that Perplexity could monetize by enabling brand-safe AI conversations rather than pushing noisy ads into answers. In this frame, Perplexity could attract advertisers who value accuracy, safety, and measurable lift tied to specific query contexts. Source

Key insight: A credible, measurement-driven ad model anchored in governance and publisher partnerships may outperform brute-force ad saturation in AI answers. Source

Quotables to note:

“Monetization will come from trusted, attribution-forward partnerships that preserve answer integrity.”

“The next wave of AI-driven ads will be built on credibility and context, not volume.”

[AI adoption and monetization] Perplexity’s approach in 2025–2026 will likely hinge on governance, credibility, and publisher collaboration as core differentiators. Source

TL;DR

• Perplexity paused new advertisers to reassess ads within AI answers. Source
• Content licensing with Comet anchors a trusted monetization path. Source
• Future ads hinge on governance, attribution, and publisher partnerships. Source

Platform power shifts from apps to conversations

OpenAI’s DevDay 2025 reframed the internet as a conversation engine with apps embedded directly in ChatGPT. Users can plan trips, compose playlists, or book services without leaving chat, collapsing the web into a single dialogue thread. This is not a gimmick; it represents a fundamental shift in where and how action happens. As OpenAI demonstrated, the Apps SDK and Model Context Protocol (MCP) enable third‑party tools to render interactive interfaces inside ChatGPT, making discovery, selection, and action a single, conversational experience. For marketers, this is a call to rearchitect funnels around dialogue, not destinations. OpenAI’s app platform details outline how developers connect data sources to an AI runtime, creating a new surface for brand experiences within chat.

Meanwhile, the market has already seen early partners—Booking.com, Canva, Expedia, Spotify, Zillow, Figma, Coursera and more—demonstrating that commerce, education, media, and design can be activated inline. The practical implication is that marketers must think about how their products behave as conversational commands, not just as websites or ads. This shift is underscored by industry coverage of the Apps SDK launching with tangible use cases and with clear privacy guardrails that seek to balance surface area and consent. BusinessGhana’s coverage and Tempo’s DevDay context frame the scale and user reach now at stake for marketers.

As a result, “distribution” becomes a function of intent, not a marketplace listing. This is echoed in analyses of the ecosystem shift, including AgentKit and no‑code toolchains that let teams prototype and deploy conversational workflows rapidly. In short: the platform is the product, and conversations are the new channels. TechLatest on AgentKit and the MCP standard underpinning Apps SDK illuminate how brands can map data, actions, and UI into chat narratives.

Key insight: The Apps inside ChatGPT move marketing from destinations to orchestration, forcing a redefinition of where value is created in a customer journey. TechLatest coverage

Marketing strategy playbook for an era of conversational apps

The new reality asks marketers to rethink discovery, attribution, and loyalty in a world where a single chat can surface, compare, and execute across services. The first wave of Apps inside ChatGPT creates practical templates: location‑based home searches inside Zillow, booking flows inside Expedia, and design tasks inside Canva. The lesson for marketers is to design for “in‑chat journeys” where the user’s intent drives app surface and action in one thread. Coursera’s partnership signals how education and credentialing can migrate to conversational surfaces, turning chats into guided, interactive learning experiences. YourStory’s DevDay digest and Globe and Mail on Coursera provide concrete examples of this trend in action.

For CMOs, it’s no longer about ranking in an App Store but about surfacing in conversations where users plan, decide, and transact. The Apps SDK promises an ecosystem where millions of users encounter brand experiences through natural dialogue, bypassing traditional discovery barriers like search and pure ads. This requires new go‑to‑market thinking: co‑investments with AI developers, joint go‑to‑market content that showcases conversational flows, and clear privacy contracts that reassure users about data handling. GeekWire’s take on risk and traction and Tempo’s DevDay user scale frame the growth opportunities and cautions for marketers.

Key insight: Discovery migrates into conversation, so alliances with AI platforms and developers become the central component of a successful marketing strategy. GeekWire platform take

Measurement, governance, and risk in the new interface economy

The convergence of data, apps, and chat raises new questions about measurement and governance. In the OpenAI model, data surfaces are bounded by guardrails, privacy policies, and minimum‑data permissions, but the depth of access to conversational context remains a live topic. The Cable’s reporting on OpenAI’s Apps rollout highlights questions about data surface, user consent, and who controls the surface area of the data that apps access. It also notes planned granular data controls and the need for transparent permissions, a critical baseline for marketers operating in regulated regions. TheCable coverage and Mathrubhumi summary illustrate the cross‑border governance challenges and opportunities.

