Marketing Leadership Minus the hype 4: the platform trap

Posted by Mike O'Brien on May 28, 2026 in Marketing Leadership | Leave a comment

Why search fragmentation is a sovereignty problem and what marketing leaders should build instead

Every significant shift in the search landscape over the past decade has produced the same organisational response. A new platform emerges, or an existing one changes its behaviour, and the marketing industry mobilises around it. Budgets are reallocated. Specialists are hired. Frameworks are published. Conferences are convened. And brands invest, often substantially, in adapting their presence, their content, and their technical infrastructure to the new environment, until the environment changes again, which it always does, at which point the cycle begins once more.

The response is rational given the belief driving it: that search fragmentation is fundamentally a discoverability problem, solved by being present across more surfaces, producing more content optimised for each, and adapting more quickly than competitors to algorithmic change. Execute that well enough, consistently enough, and the brand will be found wherever its audiences are looking.

This belief is held sincerely and acted on expensively across the industry. It is also the belief that most reliably produces deeper platform dependency rather than less of it, because every investment made in optimising for a platform’s current requirements is an investment in a relationship whose terms are controlled entirely by the other party.

The question that most marketing organisations have not seriously asked is whether the game itself is the problem.

The feudalism hiding inside the platform economy

In Technofeudalism, economist Yanis Varoufakis argues that the major technology platforms have achieved something qualitatively different from conventional market capitalism. They are not simply large companies competing in markets. They have, in significant domains, replaced markets with what he calls cloud fiefs: enclosed environments where economic activity happens not through market exchange between roughly equal parties but through the extraction of cloud rent from those who have no practical alternative but to operate inside them.

The application to marketing leadership is direct and largely unmade in the industry conversation. A brand operating on Google, Meta, Amazon, or TikTok is not a buyer of services in a competitive marketplace. It is a tenant. It pays perpetual rent, in data, in attention, in margin, in the labour of content production, for access to audiences it cannot own, building presence on land whose terms of tenure can be altered unilaterally, without negotiation, and without compensation.

Varoufakis’s framework makes visible something that the language of “platform partnerships” and “channel strategy” consistently obscures: this is a power relationship, not a commercial one. And the power sits entirely with the platform. When Google restructures its search results to prioritise AI-generated answers, the brand that has invested years in SEO finds its traffic evaporates. When Meta changes its algorithm, organic reach collapses overnight. When Amazon adjusts its ranking logic, product visibility shifts in ways that no amount of optimisation spend can reliably control. The brand has no appeal, no recourse, and no compensation. It simply adapts or pays more for the visibility it used to earn.

The marketing industry has largely accepted this as the natural condition of operating in a digital environment. Varoufakis would suggest it is worth examining more carefully. Accepting feudal dependency as the cost of doing business is a choice, not an inevitability, and it is a choice with compounding strategic consequences that most marketing leaders have not fully priced. The scale of the technology infrastructure that has grown up around platform optimisation makes that cost visible in a different way: Scott Brinker’s annual marketing technology landscape has grown from 150 tools in 2011 to over 15,000 today, the majority of new entrants promising to help brands navigate, optimise for, or extract value from platform systems they do not control. The tools have multiplied in direct proportion to the dependency.

What dependency costs

The commercial cost of platform dependency is usually framed as efficiency risk: the possibility that an algorithm change will disrupt performance metrics and require additional investment to restore. That framing significantly underestimates what is actually at stake.

The deeper cost is strategic. A brand that has allowed its discovery, its engagement, and its customer retention to become primarily platform-dependent has outsourced the most valuable part of its commercial relationship, the direct connection with the people who buy from it, to entities whose interests are structurally misaligned with its own. Platforms need brands to keep spending. They do not need brands to succeed. Those are different objectives, and over time the difference matters enormously.

For marketing leaders specifically, platform dependency creates a performance volatility problem that connects directly to the short tenure the first piece in this series documented. When discovery, engagement, and conversion are mediated primarily through platforms, the marketing leader’s commercial case rests on metrics that can be disrupted by decisions made in Menlo Park or Mountain View with no warning and no consultation. When performance drops, as it periodically will, the accountability falls on the marketing function regardless of where the actual cause sits.

The CMO who builds direct customer relationships, owned engagement infrastructure, and genuine epistemic authority in their category is building something different: a commercial foundation that does not depend on algorithmic favour, that compounds independently of platform decisions, and that makes the case for marketing’s contribution in terms that are durable rather than volatile. That is not only a better brand strategy. It is a better survival strategy for the leader responsible for delivering it.

