WHY AI HYPE IS OUTPACING AI READINESS IN SPORTS
Every athletic department, team, and league office is being told the same thing right now: adopt AI or get left behind. Vendors are pitching predictive churn models, generative content tools, and AI-powered fan engagement platforms. Conference calls are filled with questions about ChatGPT, Claude, and custom copilots. Budgets are being carved out for pilots.
And yet, most of these initiatives will underperform, stall, or quietly disappear within eighteen months. Not because the technology is bad. Because the organizations deploying it haven't done the foundational work that makes AI useful in the first place.
That foundational work is systems thinking. And the most important expression of systems thinking inside a modern sports organization is something deeply unglamorous: a system of record.
Before a college athletic department can reasonably ask an AI to predict which donors are likely to lapse, which ticket buyers will upgrade to season plans, or which fans are worth re-engaging, it has to answer a much simpler question. Where does the authoritative version of that fan live? If the answer is "in seven different places, depending on who you ask," then AI isn't the next step. Systems are.
WHAT A SYSTEM OF RECORD ACTUALLY IS
The term "system of record" gets thrown around loosely, so it's worth defining clearly. A system of record is the authoritative source of truth for a given type of data inside an organization. It's the place where, if two systems disagree, you go to settle the argument.
In finance, the general ledger is the system of record for transactions. In HR, the HRIS is the system of record for employees. In a hospital, the electronic health record is the system of record for patients.
In sports organizations, the question of "what is the system of record for a fan" has historically been muddy. Ticketing platforms hold purchase history. Donor management systems hold giving history. Merchandise platforms hold transaction data. Email marketing tools hold engagement data. Media and streaming platforms hold consumption data. Each of these systems was built to be authoritative for its own slice of the relationship, but none of them were designed to be authoritative for the person.
The result is that most sports organizations don't have a system of record for their customers. They have fragments. A single fan might exist as seven partial records across seven systems, with seven different email addresses, seven different spellings of their name, and seven different behavioral histories that nobody has ever reconciled.
This is not a problem AI can solve. This is a problem AI inherits.
THE SYSTEMS THINKING GAP
Systems thinking, as a discipline, is the habit of understanding how parts interact to produce outcomes. It's the opposite of point-solution thinking, where you address problems in isolation and hope the pieces add up.
Most sports organizations have been operating in point-solution mode for a long time, and for understandable reasons. Budgets are constrained. Staffs are lean. Vendors show up with a specific problem and a specific fix, and the fix gets implemented, and the team moves on to the next fire. Over a decade of this, you end up with a stack of tools that each solved yesterday's problem and collectively created today's problem.
The rise of AI is exposing this gap in an uncomfortable way. AI models are, fundamentally, systems that reason over data. Their quality of reasoning is bounded by the quality, completeness, and consistency of the data they reason over. You cannot paper over fragmented data with a sophisticated model. The model just learns the fragmentation.
A CRO who asks an AI to predict renewal likelihood for season ticket holders will get a prediction. Whether that prediction is worth acting on depends entirely on whether the underlying customer records are complete and unified. If the AI only sees the ticketing system's view of the fan, it's making a prediction based on maybe 20% of the relevant signal. The donor history, the merchandise purchases, the email engagement, the streaming behavior, the stadium concession spend, the family membership structure—none of that is visible to the model unless the data has been brought together first.
That bringing together is systems work. It happens before the AI ever runs.
WHY THE CDP CONVERSATION MATTERS NOW
A customer data platform is, at its core, an attempt to create a system of record for the customer across a distributed stack of operational tools. It sits underneath the ticketing system, the donor system, the merchandise system, and the marketing system, and it assembles a unified, persistent, queryable view of each fan.
This is not a new idea in other industries. Retail, travel, media, and financial services have been building toward unified customer records for fifteen years. What's new is the urgency of the conversation in sports, driven by two forces.
The first is the collapse of third-party tracking. Cookie deprecation, Apple's privacy changes, and tightening regulation mean that the old playbook of renting audience data from intermediaries is dying. First-party data, the data an organization owns directly about its own fans, is becoming the only durable asset.
The second is AI. Every AI use case a sports organization might pursue depends on having clean, unified, first-party data about its fans. Predictive lifetime value requires it. Personalized content recommendations require it. Automated donor cultivation requires it. Even simple use cases like "write a segment-specific email" fall apart when the segment definitions are based on fragmented data.
Organizations that treat the CDP conversation as a precursor to the AI conversation are thinking in systems. Organizations that treat them as parallel or interchangeable are setting themselves up for expensive disappointments.
THE THREE LAYERS OF SYSTEMS MATURITY
It's useful to think about systems maturity in a sports data context as a stack of three layers. Each layer is a prerequisite for the next.
LAYER ONE: CONNECTION
The first layer is simply getting the data out of the operational systems and into a place where it can be worked with. This sounds trivial and almost never is. Ticketing platforms have their own APIs and exports. Donor systems often require manual pulls. Legacy systems might require custom integrations or vendor cooperation that has to be negotiated. Marketing platforms have rate limits and schema quirks.
Most sports organizations are still somewhere in this layer, with some data flowing automatically and other data being pulled manually by a staffer running reports every Monday morning.
LAYER TWO: RESOLUTION
The second layer is identity resolution. Once you have the data in one place, you have to figure out which records belong to the same person. The same fan might appear as "Michael Smith" in the ticketing system, "Mike Smith" in the donor database, and "M. Smith" on the merchandise side, with different email addresses across each.
Deduplicating and resolving these records is the work that turns a pile of data into a system of record. It's also the work that most organizations dramatically underestimate in both complexity and importance. Bad identity resolution produces bad segments, bad predictions, and bad fan experiences.
LAYER THREE: ACTIVATION
The third layer is activation. Once you have unified, resolved records, you need to be able to use them. That means pushing clean segments back into the email tool, the ad platform, the donor outreach system, and the ticketing campaigns. It also means making the unified record accessible to the people who need it, whether that's a development officer preparing for a major gift conversation or a marketing coordinator building a renewal campaign.
Only after all three layers are in place does AI become a reasonable next conversation. An AI model trained on unified, resolved, activatable fan data can produce meaningful predictions and recommendations. An AI model pointed at fragmented operational data will produce expensive nonsense.
WHAT THIS MEANS FOR CROS, CFOS, AND DATA LEADERS
The practical implication for sports executives is that the sequencing of investment matters. A department that spends its AI budget on a generative content tool before it has resolved the fragmentation in its fan database is buying an output layer without an input layer. The content will get generated. Whether it will be relevant, personalized, or effective is a different question.
The more durable path is to invest in the systems layer first. Establish a system of record for the fan. Get the ticketing, donor, merchandise, marketing, and media data flowing into one place. Do the identity resolution work. Make the unified record the foundation of how the organization operates.
This is unglamorous work. It doesn't produce a demo-able moment. It doesn't generate a press release. But it is the work that determines whether every subsequent AI investment pays off or evaporates.
The organizations that will benefit most from AI over the next five years are not the ones moving fastest on AI. They're the ones that did the systems work first and now have a foundation that AI can actually leverage. Thinking in systems is not a detour from the AI conversation. It is the AI conversation, just at the layer where the real leverage lives.
