Cases, patterns and observations from twenty years inside the system.

In every other part of the marketing process, the work gets reviewed.

The creative is scrutinised.

The media plan is questioned.

The results are analysed.

The brief is treated as given.

In a process improvement exercise at a major agency — an effort designed to benefit both client and agency — nearly all the client briefs arriving in a single quarter were found to be incomplete before anyone had assessed whether the thinking was any good.

The commercial context could usually be researched and filled in. What was missing was the hard information only the client could provide — budget, timeline, target audience, success metrics. And the strategic thinking only the client could supply — why this campaign, what the real problem is, what success actually looks like.

Those two things together are what an agency cannot fabricate.

If they are absent, the work proceeds on guesswork.

What happens next

What happens next depends on who receives the brief.

A stronger account lead or strategist will push back and ask for more.

A weaker or more relationship-cautious one will work with what they have.

There is no standard process for this. The quality of the response is determined by whoever happens to catch the brief, and how much friction they are willing to create with the client.

Which means any CMO who assumes their agency has a reliable system for handling incomplete briefs is assuming something that is not usually true.

Why this isn't fixed

This is not a failure of agency process.

Agencies review what they can review. The brief comes from the client. Sending it back creates friction, risks the relationship, delays the project. So the agency works with what it has, makes assumptions in place of strategic clarity, and produces work against an objective that was never sharp enough to begin with.

The BetterBriefs Project surveyed both sides of the relationship and found the full shape of the problem. 78% of marketers believe their briefs are clear. 5% of agencies agree.

The gap between those two numbers describes the volume of work being produced every year against objectives nobody on either side fully understood.

There is an uncomfortable layer underneath.

The Ipsos study commissioned by Mark Ritson in 2026 found that two-thirds of British marketers cannot pass a basic test on the fundamentals of their own discipline. Among those who failed, 83% rated themselves above average.

So the brief is not challenged because challenging it is commercially uncomfortable.

And the person writing it often lacks the foundational knowledge to write a good one in the first place.

Both of those things are true at the same time.

The document that sets the direction for the spend is the one document in the process that never gets a proper review. And the money moves accordingly.

In early 2020, a large health and wellness retailer was cutting costs.

Like every retailer that month, the company was preparing for an unknown. Budgets were being reduced across the board. Paid search was one of the lines that took a significant cut — not to zero, but materially.

The expectation was clear. The attribution model had been crediting paid search with around a third of online revenue. A reduction of that scale would mean a meaningful drop in conversions. The team prepared for it.

The drop did not come.

Online revenue stayed flat. Organic search picked up almost exactly what paid search had been claiming credit for. The customers who had been showing up through the paid ads were still showing up — they were just arriving through a different door.

That distinction matters. The attribution model had not been measuring incremental contribution. It had been measuring overlap. Paid search was capturing demand that organic would have captured anyway.

This is not a new finding in the academic literature. A well-known field experiment on eBay's paid search activity in 2015 found that the vast majority of conversions being credited to paid ads would have happened anyway. The customers were already going to buy.

What is unusual is for a company to discover this in its own data, in its own business, with its own customers. Most companies do not run this test deliberately. The test only happens when something else forces it — a budget cut, a system outage, an accidental switch-off. The 2020 cuts created a moment of forced honesty that no marketing team would have asked for.

What happened next

The honest finding did not produce an honest response.

Some of the paid search budget was put back. Not all of it. Partial restoration.

The reasoning was not that anyone in the team had stopped believing the data. The reasoning was that nobody wanted to be the person who had killed paid search if revenue dipped later for any other reason.

Cutting something that looks important and being wrong is a worse outcome for a career than keeping something unnecessary and being right.

So the budget went back.

This is how a great deal of marketing spend is actually decided. Not against evidence. Against the personal risk of being the one who made the call.

The channel that wasn't doing what it looked like was discovered, measured, and quietly funded again. Because the structural pull of the system is stronger than the data the system produces.

In any other part of a business, when significant money is being spent and the return is unclear, an independent view is available.

External audit for the financial statements.

Commercial due diligence before an acquisition.

In marketing, there is no equivalent.

This is unusual. Marketing is a major line on the P&L of most consumer-facing businesses. The amounts involved are not small. The strategic stakes are not low. And yet there is no established service whose job is to look at the spend, independently, and tell the board whether it is producing what it should.

The absence is not because nobody has thought of it.

It is because the marketing services industry is structurally arranged so that nobody inside it has a commercial reason to provide the answer.