At the same time, there is a real incentive for brands to embrace attribution within the context of conversational actions. If a user asks for a travel plan and a playlist, the platform could attribute engagement to the participating app surface rather than a page click. The risk is potential data leakage or misalignment with consumer expectations about privacy; the industry will evolve standards for surfaced data, consent prompts, and monetization streams tied to app surfaces. BusinessGhana report and Tempo DevDay coverage anchor these debates in real world policy and practice.

Key insight: The new surface requires rigorous data governance and measurement models that tie in‑chat actions to business outcomes, not just impressions. TheCable analysis

Execution blueprint for CMOs in an AI‑first era

CMOs should pursue a two‑track strategy: (1) embed with OpenAI’s app ecosystem to surface branded capabilities directly in conversations, and (2) invest in internal conversational experiences that can be offered as a service to external developers via MCP‑compliant backends. This mirrors the AgentKit approach, which provides a no‑code builder to assemble AI agents and a robust governance framework for production use. The Apps SDK offers a directory‑style discovery path once surfaced publicly, enabling marketers to ride the network effects of a growing app ecosystem. TechLatest on AgentKit and OpenAI app platform overview provide concrete starting points for partnerships and experimentation.

Operationally, brands should design conversational flows that demonstrate value in a single chat: prompt the user with a natural language task, surface a branded app, and complete the action seamlessly. Coursera’s app partnership shows how a trusted learning platform can be woven into chat, turning a question into a guided learning path. Coursera–OpenAI collaboration and Mathrubhumi summary offer practical case studies.

Quote: “Agents inside our chat become the operating system for the internet, not just a search box.” This captures the strategic implication of a platform where intent drives actions across apps. AgentKit perspective

Quote: “Distribution is now a sentence.” A vivid reminder that reach comes from conversational surfaces, not banners. OpenAI app platform note

Key insight: CMOs must treat the OpenAI app economy as a joint venture with developers, balancing growth with governance to unlock reliable, scalable, conversation‑driven marketing. GeekWire risk‑and‑growth analysis

TL;DR: The shift to in‑chat apps demands a new marketing operating system built around intent orchestration, developer partnerships, and strong governance. TheCable coverage

TL;DR

• Apps inside ChatGPT turn conversations into commerce surfaces, compressing discovery, intent, and action. Tempo DevDay context
• Marketers should invest in partnerships, co‑branded conversational flows, and internal apps that can be surfaced by AI agents. AgentKit insights
• Governance, data privacy, and measurement are now front and center as data surfaces and permissions evolve. TheCable governance notes

The App Store Moment: OpenAI Turns ChatGPT Into an App Platform

OpenAI’s DevDay arc is turning conversations into a platform, not just a chat. By embedding third‑party apps inside ChatGPT via an Apps SDK, OpenAI aims to compress discovery, intent, and action into a single chat thread. In practice, you can summon apps like Spotify for a playlist or Zillow for an interactive property map without leaving the chat window. The reach is staggering: OpenAI projects hundreds of millions of weekly users, with early partner momentum from brands like Booking.com, Canva, Coursera, Expedia, Figma, Spotify, and Zillow. As of late 2025, this looks less like a gadget and more like a new operating surface for consumer and enterprise tasks. Source

Technically, the Apps SDK sits on top of OpenAI’s Model Context Protocol (MCP), enabling apps to connect data, trigger actions, and render interactive UI inside ChatGPT. That stack allows data-driven workflows to live within a chat thread, rather than bouncing users to a separate web page. OpenAI frames this as a full-stack platform for developers to build browser‑free, chat-first experiences with real data connections. Source

Beyond the tech, the business narrative is clear: OpenAI wants to reach hundreds of millions of users through a new software layer that sits between you and external services. Sam Altman envisions apps inside ChatGPT that are interactive, adaptive, and personalized—essentially an ambient operating layer that can transact, present data, and drive actions without leaving the chat. In other words, this is not a single app store but a platform that could rewire how people discover and engage with software. Source

Key insight: OpenAI’s app platform compresses discovery, decision and action into one chat thread, signaling a new operating layer for the AI era that could redefine app economics and marketing playbooks. Source

Implications for the App Ecosystem and Marketing: Redefining Discovery, Monetization, and Control

Redefining ASO: App Presence Optimization in a Chat‑first World

Traditional app store optimization (ASO) depends on search, keywords, ratings, and in‑market media. Inside ChatGPT, discovery is contextual and conversational. OpenAI’s directory, forthcoming monetization, and a focus on higher design standards for featured apps suggest marketers will need a new playbook—one that emphasizes in‑chat presence, seamless data integrations, and trust signals around privacy and UX. Early reporting highlights a directory and a path to monetization later in 2025, with apps like Booking.com, Canva, Coursera, Expedia, Spotify, and Zillow leading the way. Source

