Another seven days in Austin. Another complete dismantling of assumptions I didn't realise I was still making.
SXSW 2026 confirmed what many of us have sensed: the marketing world as we built it over the last two decades is not evolving. It is being replaced.
Observation 1: The Customer Journey as We Know It Is Ending
The traditional customer journey assumes progression. A consumer becomes aware, considers options, converts, and, if everything goes well, becomes loyal. This model has shaped decades of marketing strategy, from funnel diagrams to nurture sequences to lifecycle stages.
But this assumption no longer reflects how people actually behave. The reality is far messier and far more human. Through the use of AI tools, today's consumers:
- Enter journeys at different points
- Move forward, backward, and sideways
- Pause engagement for weeks or months
- Re-engage when timing feels right
Why the funnel breaks down
The traditional linear journey assumes three things that do not always hold true: a shared starting point, a consistent pace, and a single definition of "progress." AI-literate consumers challenge all three. Some arrive deeply informed before ever considering a purchase. Others investigate early, then disengage until a specific need arises. Many move in and out of consideration multiple times, especially in long-cycle decisions.
For a brand, pauses may look like drop-off. Skipped steps look like disinterest. Nonlinear behaviour looks like failure. But from the consumer's point of view, these are entirely rational choices.
The shift to state-based thinking
Instead of thinking in terms of linear steps and stages, adaptive brands are shifting to a state-based model. In this model, consumers are not necessarily "moving forward." They are occupying states: temporary phases defined by context, intent, and readiness.
A consumer state reflects a moment in time shaped by immediate needs, available attention, external pressure such as timing or deadlines, and readiness to act. The most important caveat: states are non-sequential. A person can move from high intent to low intent without warning or explanation. The brands that learn to read and respond to states, rather than stages, will be the ones that stay relevant.
Observation 2: PPC Will Not Die, But Its Core Primitives Are Breaking
Pay-per-click advertising was built on a specific behaviour: a human typing a query, scanning results, and clicking a link. That behaviour is changing, and the consequences for PPC are structural, not superficial.
Four shifts rewriting the rules
1. AI removes the need for "search" as a behaviour
PPC fundamentally depends on user-initiated queries. Large language models are shifting interaction from a "search, click, browse, convert" model to an "ask, receive a synthesised answer, act" model, often without a single click. If users stop clicking links, there is no SERP real estate, no ad inventory, and no cost-per-click model. The intent capture layer collapses.
2. Zero-click journeys eliminate monetisable touchpoints
We are already seeing this with Google AI Overviews, ChatGPT browsing, and Perplexity-style answers. These systems aggregate information, provide recommendations, and, increasingly, complete actions such as booking, buying, and subscribing. If a consumer can ask "What is the best CRM?" and receive a recommendation with integrated purchase options, there is no click auction. Monetisation shifts to placement inside the answer, not outside it. That is not PPC. That is something closer to answer engine optimisation (AEO) combined with paid inclusion.
3. AI agents will act on behalf of users, not browse like them
This is the shift most people are underestimating. The future flow is: a user gives intent to an AI agent, the agent evaluates options programmatically, and the agent executes the purchase decision. Agents do not see ads. They do not click links. They do not respond to creative or copy. They optimise for structured data, price, trust signals, and API access. Instead of bidding on keywords, brands will need to win machine-readable preference signals and integrate directly into agent ecosystems. This kills the psychological and auction-based mechanics of traditional PPC.
4. Platform incentives are shifting from traffic to retention ecosystems
Google, OpenAI, Meta, and others are all moving toward closed-loop environments, native transactions, and subscription-plus-API monetisation. In that world, sending users off-platform becomes less valuable. Keeping users inside the ecosystem becomes more valuable. Monetisation shifts toward sponsored answers, native product placements, and revenue-share or transaction-fee models.
| Today (PPC) | Future (Emerging Model) |
|---|---|
| Keywords | Intent understanding |
| Clicks | Outcomes and actions |
| Ads outside content | Ads embedded in answers |
| Human persuasion | Machine selection |
| Auction-driven | Model-driven |
PPC will not die overnight. But the primitives it was built on are being systematically removed.
Observation 3: The Measurement Crisis Is Here, and It Is Getting Worse
Marketing measurement was already imperfect. It is increasingly becoming unreliable in ways that have real consequences for planning, budgeting, and accountability.
Three reasons the funnel is becoming harder to measure
1. Fragmented user journeys across platforms and devices
Consumers no longer move linearly from awareness to conversion. They bounce between TikTok, search, Reddit, AI tools, and direct visits, often anonymously or across multiple devices. This breaks attribution continuity. What used to look like a clean path to purchase is now a messy, non-linear web of touchpoints that cannot be stitched together reliably.
2. Signal loss and privacy-driven data limitations
With cookie deprecation, iOS tracking restrictions, and increasing privacy regulation, the observable data layer is shrinking fast. Platforms are operating more as black boxes, and modelled conversions are replacing deterministic tracking. Measurement is becoming probabilistic rather than precise. Planning based on "this channel drives X stage of the funnel" is increasingly indefensible.
3. The collapse of intent visibility
Search used to be the clearest signal of intent. Now, AI summaries, zero-click results, private communities, and dark social channels including DMs, Slack, and WhatsApp are absorbing intent without leaving measurable footprints. Discovery is happening algorithmically before intent is even consciously formed. The distinction between demand creation (brand) and demand capture (performance) is blurring, making traditional funnel planning frameworks increasingly artificial.
And to my Prediction: There Can Be Only One
In the immortal words of Highlander, there can be only one.
And that one is Alphabet.
One of the privileges of attending SXSW is access to some of the brightest minds working in AI today. Conversations with DeepMind leaders provided a perspective and a clarity that is hard to replicate. You come away with a very real sense that we are heading toward an inescapable world: an ecosystem, a closed environment, where in their vision, you never need to leave.
Gemini is breaking records in deploying better AI faster. The Alphabet AI world is seamlessly integrating with your data points, your drive, your docs and decks, with AI becoming embedded at every turn. Whether you are creating professional-level imagery through Nano Banana, composing music tracks for ads through Lyria, project managing and converting your thoughts into audio through NotebookLM, conducting scientific research through Deep Think, or creating world-class video assets through Veo, everything Alphabet is building points to a world you can never leave.
Backed by the world's largest search engine, the world's most widely used communication and collaboration platforms, and the world's largest streaming service, the scale of what is being deployed makes it easy to see how a narrowly focused player like ChatGPT could eventually be overtaken. Alphabet is not building tools. It is building gravity.
The question for every marketer is not whether this world arrives. It is whether your brand, your content, and your data structures are ready to exist inside it. Reach out to me if you'd like to chat further!
Michael Ungerboeck is the Founder and CEO of ADMATIC, a digital marketing agency specialising in performance marketing across Australia and New Zealand.