ChatGPT Enterprise: How the Spurs Scaled Fan Impact

Most organizations trying to “roll out AI” make the same mistake: they treat adoption like a software deployment instead of a culture change. The San Antonio Spurs took the opposite approach—and got measurable results fast.

In a public case study, the Spurs reported 1,800 staff hours saved per month, AI fluency rising from 14% to 85%+, and 94% of users saying they feel more confident using large language models (LLMs). Those numbers matter beyond sports. They’re a clean example of how AI is powering technology and digital services in the United States—not through flashy demos, but through repeatable operational practices.

This post is part of our AI in Media & Entertainment series, where we track how AI is reshaping content, personalization, audience insights, and internal workflows. The Spurs story is a strong reminder: entertainment brands don’t “do AI” as a side project. They build an AI operating rhythm that helps them ship better experiences, faster.

The real win: scaling communication without scaling headcount

AI creates outsized returns when it improves how information moves—between teams, and between brands and audiences. That’s the Spurs’ core insight, and it’s the same one driving adoption across U.S. SaaS, media, and digital service companies.

Sports is basically a high-frequency media business. You’ve got constant content deadlines (games, highlights, social posts), constant customer feedback (fans), and constant partner reporting (sponsors). That combination produces a familiar pain:

• Feedback piles up faster than humans can summarize it
• Content needs to be localized for different markets and languages
• Staff spend hours reformatting insights for different stakeholders
• Knowledge lives in scattered docs, Slack threads, and someone’s memory

The Spurs used ChatGPT Enterprise to reduce that friction. Not by replacing jobs, but by compressing the time between signal → decision → action.

Here’s the stance I’ve developed after watching AI rollouts succeed and fail: “Productivity” is a weak goal. Cycle time is the real goal. When teams shorten cycle time, the fan experience improves, partners get answers sooner, and internal teams stop drowning in manual busywork.

A people-first AI rollout beats a tool-first rollout (almost every time)

The Spurs optimized for comfort and fluency first, and usage quotas second. That’s why adoption stuck.

According to the case study, they started with a 150-person pilot, supported by in-person training, onboarding guides, and internal experimentation events (including hackathons). The key phrase to focus on is “phases.” They didn’t force everyone onto day-one change.

Why “phased adoption” works in U.S. organizations

U.S. companies typically have a few traits that make AI rollouts tricky:

• Teams move fast and hate extra process
• Many roles already feel overloaded
• Legal/compliance concerns can slow decision-making
• Different departments have different definitions of “value”

A phased rollout gives you a place to learn what breaks before you scale it. It also gives you internal champions—people who can say, “I used this on a real deliverable and it helped.” That peer credibility beats any top-down memo.

What to copy: a simple adoption playbook

If you’re building an AI program inside a media, entertainment, or digital services organization, steal these operating choices:

1. Pilot with motivated teams, not “representative samples.” You want momentum.
2. Train in-person when possible (even one session). AI anxiety drops when people can ask blunt questions.
3. Reward prototypes, not polish. Early wins should be quick and slightly messy.
4. Publish internal examples (prompts, workflows, before/after). People learn by imitation.

The Spurs’ own leaders framed it clearly: start with goals, not tools. That’s a mature approach—and it’s why their results sound like operations metrics, not marketing.

Purpose-built GPTs: where AI becomes a real business system

Generic chat is helpful, but purpose-built GPTs are where organizations start to compound value. The Spurs built multiple custom GPTs for specific workflows across community engagement, operations, partnerships, retail, and internal culture.

This is the part that maps cleanly to SaaS and digital services: a purpose-built GPT is basically a lightweight internal product. It’s repeatable. It standardizes quality. It turns “one person’s talent” into “a team capability.”

Example 1: Fan sentiment at the speed of the internet

The Spurs’ Fan Voice GPT scans thousands of post-game comments and produces sentiment summaries in minutes—work that previously took hours.

For media and entertainment brands, this is the most transferable pattern in the whole story:

• Fans generate high-volume feedback (comments, reviews, support tickets)
• That feedback contains insights, but it’s unstructured and emotional
• Teams need summaries quickly enough to act before the next content cycle

If you run a streaming brand, a sports franchise, a publisher, or even a large creator business, the “Fan Voice” pattern translates directly to:

• Episode-level reaction summaries
• Launch-day sentiment for a new feature
• Creator community feedback triage
• Customer support trend detection

AI doesn’t replace listening. It makes listening fast enough to matter.

Example 2: Localization that doesn’t dilute your brand voice

The Spurs also reported GPT solutions that engage fans in Spanish and French, supporting their push to connect authentically in international markets.

