The next black box is already being built. Media leaders need to own their governance model.
Governance has always been core to the job.
Brand-side media leaders have been doing it for years. We’ve acknowledged the misaligned incentives across the industry: most of the media ecosystem gets paid on something other than the brand’s results. Agencies can earn more on inventory they resell than on the fees they disclose. Ad tech takes a percentage of every dollar that passes through it, so more spend means more revenue, whether or not it works. Everyone in the chain can hit their number while the brand misses its own.
That’s why the industry has spent the better part of a decade circling the same problems: transparency gaps in the programmatic supply chain, non-transparent buying practices at media agencies – and newer gaps taking shape, like how much influencer and creator investment actually reaches creators versus the agencies behind them.
The industry isn’t designed to maximize advertiser value. There are no shared incentives around media quality, and there are no standards for the data provided to brands to audit their own spend. Because no one else in the chain is incentivized to fix this, it falls to the media leader: build the standards for data, tech, and best practice that steward the investment toward return.
So why the additional urgency to own governance now?
The incentive problem is staying. The complexity is getting worse.
The media landscape is at another crossroads with agentic capability. If built with transparency in mind, it could give media leaders more visibility into their investment than they’ve ever had. However, the risk is the opposite: a new generation of black boxes, run by algorithms, with less observability than what came before. Which way it goes depends on who owns the governance.
- Agentic is already reshaping buying. Even if you’re not yet scaling agentic capability inside your walls, automated buying is reshaping the media landscape around you.
•
Search is the clearest example. AI answers are changing how people find information and pulling clicks away from traditional paid search and eyeballs away from traditional publishers. Automation has also handed more of the buy to the platform: which queries you match, where your ads show, and how bids get set. As a result, advertisers are getting more automation and less visibility into what it’s doing.
•
Programmatic is next. DSPs, SSPs and agencies are all building agents that plan, negotiate and optimize buys, and how buyers and sellers engage and transact will transform fundamentally within a few years. Will those new systems be built with brand transparency in mind? Not by default. The incentives haven’t changed, so brands will have to ask for it.
• - The lock-in is being built into the agency operating systems. Agency operating systems are designed to keep your investment, data and workflow managed inside their walls, including the agentic tools they’re building now. Change your partners – and the audiences, historical data, and agents that ran on it often stay behind. Opting into elements of your agency’s operating system and tech can be the right call. However, defaulting into it without your own governance and contractual visibility isn’t a strategy.
Fast change is the normal state now, the risk to advertiser value is inherent to it, and governance is what protects it: your standards, your decision rights, and your view into the data, however fast the buying changes. That governance has to be built for both the now and the next.
The dual horizons
Media leaders need to build their governance models with dual horizons:
- Get the fundamentals right today. Can you prove your media is running the way you think it is? That comes down to the basics: quality standards by channel, clear decision rights with your partners, measurement you trust, and a clear read on where the money goes. Get them right and you recover value today. They’re also the baseline that future capability builds on.
- Build for an agentic tomorrow. Can you trust your media as machines start making the calls? That comes down to observability. When buying is automated, you need to see the data underneath it, the logic the agents are using, and the decisions they’re making. See that, and you can trust what’s running and make it better. Miss it, and you’re taking the machine’s word for it.
It’s easy to fall into one mindset or the other: fix today’s execution or chase the pilots. Both are the same mistake. Your media fundamentals and the future aren’t two separate projects; instead, they’re moving together. Perfect today’s buy while ignoring what’s coming and you’re ready for a market that’s gone. Chase the future while the basics leak money and you’re not supporting the business. The answer is both, on purpose, at once.
What a governance model actually is
This isn’t about pulling media buying in-house. Instead, it’s about brand-side media leaders having clarity on what decisions they need to own.
A governance model starts with the operating framework. It spans people, process and tools, and sets the standards, decision rights, contracts and data access that sit across every partner and channel. Ultimately, it gives you the ability to see, govern, and decide on your own media, no matter who runs it.
It comes down to three lenses:
1. Capability & standards Your own definition of what good media looks like: the data, technology, standards, and learning agenda your media has to run against, set by you rather than only your partners. That includes named quality standards by channel, the data and tech you own outright, and an explicit agenda for what you’re testing next. Without it, you’re governing to someone else’s definition of good.
2. Operating control The operating model that makes those standards real: decision rights, the contractual obligations behind them, and a review cadence that confirms every channel and partner is actually running to your standard – across programmatic, social, search, retail media, influencer, and offline. It answers three questions. Who decides? On what basis? And how do we know it happened? Without it, your standards are a document, not a practice.
3. Observability Continuous visibility into the data, logic, and decisions underneath your media, not just the outcomes reported back to you. That means log-level and platform data you can access on your own terms, visibility into how algorithms and agents are reaching decisions, and contractual access rights that guarantee that visibility holds as buying automates. Reporting tells you what happened. Observability tells you why, and whether you’d have made the same call.
Writing the standard is the easy part. Holding partners to it, and proving they did, is the job.
Assessing your readiness
Start with these questions. Your answers show where your model is strong and where it’s exposed. The NOW column tests the foundation you have in market today. Meanwhile, the NEXT column tests whether that foundation is ready for an evolved, agentic-first industry.
Case study: start with what’s in market today
Can’t answer many of these yet? That’s the most common starting point: the first step is to get a clear read on what’s in market today. A current-state assessment does two things at once: it finds value you can recover now, and it sets the baseline your future-facing roadmap builds from.
Be ready for what you’ll see
Governance only works if you’re willing to see the answer. A genuinely transparent model means you’ll see where your partners make their margin: how much of your investment becomes someone else’s profit, and where.
The discomfort isn’t external; it’s internal. Once you can see how much of your investment is “working” versus “non-working,” you own that number. Media leaders should want it anyway. After all, it’s a better basis to establish meaningful partnerships, and it’s the only way to get more value out of the total investment.
This is the work Transparent Partners does with brand-side media leaders. We assess what’s in market today, find the value you can recover now, and build the standards, contracts and observability your agentic roadmap depends on. If any of these questions are hard to answer today, that’s where we’d start.
Adam Benaroya, VP of Media Solutions