As this unfolds, the definition of discoverability shifts from keywords and downloads to conversational relevance, data permissions, and the quality of in‑chat interactions. The platform could curate and surface apps based on context, prompting new optimization metrics around integration depth, data partnerships, and cross‑app workflows. Source

In practice, this means marketers should plan for an “App Presence Optimization” framework that signals trust, data minimization, and a strong, clear value in the chat context. The initial wave of apps demonstrates real use cases, but monetization and discoverability will depend on how transparently apps request data and how well they integrate into the chat’s intent. Source

Key insight: Discovery inside a chat demands a new optimization lens—APO, not ASO—centered on conversational relevance, privacy signals, and seamless in‑chat workflows. Source

Platform Risk and the Gatekeeper Question

The shift from a pure app ecosystem to a platform where a single player (OpenAI) curates and governs the app surface raises familiar tech‑policy questions. Will OpenAI’s platform become a neutral conduit or a gatekeeper with inherent economic and data‑control incentives? Analysts have warned about the risk of centralized control and the potential for future monetization that could shape app visibility and traffic. The concern echoes historical platform shifts, with some observers warning that an AI ‘super app’ could introduce new privacy and governance frictions. Source

On the privacy front, conversations stored in ChatGPT and the AI graph raise questions about how much of a user’s private data travels with each app interaction, and how long that data persists. The platform’s success will depend on transparent policies, clear user consent, and robust controls around data sharing between apps and the host AI. Source

Key insight: The OpenAI app platform carries governance and data‑privacy risks typical of early platform bets; success will hinge on transparent policies, user control, and sensible monetization that preserves user trust. Source

Enterprise, Partnerships and the New “Directory” Economy

InfoWorld emphasizes that Apps SDK is open source, supports login and single sign‑on for enterprise use, and that OpenAI plans to roll out business and EDU editions with a dedicated directory. This points to a multi‑tier ecosystem where internal apps, enterprise workflows, and consumer apps share the same platform fabric but with different governance and data boundaries. The enterprise angle could accelerate B2B marketing opportunities as organizations embed AI workflows directly into the chat experience. Source

Meanwhile, the consumer side—through the app directory and a forthcoming monetization framework—signals a near‑term opportunity for brands to co‑build native experiences inside the chat. The first wave of partners demonstrates a mix of consumer and productivity apps, hinting at a broad spectrum of use cases that could redefine how brands interact with people in real time. Source

Key insight: A multi‑tier ecosystem with an enterprise app layer alongside a consumer app directory could unlock new marketing channels, but only if data governance and performance standards keep pace with growth. Source

What Marketers Should Do Now: Rethink, Reframe, and Run Pilot Programs

From Ads Inside Chat to AI-native Marketing Experiences

The marketing playbook shifts from hunting for impressions inside apps to designing experiences that live inside the chat surface. Rather than traditional display or search ads, marketers should imagine conversational experiences, guided recommendations, and data‑driven interactions that feel natural within a chat. The initial partner demos show how brands can surface recommendations or tasks directly in context, which can dramatically reduce friction in the path from intent to action. Source

In practical terms, this means marketing teams should experiment with co‑branded apps and in‑chat flows that showcase value quickly, such as a Canva‑driven design task or a Zillow map embedded in a chat. The relevance of these in‑chat tasks hinges on an elegant UX that requires minimal data sharing and instant, deterministic outcomes. Source

Quote: “This will enable a new generation of apps that are interactive, adaptive, and personalized — that you can chat with,” OpenAI’s Altman said, signaling a different cadence for marketing collaborations. Source

Key insight: Marketers should start building AI‑native experiences inside ChatGPT, not just place ads; co‑creation with app partners will be the pathway to meaningful, measurable outcomes. Source

Partnerships, Co‑Branding and the Directory Opportunity

As the platform scales, partnerships will matter more than traditional media buys. The first partner set—Spotify, Canva, Zillow, Booking.com, Expedia, Coursera, Figma—demonstrates how brands can extend their ecosystems into conversational flows. A deduplicated, high‑quality app directory could feature co‑branded experiences and simplified onboarding for developers, creating a new channel for brand experiences inside a ubiquitous AI interface. Source

Revenue models will evolve too. The Apps SDK preview notes potential monetization paths, including an agentic commerce protocol that enables instant checkout within ChatGPT. In late 2025, the monetization framework and directory policies will determine how brand partnerships scale and how much control developers have over traffic and pricing. Source

Key insight: The directory and monetization framework will determine whether this becomes a thriving marketer‑driven ecosystem or a limited set of high‑visibility experiments. Source