The direct relationship imperative

The alternative to platform dependency is not tactical diversification across more platforms. Adding TikTok search optimisation to an existing Google and Meta dependency does not reduce exposure to technofeudal rent extraction. It extends it.

The genuine alternative is the deliberate, long-horizon investment in direct customer relationships, and that investment has two distinct components that are often conflated but are strategically different.

The first is CRM as strategic infrastructure rather than retention tactic. Most organisations treat customer relationship management as a post-acquisition function: something that activates once a customer has been acquired through a platform-dependent channel. That sequencing is both expensive and strategically backwards. The organisation that invests in building direct relationship capability, owned data, direct communication channels, proprietary engagement environments, genuine value exchange that gives customers a reason to maintain the relationship independently of any platform, is building an asset that appreciates over time in a way that acquisition spend cannot. The customer who has opted into a direct relationship is not rented. They are owned in the genuine commercial sense, and the value of that relationship compounds with every interaction rather than requiring repurchase. That compounding depends on one condition: the owned channel must deliver genuine value to the customer, not simply redirect the broadcast logic of platform advertising through a cheaper pipe. (Check your Gmail promotions folder for a working definition of what misuse looks like.)

The second component is building engagement environments that give audiences a reason to seek the brand out rather than waiting to be found. A well-designed application that delivers genuine utility, not a brochure in app form but something that makes a recurring task easier, more informed, or more enjoyable, creates a direct engagement channel that no platform controls. A community built around genuine shared interest rather than brand promotion creates relationship density that travels independently of algorithmic recommendation. These are not cheap investments, and they are not quick ones. But they are investments in sovereignty rather than dependency, and the return compounds in ways that platform spend structurally cannot.

Genuine epistemic authority

The most demanding and most valuable form of direct relationship investment is genuine epistemic authority, and it is worth being precise about what that means, because the term is used loosely enough to cover almost anything a marketing department produces.

Epistemic authority in a category means being the organisation that serious people in that category actually turn to when they need to think something through. Not the organisation with the highest content volume, the strongest domain authority score, or the most sophisticated distribution infrastructure. The organisation whose perspective genuinely advances the conversation: that says things other voices in the category are not saying, defends positions rather than presenting options, and is willing to be inconvenient to its own commercial interests when the evidence points that way.

That is a specific and demanding editorial commitment, and most organisations are not currently structured to sustain it. The most common failure mode for organisations attempting to build thought leadership is producing content that looks like genuine insight but is structurally incapable of generating it, because every piece has been reviewed for commercial risk, softened for broad appeal, and stripped of anything that might cause friction with a potential buyer. The result is content that is well-produced, consistently distributed, and largely indistinguishable from the content produced by every other organisation in the category. It does not build epistemic authority. It builds content inventory.

Genuine epistemic authority requires the organisation to hold and defend actual positions. Not “here are five considerations” but “here is what we believe is true, here is why the consensus view is incomplete, and here is what the evidence actually shows.” It requires an editorial function with genuine independence: the ability to publish something that a significant client might disagree with because the argument is sound. And it requires patience, because epistemic authority is not built through a single landmark piece or a well-promoted report. It accumulates through the consistent demonstration, over time, that this organisation’s thinking is worth seeking out.

The organisations building this seriously are rare. They are also, not coincidentally, the ones whose commercial conversations begin from a position of genuine credibility rather than a request for attention.

The AI dynamic: a provisional view

AEO and GEO are now established practice areas in their own right, with dedicated agencies, specialist functions, and meaningful budget allocations in most large marketing organisations. The tactical interventions they offer are real: structured data, content architecture, citation-friendly formatting, and entity optimisation all have genuine merit in specific contexts.

The strategic problem is not that these disciplines exist. It is that they are structurally identical to the SEO industry they position themselves to complement, promising brands that technical optimisation will secure algorithmic favour in systems they do not control and building a new layer of platform dependency on top of the existing one.

The brands that invested most heavily in SEO authority over the past decade did not escape the platform trap. They deepened it. There is no obvious reason to expect a different outcome here, unless the underlying investment logic changes from optimising for platforms to building the kind of genuine authority that any intelligent system, trained or otherwise, will surface on its own terms.