Localization is where many entertainment brands stumble. They either:

• Translate literally (and lose tone)
• Rewrite from scratch (slow and inconsistent)

A well-configured GPT can enforce a voice guide, standard terms, and cultural context. For U.S.-based organizations expanding globally, that’s not a “nice to have.” It’s a growth requirement—especially when international audiences expect native-level content quality.

Example 3: Partnership reporting that sponsors can actually use

The Spurs built a Partnership Insights GPT that converts sponsorship data into clear, actionable insights.

This is quietly huge. Sponsorship and advertising live or die on reporting clarity:

• What did the partner get?
• Where did visibility happen?
• What should change next month?

AI can reduce the hours spent turning dashboards into narratives. More importantly, it can standardize how those narratives are written so partners don’t get a different story depending on who made the deck.

If you sell digital services, think of this as the AI version of a “monthly business review” template—except it’s faster, consistent, and easier to tailor to each stakeholder.

Example 4: Trust and culture aren’t soft benefits—they’re operational

One of the Spurs’ most interesting builds is a Spurs Culture GPT, trained on a long internal docuseries, to help staff incorporate values and milestones into everyday work.

A lot of AI content focuses on automation. I’m more interested in alignment.

When companies grow, culture becomes harder to transmit. Entertainment brands feel this sharply because brand voice is culture. A culture/voice GPT doesn’t just make writing easier—it reduces the odds of:

• Off-brand messaging
• Inconsistent community responses
• Training gaps across new hires

It’s essentially a scalable “brand brain,” and it’s an underused pattern in media & entertainment AI strategy.

ROI in months is possible—if you measure the right things

The Spurs reported ROI quickly because they tracked outcomes people care about. Their results weren’t vague:

• AI fluency: 14% → 85%+
• Time saved: 1,800 hours/month
• Confidence: 94% report greater confidence with LLM tools

Those are adoption and productivity metrics, but there’s an even more important media & entertainment metric sitting underneath them:

The metric that actually moves revenue: speed to experience improvement

When you can summarize fan sentiment in minutes, you can:

• Fix venue friction points (lines, wayfinding, concessions) faster
• Adjust content packaging and social messaging in near-real time
• Flag problems before they become reputational issues

In digital services terms, you reduce the time between “user pain” and “product change.”

If you want a practical measurement framework, use a simple 3-layer approach:

1. Operational: hours saved, turnaround time, backlog size
2. Quality: fewer revisions, fewer inconsistencies, higher partner satisfaction
3. Experience: fan sentiment lift, engagement rates, repeat purchases

The Spurs published strong Layer 1 and Layer 2 indicators. Many organizations stop there. The next step is tying those gains to experience outcomes (Layer 3) so AI doesn’t get categorized as “just an efficiency tool.”

What leaders should copy next (and what to avoid)

The Spurs model works because it’s human-centered and specific. If you’re a U.S. media company, digital agency, SaaS platform, or sports organization trying to operationalize AI, here’s what I’d implement immediately.

Do this next: build 5 “high-frequency GPTs” before you build 50

Pick workflows that happen daily or weekly, and that touch revenue or reputation. Examples:

• Audience sentiment summaries (comments, reviews, community posts)
• Sponsor/advertiser reporting narratives
• Brand voice and localization assistant
• Customer support triage and macro tagging
• Content brief generation from research + past performance

High-frequency use creates habit. Habit creates fluency. Fluency creates better ideas for the next GPT.

Avoid this: “AI theater” and prompt hoarding

Two failure modes show up constantly:

• AI theater: leadership buys tools, announces a big initiative, and expects magic without training or workflow redesign.
• Prompt hoarding: a few power users get great results, but nothing spreads because examples aren’t shared.

The Spurs countered both with training, hackathons, and internal sharing. That’s the cultural infrastructure most companies skip.

Where this is headed for media & entertainment in 2026

Entertainment brands are becoming AI-powered service businesses. Fans expect personalization, instant responses, and culturally aware content across channels. That expectation won’t cool down in 2026—it’ll intensify.

The Spurs’ next step—building research partnerships around performance, decision-making, and cognitive training—signals something bigger: AI isn’t just a back-office tool anymore. It’s moving into how organizations learn, coach, and improve.

If you’re responsible for digital transformation, marketing ops, customer experience, or revenue partnerships, the question isn’t “Should we use LLMs?” The question is: Which repeatable communication workflows will we standardize with AI first—so we can improve experiences faster than expectations rise?

If you want help translating this case study into your own roadmap—use cases, rollout plan, governance, and measurable ROI—start with two things: one team that’s eager, and one workflow that’s painfully slow. Then build from there.