Governance, Privacy, and Risk Management

OpenAI’s platform promises powerful capabilities, but it also introduces new privacy and governance considerations. The possibility of an AI graph storing private conversations raises concerns about long‑term data exposure and user control. Platform observers stress the importance of policies that disclose privacy practices, limit data sharing, and provide granular controls for users. For marketers, this means designing consent flows and data minimization into every in‑chat experience. Source

As OpenAI opens the door to enterprise and education use, organizations will want to deploy internal apps with security assurances and single sign‑on. This adds an additional layer of compliance and risk management for marketing teams sponsoring internal AI workflows. Source

Key insight: Governance and data‑privacy controls will become a marketable differentiator; marketers must embed consent, security, and privacy as core components of any AI‑native experience. Source

Implementation Framework: A Practical Playbook for 2026

Phase 1: Partner Selection and Use‑Case Mapping

Identify 2–3 high‑intent use cases that fit conversational UX and can demonstrate measurable value inside ChatGPT. Prioritize brands with clear data partnerships or strong existing product ecosystems (e.g., travel, design, real estate) to showcase seamless “in chat” workflows. The initial partner wave already includes consumer and productivity apps, illustrating how embedded workflows can cross domains. Source

Define success metrics that reflect chat‑native outcomes: time‑to‑task, completion rate within chat, data‑sharing opt‑ins, and downstream conversion in the partner ecosystem. Monetization pilots should be time‑boxed, with clear exit or pivot criteria if the in‑chat experience underperforms relative to traditional funnels. Source

Key insight: Start with 2–3 precise, high‑intent use cases and concrete success metrics to de‑risk early adoption. Source

Phase 2: Design for Privacy and Trust

Integrate privacy and consent into every in‑chat workflow. Apps SDK constraints include careful handling of user data and limited PII exposure, so design patterns should minimize data sharing and provide obvious user choices. This is not just compliance; it’s a competitive advantage as users grow more selective about AI companions. Source

Key insight: Privacy‑by‑design will become a proxy for trust and a driver of adoption in an AI app store world. Source

Phase 3: Governance, Compliance and U/X Standards

Adopt governance frameworks that align with OpenAI’s platform policies and enterprise requirements, including single sign‑on, role‑based access, and data retention rules for internal apps. The risk/enterprise angle should evolve in tandem with consumer offerings, ensuring that the platform remains a trusted interface for business users and developers alike. Source

Key insight: A strong governance baseline is not a barrier to growth; it’s the fastest path to scalable, enterprise‑grade AI experiences. Source

Conclusion: A New Marketing Arena and a Call to Action

The OpenAI app platform presents a rare inflection point: a consumer and enterprise AI surface that compresses discovery, decision, and action into one chat thread. If developers can deliver high‑value, privacy‑respecting in‑chat experiences, and if OpenAI can design sane monetization and governance, the platform could become a major new marketing channel—one that rewards app depth, cross‑brand collaboration, and user trust over volume alone. The opportunities are immense, but the risks—privacy, control, and platform dependency—will require deliberate risk management and leadership from CMOs who want to future‑proof their brands. Source

Actionable takeaway: start with AI‑native co‑brands and a tight, chat‑centric value proposition, build a pilot inside ChatGPT, and measure success through in‑chat task completion, consented data shares, and downstream impact on brand experience. The era of ASO may evolve into APO—App Presence Optimization—where trust and conversational fidelity win more than keyword rankings. Source

TL;DR

• OpenAI’s Apps SDK and MCP enable third‑party apps to live inside ChatGPT, reaching hundreds of millions of users. Source
• Marketing opportunities will center on AI‑native in‑chat experiences and co‑branded workflows, not traditional ads. Source
• Privacy, governance and data controls will define platform trust and adoption; monetization plans are still evolving. Source
• Early adopters should pilot a 2–3 use cases with clear success metrics, focusing on in‑chat task completion and data consent signals. Source

Opening

AI-driven conversational ads are the next frontier where dialogue and commerce meet in real time.

Why this matters now: AI assistants are increasingly central to search, shopping, and information experiences. Brands are testing ad formats that appear inside dialogue, not just on pages, prompting a need for standardized measurement, governance, and scalable creative frameworks across platforms like OpenAI, Claude, and Perplexity.

Factual takeaway: Conversational ads surface within real-time dialogue and monetize through context-driven interactions, not static placements.

Definitions & Scope

Definitions: AI-driven conversational ads refer to paid placements embedded within AI chat interfaces or conversation surfaces, including sponsored answers, chat placements, and native responses within retail or assistant experiences. They are distinct from traditional banners, pre-rolls, or influencer promotions that do not participate in the dialog stream.

Scope: This article covers AI-ad surfaces across leading platforms (OpenAI, Claude, Perplexity) and adjacent chat-enabled ecosystems, focusing on actual or imminent ad formats, governance considerations, and measurement approaches. It intentionally excludes non-conversational display ads that appear outside chat contexts.

