The Monday Morning Nightmare for Marketers: Every Platform Claims It Closed the Sale

A couple of days ago, I grabbed dinner with a friend who works in marketing.

He ordered an Americano, then pulled out his phone to show me something — a screenshot of his Monday morning report.

The search platform said that of last week's 3 million in sales, 1.8 million came from search.

The short-video platform said, nope — 2 million of that was ours.

The e-commerce platform's attribution report told a different story altogether.

Add those three numbers up, and you get more than double the actual sales.

He said, "You know what this is called?"

I said, "What?"

He said, "This is called 'every platform claims it closed the sale.'"

I laughed. But after I stopped laughing, something didn't sit right.

Because this isn't just his problem. Almost everyone who works in brand marketing is being tormented by this.

Why Does This Happen?

Think about it — today's consumers, from the moment they first see a product to the moment they place an order, how many steps does that journey take?

Maybe they first scrolled past it in a short video. Didn't buy.

A couple of days later, they searched for it, checked the reviews. Still didn't buy.

Another three days go by, a friend sends a link. Finally, they buy.

So tell me — whose sale is that?

The search platform says: "They clicked through from search last. That's ours."

The short-video platform says: "We planted the seed. Without us, none of the rest would've happened."

The e-commerce platform says: "The purchase happened on our platform. Obviously, it's ours."

Everyone has a point. But the numbers don't add up.

This is what's known as the "attribution paradox."

The data is fragmented, the signals are disconnected, and every platform is talking past each other from inside its own silo. That weekly report sitting in your hands is, essentially, several contradictory reports fighting it out.

There's a Number That Surprised Me

I recently came across a set of data showing that agency partners are 35% ahead of brands themselves when it comes to key marketing capabilities.

Especially when it comes to "measurement."

35%. That's not a small number.

What does it tell us? It tells us that in this tangled mess of attribution chaos, there's a group of people who've already found their way out. They're not spinning in circles — they've figured out how to piece together fragmented signals and establish a single source of truth.

How did they do it?

I looked into it, and it turns out the whole thing comes down to three key moves.

Move One: Build a Solid Data Foundation First

No matter how powerful AI is, it digests whatever you feed it.

Feed it disconnected data and broken signals, and what it spits back out is a muddled mess.

So step one is transforming your data from "a tangled mess" into "a single unbroken thread."

What does "a single unbroken thread" mean?

It means your online ad clicks can be connected to users' offline behaviors — like whether they actually walked into your store, whether they left a trace in your CRM system.

Some agencies use a technology called Google Tag Gateway (GTG) to do this. In plain terms, it builds a bridge between your ad systems and your business data.

How well does it work?

A company called TRKKN helped DoYouSpain, a Spanish online travel platform, do exactly this. They used a server-side tag management solution to rework the entire data pipeline. The result? Observable conversions in Google Ads jumped by 11.7%.

11.7%. Think about that — it's not about spending more budget. It's about, with the same budget, finally "seeing" conversions that were previously invisible.

Then there's Kepler, who went even deeper for a U.S. B2B software client. They used their own data platform, connected it to Google's Meridian (an open-source Marketing Mix Model), standardized data across all channels, and fed it to AI as a unified stream.

The result? Cross-channel outcomes went from "guessing" to statistical certainty. Conversions were up 8% year over year.

Here's the fundamental truth: give AI good data, and it'll give you good decisions. Give it bad data, and it'll help you run faster in the wrong direction.

Move Two: Don't Just Look at "What Happened" — Figure Out "Why"

Data tells you "what happened."

Last week you spent 500,000 and got 20,000 clicks. Those are facts.

But do you know, out of those 20,000 clicks, how many would have come anyway — without you spending a dime? And how many were truly incremental, brought in only because you spent that money?

You don't know?

Then how do you judge whether that 500,000 was well spent?

This is the second thing agencies are doing — experimentation.