The more important observation, and a deliberately provisional one, is directional rather than prescriptive: feeding large language models looks very different from building a distinctive brand narrative. Optimising content for AI ingestion is a production and technical challenge. Building the kind of genuine authority and behavioural coherence that means any intelligent system encountering a brand will represent it accurately and favourably is an editorial and strategic challenge of an entirely different order.

What does seem clear is that AI answer engines draw on breadth and depth of source material, and that sources with genuine authority, consistent positioning, and substantive content appear to carry more weight than sources optimising primarily for algorithmic visibility. If that dynamic holds and develops, and there are reasonable grounds to expect it will because it serves the user interest these systems are ultimately designed to serve, then the investment in genuine epistemic authority is also the investment most likely to produce durable AI visibility. Not because it has been optimised for that outcome, but because it has produced something real enough to be accurately represented.

That is a different brief than most organisations are currently giving their content functions. And it is a brief that most content functions are not currently structured to deliver.

The coherence connection

There is a direct line between the argument of the previous piece in this series and the argument of this one, and AI search makes it visible in a way that was previously easier to ignore.

When an AI answer engine summarises what a brand is, what it stands for, and what it delivers, it draws on the totality of what exists about that brand across the open web: its own communications, yes, but also customer reviews, employee commentary, editorial coverage, community discussion, and the gap, where it exists, between what the brand claims and what its audiences have actually experienced. A brand with genuine coherence between its stated identity and its actual conduct will tend to be summarised in ways that reinforce its positioning. A brand managing that distance through communications will find that AI surfaces the distance rather than the messaging.

This is not a new threat requiring a new tactical response. It is the authenticity argument restated in a search context. The answer is the same: close the distance between what you say and what you do, build direct relationships that give you standing independent of platform mediation, and develop the kind of genuine authority that travels on its own merits. No optimisation strategy substitutes for that. And no optimisation strategy is necessary for a brand that has built it seriously.

Four questions worth asking

The dependency audit. Map the journey from discovery through conversion to retention for your most valuable customer segment. At each stage, identify which touchpoints are platform-dependent, reliant on algorithmic favour, paid visibility, or rented reach, and which are owned. The ratio tells you more about your strategic exposure than any channel performance report.

The CRM investment question. Is customer relationship management in your organisation a strategic priority with its own investment case, or a retention tactic funded from what remains after acquisition spend? If the latter, the sequencing is backwards, and the commercial case for reversing it, built on customer lifetime value rather than acquisition cost, is almost certainly stronger than the current budget allocation reflects.

The epistemic authority test. Look at the last ten pieces of content your organisation published under a thought leadership or brand narrative brief. How many of them defend a specific position that a significant proportion of your audience might initially disagree with? How many advance the category conversation rather than summarising it? How many would be published unchanged if a major client asked you to reconsider? The answers locate you on the spectrum between genuine epistemic authority and expensive content inventory.

The editorial courage question. Does your organisation have the structural conditions required to build genuine epistemic authority: an editorial function with meaningful independence, leadership willing to hold positions under commercial pressure, and the patience to measure authority accumulation over years rather than quarters? If not, the investment in content production is unlikely to produce the return the brief implies, and it is worth being honest about that before the budget is committed.

Building toward resilience

The argument this piece makes is ultimately about foundations. Platform dependency is not a search strategy problem. It is a sovereignty problem, and sovereignty, once lost to the compounding logic of feudal dependency, is expensive and slow to recover.

The organisations that will navigate the continued fragmentation of the search landscape most effectively are not those with the most sophisticated optimisation capabilities. They are those that have built direct customer relationships deep enough to survive platform disruption, epistemic authority genuine enough to travel independently of algorithmic recommendation, and the kind of behavioural coherence that means any surface on which they appear, human-curated or AI-generated, represents them accurately.

That is a long-horizon investment with a compounding return. It is also, as the final piece in this series will argue, the foundation on which brand resilience during genuine market volatility is built. The brand that owns its customer relationships and its authority in a category is not invulnerable to disruption. But it enters disruption from a position of genuine strength rather than managed dependency, and that difference, when the disruption arrives, turns out to matter enormously.

Further reading

Yanis Varoufakis, Technofeudalism: What Killed Capitalism | Shoshana Zuboff, The Age of Surveillance Capitalism | Cal Newport, Deep Work  – calnewport.com/writing | Forrester research on direct customer relationships  – forrester.com | Edison Research on audio and search fragmentation  – edisonresearch.com | Harvard Business Review on platform dependency  – hbr.org | chiefmartec on the evolving search landscape  – chiefmartec.com