Factual takeaway: The taxonomy centers on ad formats that integrate with, or appear inside, AI-generated dialogue and assistant interactions.

Market Trends & Signals

• Brand experimentation is broadening beyond hype to real-world performance tests in chat surfaces, with campaigns that blend sponsorships into AI dialogues and shopping prompts.
• Storytelling pivots are moving from tech-centric narratives to human-centered use cases (recipes, travel planning, daily tasks), signaling a shift in how AI brands position utility and trust.
• Public backlash risk remains a factor as some campaigns test provocative or social-impact messaging within transit and street-level media. Brand safety discourse is intensifying around AI-generated content and prompts used in co-creative processes.
• Measurement maturity is evolving. Early pilots emphasize intent signals, time-in-dialogue, and downstream conversions, but standardization across platforms is still emerging.

Factual takeaway: The AI-ad surface market is moving from experimental hype toward performance-focused deployments, with increasing attention to safety, governance, and standardized metrics.

Platform & Feature Analysis

OpenAI: Campaigns are broadening to human-centered storytelling across media, signaling a brand-safe shift toward practical, everyday use cases (recipes, planning, personal tasks). While not all formats are public-facing as native in-chat ads, OpenAI’s branding work foreshadows potential sponsorships or integration points within future chat experiences and co-creative workflows.

Claude (Anthropic) and Perplexity: Perplexity is exploring paid placements that strive to feel native within AI-generated answers, aiming to preserve answer integrity while introducing sponsor signals. Claude’s brand positioning efforts emphasize a broader shift to responsible AI storytelling in ads and developer-facing contexts.

Emerging formats: Sponsored answers, chat-embedded promotions, and shopping prompts that originate in conversation rather than on search result pages. These formats require alignment with platform governance, prompt-design guidelines, and clear disclosure to maintain user trust.

Factual takeaway: Platform-specific ad formats are converging on native, dialogue-integrated placements, with governance and disclosure as prerequisites for scale.

Measurement & Attribution Framework

KPIs: incremental lift (sales, sign-ups, or intent), engagement depth within the chat, brand lift (familiarity, trust), and time-to-conversion from initial dialogue. Incrementality requires randomized holdout experiments within AI surfaces and cross-channel attribution to isolate dialog-driven impact.

Measurement approach: Combine experimental designs (A/B tests in chat contexts) with robust attribution models that span chat, search, social, and retail touchpoints. Include privacy-preserving analytics and disclosure considerations for AI-generated content in measurement pipelines.

Quality signals: compute coherence of AI responses, usefulness of recommendations, and transparency of sponsorship disclosures to protect brand integrity.

Factual takeaway: A rigorous, cross-channel measurement framework with clear disclosure is essential to quantify the true impact of AI-driven conversational ads.

Risk & Governance

Key risks include privacy concerns, data governance, potential AI hallucinations, and reputational exposure from provocative or misaligned messages within a dialogue. Brand safety policies must address how prompts, co-creative ideation, and sponsor signals interact with user trust. Regulatory and platform-specific disclosures are critical as ads become embedded in conversations rather than separate media units.

Mitigation: implement inbound risk reviews, guardrails for prompt design, clear sponsorship disclosures within dialogue, and risk-based budgeting to avoid high-variance campaigns in sensitive topics.

Factual takeaway: Governance and transparent disclosures are non-negotiable for scalable, responsible AI-ad experiences.

Strategic Implications & Recommendations

• Adopt a cross-functional governance model (marketing, legal, privacy, product, and engineering) to oversee dialog-based advertising standards and disclosure practices.
• Prioritize controlled pilots that test specific use cases (e.g., recipe recommendations or travel planning) with predefined success metrics and holdout groups.
• Develop creative playbooks for human-AI co-creation workflows, including templates for sponsored dialogue and ethics guidelines for tone, transparency, and inclusivity.
• Invest in measurement infrastructure that integrates AI-dialog engagement with traditional marketing metrics, ensuring incrementality and brand safety signals are aligned.
• Plan for multi-platform consistency while recognizing platform-specific constraints and opportunities in OpenAI, Claude, and Perplexity ecosystems.

Factual takeaway: Successful adoption requires disciplined governance, targeted pilots, and integrated measurement that connects chat-driven signals to business outcomes.

Future Outlook

Near-term signals point to broader adoption of native conversational ad formats, deeper integration of commerce within AI chat experiences, and the emergence of standardized reporting for dialog-based impact. As AI assistants become more central to decision-making and shopping, brands should expect a more mature marketplace of sponsor-aware, compliant ad formats that prioritize user trust and transparent disclosures.

