In this article
Trust but verify means starting from an assumption of good faith and then making delivery visible enough that no one has to take anything on faith. Applied to your media agency, the principle is straightforward: trust is the working basis of the relationship; verification is what makes that basis sustainable. The two are not in competition.
You have appointed an agency because you believe it can deliver. That belief is not naive. It reflects a selection, a contract, and an agreed scope of work. The question this article addresses is what happens after the agreement is signed, when the work is live and the invoices are arriving.
Why trust and verification are not opposites
The phrase “trust but verify” is sometimes read as a polite way of saying you do not quite trust someone. That reading is wrong, and acting on it leads to the wrong kind of relationship with your agency.
Trust, in a commercial context, is the assumption you extend at the start. You assume that your agency is acting competently, in good faith, and in your interest. That assumption is a working condition. Without it, every conversation becomes an interrogation and the relationship degrades quickly. A damaged working relationship may become expensive: the agency you depend on can become less candid, less proactive, and harder to hold to account precisely when you need to.
Verification is a different thing entirely. It is the mechanism that keeps the assumption of good faith from having to be a leap of faith. When delivery is visible, when what was agreed, what was delivered, and what was charged can be laid alongside each other, no one needs to trust that things are fine. They can see that things are fine. Or see where they are not.
The two together produce something that neither provides alone.
Why an unverifiable arrangement is structurally weak
Where nothing can be checked, the quality of the outcome rests entirely on the quality of the people on the account on any given day. That is a significant dependency. Good people can have difficult periods. Account teams can change. Definitions can drift because no one wrote them down precisely in the first place. None of that requires bad intent, and all of it can cost money.
A trust but verify arrangement removes that dependency. The reference point was agreed in advance. The information arrives in a form that allows comparison. The comparison happens at defined moments. Whether the team changes, whether definitions have shifted slightly, whether the quarter was unusually pressured, none of that changes the standard against which delivery is measured.
This is a structural argument, not a judgment about any particular agency. Even excellent agencies benefit from a fixed reference. When delivery is measured against something agreed in advance, disagreements about what “delivering well” means do not have to be resolved in conversation. The data resolves them.
What verification does for the agency
This is the part most guidance on this topic misses. An agency that delivers well has a direct interest in that delivery being visible.
Where nothing is checked, good work and adequate work look the same. The agency that goes further, that negotiates sharper rates, achieves better placement, or delivers above the agreed metrics, gets no credit for it, because there is no record against which to demonstrate the outperformance. A verified arrangement is the only one in which delivering above commitment can be shown rather than asserted.
And that cuts both ways. A check can equally reveal that an agency has outperformed what it committed to, and that finding deserves to be recorded exactly as a shortfall would. A verified relationship benefits both sides: it protects you against undetected underdelivery, and it gives your agency a mechanism for demonstrating when it has done better than the contract required.
This is why framing verification as a sign of distrust is not only wrong but actively unhelpful to the agency. A good agency should prefer an arrangement that makes its performance visible.
A verified arrangement also settles what “delivering well” means. The agreement is a floor, not a target: on or above what was committed is what is expected, in every period. That matters in both directions. Where results come in better than expected because the market moved - a school strike that puts teenagers in front of the television or behind their computer, a competitor pulling a campaign - the reach is real and it counts, but it is not evidence that the buying was sharp. Without a reference point, nobody can tell the difference between a good buy and a good week. With one, both sides can.
The difference between an error and a pattern
In any relationship of this complexity, individual deviations from what was agreed will occur. A placement runs on the wrong platform for a week. A rate agreed in the contract appears incorrectly in one month’s invoice. These things happen. They are corrected and closed.
But here is where verification earns its place. The distinction it makes possible is between a single deviation and a pattern. A deviation that recurs in the same direction, consistently in the same area, consistently in the same favour, is a managed inconsistency. That term describes what the figures show without attributing a motive. It may be a system error. It may be a misunderstood clause in the contract. It may be something else. What matters is that without a fixed reference point and a regular comparison, every deviation looks like a one-off, and no pattern ever becomes visible.