The agency

Agencies are paid on spend. The commercial model of the agency industry — fees, commissions, retainers, principal trading margins — rewards the existence of the budget, not the productivity of it. An agency that helped its client conclude that a third of its marketing spend was unnecessary would be advising itself out of revenue. Agencies do not, as a rule, do this.

This is not a moral failure. It is a structural one. The agency is doing exactly what its commercial model rewards it for doing. Asking it to act against that model is asking the wrong party to solve the problem.

The measurement vendor

Most of the major measurement tools used to assess marketing performance are sold by the platforms whose performance they are measuring.

The tools that tell you whether Google is working are often sold by Google.

The tools that tell you whether Meta is working are often sold by Meta.

There are independent measurement firms, and there are independent methodologies. But the dominant tools used by most companies are not independent in the structural sense. They are produced by parties who have a direct commercial interest in the answer they generate.

A reader from outside marketing would find this remarkable. Inside marketing, it is so normal that most practitioners do not think about it.

The marketing team

The third reason the answer does not exist is the most uncomfortable.

The people inside the company who are best positioned to ask whether the marketing spend is working are also the people whose careers depend on the budget staying where it is. A marketing director who concluded that their function was overspending and successfully argued for a budget reduction would be diminishing the scope of their own role. The structural pressure on that person is to defend the budget, not to question it.

Again, this is not about individuals. Marketing directors are doing exactly what the system rewards them for doing. The point is that there is no one inside the function whose job is to produce an honest answer, because nobody inside the function has an incentive to.

The shape of the absence

So: the agency cannot produce the answer. The measurement vendor cannot produce the answer. The marketing team cannot produce the answer.

The consultancies that do exist in this space tend to sell transformation programmes, not diagnostic engagements. They are too expensive, too slow, and too closely involved in delivery to be independent in the way this question requires.

That is the entire established marketing services landscape. Nobody in it is positioned to give a CEO or CFO a clear, independent, short answer to the question: is this spend producing what it should?

The absence is not a market gap. It is the shape of a market working exactly as its incentives suggest it should.

In September 2025, the interim CEO of Diageo stood up at an investor conference in Boston and described the company's marketing budget in terms a serving CEO of a major company almost never uses.

He said it was not working.

Diageo had been spending more on marketing every year. The spending had not been producing the growth it was meant to produce. So the company had spent more again the following year. And then more again the year after that.

He called it a vicious cycle.

"We've allowed that to continue happening," Nik Jhangiani told investors, "because it was almost this vicious cycle around, well, I need to continue investing there to get the growth. If I'm not getting the growth, I'll put more money in and I'll hope that the growth will come."

Diageo had spent $3.66 billion on advertising and promotional investment in its 2025 fiscal year. The investment had been growing faster than net sales. More money was being spent each year than the previous year's spend had produced.

Acting on the admission was quick. Within weeks the British market had cut its development spend in half and reduced its agency roster by thirty per cent. Advertising and promotional spend was down ten per cent in the half that followed. None of this was framed as cost-cutting. It was framed as a response to a question the company had not previously been able to answer.

Why the question was finally asked

Jhangiani is a finance person.

He joined Diageo as Chief Financial Officer in September 2024 and stepped up to interim Chief Executive in July 2025. The vicious cycle observation came two months into his time in the CEO role.

The structural problem had not appeared overnight. Diageo's marketing budget had been the same shape under previous leadership.

What changed was that the person now responsible for the budget held it to the standards finance applies to every other line on a P&L.

The vicious cycle was not a discovery. It had been visible for years.

What was missing was a person with both the financial discipline to ask the question and the operational authority to act on the answer.

Most companies do not have such a person. The Chief Financial Officer has the discipline. The Chief Executive has the authority. Neither has both. So the question gets asked indirectly. And gently deflected.

Jhangiani had both at once, briefly.

The first thing he did was name the cycle.

The marketing budget at most companies the size of Diageo is being spent the same way Diageo was spending its marketing budget. It will continue being spent that way until someone with both the discipline and the authority asks the same question Jhangiani asked.

In January 2017, Marc Pritchard, P&G's Chief Brand Officer, stood up at the Interactive Advertising Bureau's annual conference and gave a speech that changed the conversation about digital advertising.

He demanded transparency from the platforms. He demanded accountability from the agencies. He named viewability fraud. He named the opaque programmatic supply chain. He set a deadline by which P&G would stop spending money with parties that did not meet basic transparency standards.

The transparency problems Pritchard named in 2017 were real, well-documented, and had been ignored by the industry for years.