Factual takeaway: Expect rapid expansion of dialog-integrated ad formats, with governance, measurement, and cross-platform collaboration becoming prerequisites for scale.

FAQ

• Q: What exactly are AI-driven conversational ads?
  A: Paid placements embedded within AI chat interfaces or dialogue surfaces, such as sponsored answers or chat-native promotions, designed to appear as part of the conversation rather than as separate banners.
• Q: How should we measure their impact?
  A: Use randomized experiments within chat contexts, together with cross-channel attribution and holdout groups to isolate dialog-driven lift; include brand-safety and disclosure metrics.
• Q: What are the main risks?
  A: Privacy concerns, AI misalignment or hallucinations, potential backlash from provocative content, and risks to trust if disclosures are unclear or inconsistent.
• Q: Where should a brand start?
  A: Launch a small, well-scoped pilot on a single platform, define success metrics, and establish governance for disclosures and creative prompts before expanding.
• Q: Will these ads be ROI-positive soon?
  A: Early pilots focus on learning and initial lift; ROI depends on rigorous experiments, measurement integration, and cross-channel orchestration—timeline varies by category and platform.

Key Takeaways (TL;DR)

• AI-ad surfaces inside conversations are moving from hype to measurable experiments; governance and disclosure are essential for scale.
• Cross-platform measurement is non-negotiable; tie dialog-driven signals to downstream business outcomes.
• Co-creation with AI can accelerate concepting but requires clear ethical and tonal guidelines to protect brand trust.
• Start with focused pilots, build a governance framework, and iterate toward standardized, evaluable metrics across OpenAI, Claude, and Perplexity ecosystems.

Conversational marketing is the design of marketing experiences that engage customers through interactive, AI-powered dialogue across chat, voice, and natural language interfaces. As of 2025, advances in large language models and AI assistants from OpenAI, Google, Anthropic, and others enable brands to replace static funnels with real time conversations that guide, inform, and convert.

Why this matters now: consumer expectations have shifted toward immediacy, personalization, and two‑way dialogue. Platform shifts—from AI chat in search results to brand‑owned conversational experiences—have expanded the touchpoints where marketers can influence outcomes. For senior leaders, the implication is clear: success requires governance, measurable impact, and a disciplined pilot rhythm across channels and partners.

Definitions and scope

Definition: AI‑powered conversational marketing designs marketing experiences that engage customers through real‑time dialogue using chat and voice interfaces, enabled by large language models and retrieval augmentation. It encompasses chat ads, AI‑driven search results, on‑site chat, and voice assistants used for brand engagement and conversion. It does not include non‑interactive display banners or generic outbound messaging that lacks two‑way dialogue.

Scope and boundaries: align marketing, CX, and data governance. Include two‑way dialogue that can be personalized, contextually aware, and capable of guiding a user toward a decision. Exclude one‑way push messages and non conversational placements.

Factual takeaway: Conversational marketing combines two‑way dialogue with AI to move customers through decision cycles in real time.

How it works: workflow, data, models, interfaces, targeting, measurement

Data and consent: rely on first‑party data, consented signals, and privacy‑preserving techniques. Maintain clear disclosures when AI responds autonomously. Use retrieval augmented generation to ground responses in trusted sources.

Models and guardrails: leverage leading LLMs (OpenAI, Google Gemini, Anthropic Claude) with retrieval layers and strict guardrails to minimize hallucinations and bias. Monitor model behavior and implement escalation paths to human agents when needed.

Interfaces and experiences: deploy chat widgets on brand sites, voice experiences on devices, and AI‑enhanced search results. Integrate with CRM, commerce, and helpdesk systems for real‑time context sharing.

Targeting and orchestration: combine query‑level intent signals with CRM attributes, behavioral data, and consented demographics. Orchestrate experiences across channels to ensure consistency and avoid conflicting prompts.

Measurement design: use conversation‑level metrics (start rate, completion rate, sentiment, escalation rate) and outcome metrics (lead quality, bookings, revenue per conversation). Apply attribution models that connect conversations to downstream conversions.

Platform examples and features: OpenAI API and Chat models for dialogue; Google Gemini for multimodal context; Anthropic Claude for safety‑first flows; Perplexity Ads for AI search‑context conversations; Microsoft Advertising and Bing AI for chat‑enabled search results; CRM integrations (Salesforce, HubSpot) for context sharing; live chat vendors (LivePerson, Intercom) for brand site experiences.

Factual takeaway: A practical conversational stack combines first‑party data, guarded LLMs, and chat/voice interfaces across owned and partner surfaces to drive measurable outcomes.

Comparisons: conversational formats vs traditional ad experiences

The table below highlights how conversational formats differ from traditional formats in placement, intent, targeting, measurement, cost, scale, and risk.