Going with your gut doesn't cut it. "I feel like this channel performs well" doesn't cut it either. You need to design real scientific experiments to validate your assumptions.

How?

Here's an example. Jellyfish discovered that some brands pour all their money into one or two channels, and after a while, returns go flat. That's because the same ads are being shown to the same audience, over and over — diminishing marginal returns.

They used their Now-Next-Soon platform to run Meridian's modeling framework, precisely diagnosing where the money stops working. Then they reallocated the freed-up budget to new growth engines, and validated the move with geo-based A/B experiments.

The result? ROAS was up 34%, and reach doubled.

Or take Dentsu. Display advertising has always had a problem: in last-click attribution reports, its contribution always looks disproportionately small, which makes brands afraid to invest. Dentsu ran a geo-based incrementality test in the UK, using CRM and behavioral data to balance control groups across different regions.

What did they find? Display advertising drove a 5.5% increase in revenue, with approximately 9x ROAS.

9x.

Then they cross-referenced this result against the brand's own Marketing Mix Model — two independent sources of truth, and the conclusions matched. That's what you call certainty.

Data tells you what happened. Experimentation tells you why. Without understanding the "why," your next budget decision is just a gamble.

Move Three: Weave All the Signals Into One Rope

You've cleaned your data. You've run your experiments. But if those results are scattered across half a dozen tools and reports from three or four different departments, it's still useless.

Because when the moment comes to make a decision, you're still staring at a pile of contradictory numbers.

So the final step is unification.

What does unification mean?

It doesn't mean getting everyone to look at the same report. It means bringing different models and different test results together, converging them onto one validated source of truth.

Google's open-source framework Meridian was built for exactly this purpose.

Power Digital spent months leading a brand through this transformation. They built a customized Marketing Mix Model on top of Meridian, designed a roadmap of experiments, and ran rigorous incrementality tests.

After they had clear conclusions, the brand reallocated budget toward high-impact strategies. The result? Incremental marketing revenue rose significantly, and total annual revenue grew by double-digit percentages year over year.

Hakuhodo DY Group rolled out Meridian on a much larger scale across the Asia-Pacific region. They integrated it directly into their analytics platform, transforming measurement from "static dashboards" into "dynamic ROI simulations."

For a beverage brand, they ran a frequency analysis on YouTube TV and found that after optimizing the ad frequency, ROI improved by approximately 10%.

For a healthy food brand, they reallocated budget from offline to online — order volume rose 6% to 11%, while customer acquisition cost dropped 5% to 10%.

Level Agency took an even more direct approach. They built a tool called Level Signal, tying Google Cloud together with search prediction signals. Instead of staring at "rearview-mirror reports," they look proactively ahead.

One client saw application submissions jump 48%, while simultaneously reducing the cost per application by 30%.

These numbers look impressive. But the logic underneath them is the same: weave fragmented signals into a single rope, find that one trustworthy source of truth, and then make decisions based on it.

So, Is Your Partner Ready?

The gap between "passively reading weekly reports" and "proactively driving growth" really comes down to a mindset shift.

How do you judge whether your current partner is up to the task?

Ask them three questions. That's all it takes.

First: Are we using a first-party data solution to feed AI clean data?

Second: Are we running scientific geo-based experiments to prove which investments are actually driving growth?

Third: Do we have a process in place to resolve the double-counting across different platforms?

If the answer to all three is "not yet," then you might want to rethink this relationship.

A good partner won't leave you staring blankly at three reports that don't reconcile in your Monday morning meeting.

They'll help you untangle the signals, design rigorous experiments, and unify your measurement.

And then, finally, you can say this in that Monday morning meeting:

"Of last week's 3 million in sales, search drove 800,000, short-video demand generation drove 1.2 million, and the e-commerce closed loop drove 1 million. They add up perfectly."

Doesn't that feel a whole lot better than before?

Here's wishing you a weekly report next Monday that actually adds up.