The speech worked. Within eighteen months the major platforms had moved on viewability standards. Agency contracts began including transparency clauses that had not existed before. Other large advertisers followed P&G's lead. Pritchard had used the buying power of one of the world's largest marketing budgets to force the supply chain to change, and the supply chain changed.

In the years that followed, P&G itself cut around $200 million of digital advertising spend with no measurable loss in sales. The cuts were widely reported as evidence that significant portions of the company's previous digital investment had not been producing what had been claimed for it.

That admission — that P&G had been overspending without knowing it — is the part of the story worth sitting with.

Where the reform went

The reform Pritchard called for was directed at one part of the system. The part external to P&G's own marketing function.

It did not extend to the other parts. The quality of the briefs P&G was writing. The way its marketing budget was being allocated against the long-term effectiveness evidence. The incentive structures inside its own marketing function. The alignment between marketing's reporting and finance's reporting.

These were not the subject of a comparable public push.

This is not a criticism of Pritchard. It is an observation about which kinds of reform are easier to lead.

A reform directed at agencies and platforms is a reform of suppliers. The CMO who leads it appears as the strong client demanding better service. The internal politics are favourable. The agencies cannot fight back without losing the account. The platforms cannot fight back without losing the spend.

A reform directed at the inside of the marketing function is a reform of colleagues. The CMO who leads it has to challenge the people who report to them, the briefs they have signed off, the budget allocations they have approved, the incentive structures they have helped design. The internal politics are difficult. The colleagues can fight back, and they often do.

The same pattern recurs across the industry. Senior marketing reformers consistently direct their reform energy at the parties they can demand transparency from rather than at the parties they would have to confront internally. Supply chain reform makes the reformer look like a hero. Internal reform makes the reformer look like a critic of their own team.

This is structural. It is why marketing reform, when it happens, almost always points outward.

The supply chain was reformed. The demand side was not. The marketing accountability gap that existed in 2017 still exists today, in many of the same companies whose CMOs once stood beside Pritchard and applauded.

A large bank engaged a major consultancy to review its marketing spend.

The consultancy was a famous one. The kind whose recommendations are difficult to dismiss because of who they come from rather than because of what they say.

The recommendation, when it came, was to optimise the company's paid search investment against the ratio of revenue to cost. Spend more where the revenue per pound spent looked highest. Spend less where it looked lowest. The industry calls this metric Return on Ad Spend.

The principle sounds reasonable.

It was wrong.

Why the metric was the wrong one

The revenue-to-cost ratio is a useful number for some purposes. It is not a measure of whether a channel is contributing to growth.

The two things sound similar. They are not the same.

A channel can show a very high ratio and contribute very little to growth. The clearest example is paid search on the company's own brand name. When someone searches for a company by name and clicks the paid advertisement, the resulting purchase gets attributed to paid search. The ratio looks excellent. But that customer was already searching for the brand. They were going to find the website. The advertisement intercepted demand that would have converted anyway.

A channel can show a much lower ratio and contribute meaningfully to growth. Generic searches, where the customer is exploring options and has not yet decided who to buy from, produce a lower ratio because more of those searches do not lead to a purchase. But these are the searches where advertising is actually doing work — capturing people earlier in the decision and influencing where they end up.

The relationship is the opposite of what intuition suggests.

The higher the ratio, the less likely the spend is contributing to growth that would not have happened anyway.

This is well understood by anyone who has worked seriously in paid search. It is in the academic literature. It is in any decent media planner's basic training. It is not a controversial point.

The recommendation in the bank case was heading in the direction of optimising harder against the ratio — concentrating spend on the activities producing the highest numbers. Which would have produced excellent reported figures and quietly degraded the underlying contribution to the business.

Why it was followed anyway

Everyone closest to the account knew this.

The client did not act on what they knew.

The consultancy was a famous one, and the recommendation came with the authority of the brand on the cover sheet rather than the depth of the expertise inside the work. The internal posture was that the consultancy must be right because the consultancy was the consultancy.

This is the structural observation the piece is really about.

When marketing decisions are made on the basis of which advisor's brand carries the most authority rather than which advisor has the deepest domain expertise, the wrong recommendations win. Not because anyone is being malicious. Because nobody in the room wants to be the one who told the major consultancy they were wrong.

The cost shows up later. In the spend that has been reallocated against the wrong metric. In the growth that does not appear because the activities actually doing the work have been starved.

The audit was technically correct against the metric it had chosen. The metric was the wrong one. And the people who knew the metric was the wrong one were not the people in the room when the decision was made.