Format	Where it appears	User intent	Targeting	KPI / outcomes	Cost model	Key risks
Conversational Ads in AI search	AI‑driven search results (e.g., Perplexity Ads, Bing AI surfaces)	Discovery with next steps and guidance	Query‑level intent, CRM signals	Conversation start rate, time to value, qualified leads	CPC / CPV with premium for integration	Hallucinations, misalignment with brand safety, disclosure
Chat‑native placements on brand sites/apps	Brand web and mobile apps, embedded chat	Support, product discovery, lead capture	CRM data, on‑site behavior, consented attributes	Completion rate, conversion rate, average order value	Flat rate or usage‑based, per‑conversation	Data privacy, integration complexity, agent handoff quality
Voice‑assisted commerce and chat	Amazon Alexa, Google Assistant, other voice platforms	Purchase guidance, quick actions, recommendations	Contextual signals, device ownership	Order rate, basket size, repeat interactions	Cost per interaction or per sale	Privacy concerns, misrecommendations, user consent for voice data

Factual takeaway: Different conversational formats require distinct placements, intents, and measurement approaches; the right mix aligns channel capabilities with business goals.

Examples: platforms, formats, and use cases

• Perplexity Ads: sponsored conversations within AI search results that surface when users seek information and guidance.
• Bing AI chat ads: AI‑powered chat surfaces in search results that can include sponsor integrations and product recommendations.
• Brand site chatbots: live chat experiences powered by LivePerson, Intercom, or custom AI agents that qualify leads and support purchases.
• Voice assistants for commerce: shopping interactions via Amazon Alexa or Google Assistant that guide product discovery and checkout.
• Retail media assistants: AI chat copilots within ecommerce storefronts to compare products and surface promos in real time.

Factual takeaway: Real‑world use cases show conversational marketing spanning discovery, comparison, and conversion across search, site, and voice interfaces.

Measurement and ROI

KPIs should connect conversations to business outcomes. Core metrics include start rate, completion rate, conversion rate, average order value, revenue per conversation, and time to decision. Attribution requires multi‑touch models and, when possible, incrementality tests (holdout groups, randomized experiments) to isolate conversational impact.

Quality governance is essential: track sentiment, escalation rates, and brand safety indicators; ensure disclosures when AI answers are AI‑generated; monitor for data leakage across CRM and chat transcripts.

Factual takeaway: ROI from conversational marketing depends on reliable measurement, clean data inputs, and disciplined attribution that links dialogue to revenue.

Risks and challenges

Privacy and data handling remain top concerns. Comply with GDPR, CCPA, and sector‑specific rules; obtain explicit consent for recordings and personalized interactions. Platform policies can constrain or ban certain prompts or data sharing practices.

Hallucinations and bias pose risk to credibility. Implement guardrails, monitoring dashboards, and escalation to human agents for high‑stakes advice (financing, legal, health, etc.).

Disclosure and trust matter. Clearly label AI‑driven responses when appropriate to maintain consumer trust and comply with marketing regulations. Dependency on a single vendor increases risk if the platform changes policies or pricing.

Factual takeaway: The major risks are privacy, accuracy, policy limits, and trust; proactive governance reduces negative outcomes.

Opportunities and playbook

Start with a disciplined pilot. Define a narrow objective (e.g., shorten support cycle by 20%, lift qualified leads 15%), select a single format and platform, and establish a control group. Design conversations with guardrails, brand voice, and escalation paths to human agents when needed.

Playbook steps:

• Assemble cross‑functional stakeholders from Marketing, CX, Legal, and IT.
• Define data requirements, consent frameworks, and privacy safeguards.
• Choose pilot format (AI search surface, site chatbot, or voice assistant) based on near‑term impact and feasibility.
• Set success metrics and a clear ROI target; plan incrementality testing.
• Prototype with a content and UX guardrail, then iterate with real users.
• Establish governance for model updates, data retention, and vendor risk management.

Factual takeaway: A structured pilot with clear goals, governance, and incrementality testing accelerates learning and reduces risk.

Future outlook

Near term, expect deeper integration of AI chat into search, shopping, and customer service. Expect more standardized measurement approaches, better guardrails, and deeper first‑party data integrations that improve relevance while preserving privacy. Regulatory scrutiny around AI transparency and data usage will intensify, prompting clearer disclosures and auditable processes.

Forward‑looking statements should be labeled as such. Signals to watch include platform announcements on AI search experiences, moves by privacy regulators, and enterprise adoption patterns across financial services, retail, and travel.

Factual takeaway: The landscape will consolidate around governed, measurable, AI‑driven conversations that augment human agents, not replace them.