This is also why the information you receive, and the form it arrives in, determines what can be seen at all. That is set out in what information you should receive from your media agency.
Why advertisers commission verification, and what they verify
The ordinary reasons an advertiser introduces a verification process are unremarkable. A new CFO asks for independent assurance. A corporate governance procedure requires it. A contract clause stipulates it. Finance needs confirmation that what was invoiced matches what was delivered. Or the advertiser simply has not had an independent view in several years and recognises that this is a gap.
Commissioning verification implies nothing about the relationship. Treat it as standard practice, because it is. If the process is handled competently, the agency understands this. The check is against the contract, not against the agency’s character.
In my experience, agencies understand this, though there is sometimes some resistance at first. That is a reasonable reaction: another party is being introduced into a relationship that was working, and it takes explanation. Once that explanation lands, the objection tends to fall away, because the purpose is not to catch anyone out. It is to get more out of the same media budget. No agency can reasonably be against that.
What is actually verified is straightforward: what was agreed, what was delivered, and what was charged, each expressed in a form that allows the three to be placed alongside each other. The practical shape this takes belongs to the media performance management programme rather than this article. What matters here is the principle: a reference point, information in a comparable form, and moments at which the comparison is made.
It is worth being clear about the limits. What can be verified depends on what the arrangement was built to capture. Where the information regime is set up properly, the things that matter sit inside it by design: what was agreed, what was planned, what ran, what was charged. What sits outside it is harder, and no arrangement makes that impossible. Figures can be presented in ways that are technically correct and still hard to read back, and a commitment made outside the agreed process leaves no trace in it. That is why the regime is defined at the start rather than assembled afterwards, and why a fixed reference matters more than the volume of data you receive.
This approach is part of what vendor management after the pitch addresses more broadly, the ongoing discipline that keeps the commercial relationship honest and productive after the contract is signed.
A useful independent check requires neither suspicion nor a full forensic audit. Where an agency of record is already in place, what that arrangement does and does not cover is worth understanding before deciding where the verification should sit.
How this runs through the media year, and why fixed moments catch what a single annual look does not, is set out in the annual cycle of a performance management programme.
Across complex multi-market programmes, advertising across Southeast Asia or managing campaigns across several European markets simultaneously, the case for a fixed verification structure is stronger, not weaker. The variables multiply, and the chances that a deviation in one market is noticed without a structured process diminish accordingly.
Closing
Trust is good. Verifying is better. And it is better for both sides.
An arrangement in which delivery is visible does not require anyone to extend trust as a leap of faith. It creates a shared standard: the agency knows what it is measured against, the advertiser can see what it is getting, and any deviation, whether it favours you or the agency, is recorded rather than lost. That is a stronger foundation for a working relationship than trust alone, and a considerably more honest one.
Frequently asked questions
Should you trust your media agency? Yes, and start from that assumption. Trust is the working basis of the relationship. Verification is what keeps it from having to be a leap of faith, for either side.
Does verifying delivery damage the relationship with your agency? Not when it is handled correctly. A fixed reference point agreed in advance replaces suspicion with a shared standard. It is also the only arrangement in which good delivery can be demonstrated rather than claimed, which is in the agency’s interest.
What do you actually verify? What was agreed, what was delivered, and what was charged, each expressed in a form that allows the three to be placed alongside each other. Where those three align consistently, the arrangement is working. Where they diverge, you have a finding worth investigating.
What did Reagan mean by “trust but verify”? Reagan used the phrase in the context of arms reduction treaties, extend diplomatic trust, but require independent confirmation that commitments are being met. The logic transfers directly to commercial relationships: the assumption of good faith is the starting point, and independent confirmation is what makes that assumption stable.
What is a managed inconsistency? A managed inconsistency is a deviation from what was agreed that recurs in the same direction across multiple periods. It is a descriptive term for what the data shows, without attributing a motive. Identifying managed inconsistencies is one of the key things a structured verification process makes possible.