FAQ

1. What is AI‑powered conversational marketing?Marketing that uses AI chat and voice interfaces to engage customers in two‑way dialogue across channels, grounded in data and governed by privacy rules.
2. How should I measure ROI for conversational marketing?Link conversations to outcomes via attribution models, use incrementality tests, and track conversation‑level metrics like start rate, completion rate, and conversion rate.
3. What are the main risks to manage?Privacy compliance, model safety to avoid hallucinations, platform policy changes, and the risk of eroding trust through poor experiences.
4. How do I start a pilot?Choose a focused objective, select one format and platform, define success criteria, build guardrails, and run a controlled test with a clear go/no‑go plan.
5. Which platforms should we watch?Key players include OpenAI, Google Gemini, Anthropic Claude, Perplexity Ads, and Microsoft Advertising for AI‑driven search experiences, plus CRM‑integrated chat platforms like LivePerson and Intercom for on‑site conversations.

Factual takeaway: Start with a focused pilot, clear metrics, and governance to de‑risk AI‑driven conversations.

Key takeaways (TL;DR)

• AI‑powered conversational marketing enables real‑time, two‑way dialogue that can improve conversion and loyalty when grounded in first‑party data and strong governance.
• Measure success with conversation metrics and revenue outcomes; use incrementality tests to prove lift beyond traditional channels.
• Governance matters: privacy, disclosures, guardrails, and vendor risk determine long‑term viability.
• Pilot thoughtfully with cross‑functional teams, defined success criteria, and a clear plan to scale if results justify it.

Factual takeaway: The path to impact is disciplined pilots, robust measurement, and clear governance that scales across channels.

Perplexity has been positioning itself as more than a curiosity-driven search engine. Since late 2024, the company has been testing a model for advertising that looks very different from the banner-and-click systems marketers are used to. The basic idea is simple: keep the integrity of AI-generated answers intact, while introducing paid placements that feel native to the experience.

Ads appear in two places: as sponsored follow-up questions at the bottom of a response or as placements in the sidebar. Both are marked “Sponsored,” and the copy is still generated by Perplexity’s AI. Clicking on an ad doesn’t take a user to a landing page. Instead, it prompts the system to continue the conversation with the advertiser’s brand in view. Perplexity stresses that advertiser influence stops at the placement level; answers are not edited or supplied by brands themselves Perplexity blog.

The Commercial Logic

Subscriptions and publisher partnerships alone won’t keep the lights on. In 2024, Perplexity generated about 34 million dollars in revenue while burning through almost twice that amount to cover infrastructure costs Digiday. Ads are intended to provide a steadier stream of income and give brands a transparent way to engage with the platform’s growing audience. Early partners included Indeed, Whole Foods Market, Universal McCann, and PMG.

The ad model is sold on a cost-per-mille basis, typically ranging from 30 to 60 dollars per thousand impressions, with Perplexity initially targeting CPMs above 50 WebFX. For now, opportunities are limited to select brand and agency partners as the company builds the product.

Audience and Scale

Perplexity reports around 22 million active users, which is a fraction of ChatGPT’s estimated 400 million or Google’s AI Overviews, which claims over 1.5 billion monthly users. What makes the platform interesting for advertisers is the composition of its audience: mostly college-educated, relatively affluent, and with a significant portion in senior roles WebFX. In other words, small scale, but attractive demographics.

This positioning creates a tension. Advertisers see the promise of reaching high-value users in a less crowded environment, but they also want reach and measurable ROI. Several agency buyers told Digiday they’re holding back spend because the platform is still awareness-driven, lacks performance metrics, and offers little efficiency compared to established channels. CPMs in the 50 dollar range only sharpen that concern.

Early Impressions From Buyers

The reaction after six months of testing is mixed. Marketers are curious, but many feel Perplexity hasn’t moved fast enough. As Robert Kurtz of Basis Technologies put it, brands are waiting for lower-funnel ad options before they can justify significant investment Digiday. Ryan Bopp of Eden Collective highlighted concerns around brand safety and ROI, noting that his team hasn’t advanced any actual buys yet.

There is recognition, though, that Perplexity was the first AI answer engine to make a real push into advertising, and that it continues to test ways to integrate commerce directly into its conversational flow. Debra Aho Williamson of Sonata Insights described it as “a small fish in a big AI pond” but credited the company with popularizing the notion that AI platforms can be ad destinations Digiday.

Why This Matters

Perplexity’s model hints at what advertising in AI-native environments may look like: context-aware, conversational, and less about clicks than about engagement inside the system. For now, the opportunity is limited. The audience is relatively small, the buy-in expensive, and the product still being defined. But it offers a preview of how brands might interact with users once AI platforms grow beyond experiments and into daily habits.