# The dashboard is green. The business is bleeding out.

**Written by JD Prater — 3x VP of Marketing.** This piece is part of *Open Folders*, which sits at the intersection of marketing science, consulting research, and practitioner experience, opening the parts of the discipline most marketers never got taught, so they can lead the whole of it, not just the slice they came up in.
Newsletter: https://jdprater.substack.com
Website: https://jdprater.com
Original post: https://jdprater.substack.com/p/the-dashboard-is-green-but-the-business

I have been in rooms where nothing looked obviously broken, but everything was getting harder.

CAC was up, but not enough to trigger panic. The channels were still working, just not like they used to. Every optimization gave back a little less than the one before. The team was good. The strategy had made sense when we wrote it. The dashboard still had enough green on it to keep the meeting moving.

The hardest moment comes before the cliff, when things have quietly stopped working as well as they used to and everyone in the room is trying to decide whether this is normal friction or the start of something worse.

I have seen what usually happens next. More budget behind the channels still showing signal. Tighter targeting. Faster reporting cycles. More pressure on the lower funnel. A stronger push for proof. Sometimes a new agency. Sometimes a new offer. Sometimes both.

None of those moves are stupid. Most of them look responsible. A few of them work just enough to delay the real conversation. The spiral is built from reasonable decisions.

## **The plateau is a demand condition**

The spiral does not begin when the budget gets cut. It begins earlier, at the plateau. Growth gets harder. The visible numbers still look mostly fine. The business responds by pushing harder into the same narrow slice of demand that made the plateau likely in the first place.

The plateau is different for every company. Some hit it early. Some later. It depends on the category, the rate of category growth, the size of the in-market pool, the length of the buying cycle, the intensity of competition, and how much predisposition existed before performance started carrying the load. A company selling low-ACV software in a fast-moving category will feel it differently from a company selling a higher-ACV product into a slower market.

The plateau is a demand condition, not a revenue number.

Think of it like a slow leak in a tire. The engine is not smoking. The dashboard warning lights are not flashing. The speedometer still reads 65. But your foot is pressing harder just to maintain the same speed. The leak is structural. The readout does not show it.

It is the point where your current capture system starts exhausting the accessible in-market pool faster than the business is building memory and predisposition among future buyers.

You are probably there when the same lower-funnel system needs more pressure each quarter just to hold the line. The channels may still look efficient in-period. The problem is that they need more budget, tighter targeting, or more incentives to produce the same output they used to produce more easily.

That is often the first warning sign.

## **Why teams misread it**

Most teams arrive there with tools, training, and reporting systems built to optimize the visible slice of demand, not diagnose the full system underneath it. Their dashboards are built for channel performance. Their reporting is built for short windows. Their budget conversations are built around visible proof. Their teams are trained to improve what shows up now.

So when returns flatten, they ask the questions those systems were designed to answer. Which channel is slipping. Which audience needs tightening. Which campaign needs more pressure. Which spend can be justified fastest.

The system can look like discipline while being too narrow to diagnose the full problem.

MMA's *How Marketers Make Decisions: The State of Marketing Analytics* (April 2026) surveyed 103 senior marketers across North America and found that most organizations are functional but not yet strategic in how they use analytics. KPIs are consistent, but mostly used for reporting. Only 29% could trace decisions back to a data-supported rationale. Just 22% described analytics capabilities as robust. The gap was less a technology gap than a leadership gap. The strongest teams had better governance, higher trust in their data, and clearer ownership of the metrics that mattered.

That matters here because the plateau is usually being judged inside a system that is better at reporting than diagnosis. The dashboard can tell you what happened inside the visible window. It struggles to explain the full growth system underneath it. That is why the spiral can persist inside capable teams. The organization has enough analytics to keep optimizing, but not enough diagnostic power to change the frame of the problem.

Win/loss analysis often surfaces the signal before the dashboard does. Buyers say they never had you on their shortlist. They say a competitor felt more familiar when they started looking. They say they had been following someone else's content for months before they ran an RFP. That is the demand condition making itself legible. Most organizations file it as anecdote, separate from the analytics review, and keep optimizing the channels.

## **The reasonable decisions that start the spiral**

That is when the next decisions start to stack.

More money behind the channels still showing signal. Tighter targeting. Shorter reporting windows. A stronger push on offers. Less patience for broad work that compounds slowly.

I have made some of those calls myself. I understand why people make them. When the board wants answers by Thursday and the dashboard still shows enough signal to justify one more push, the instinct is to move closer to the sale.

That instinct is rational. It is also where the spiral gets its momentum.

This is the sequence I have seen most often. Returns flatten. Pressure rises. Spend concentrates further into visible demand. Broad demand-building gets deprioritized. Lower-funnel channels saturate. More pressure is needed to hold the line. The economics get a little worse. Then the same system recommends more of the same.

You are probably entering the spiral when three things are true at once: lower-funnel channels still look acceptable in-period, each quarter needs more pressure to hold the same output, and the answer to softer growth is always some version of "push harder on what is still converting."

## **What gets quietly spent down**

The business thinks it is tightening discipline. Underneath, it is often spending down the conditions that made performance work in the first place.

By brand here, I mean the broad, memory-building work that increases the probability of being noticed, recalled, and considered before a buyer is in-market. That includes advertising, but it is not limited to advertising. It can also come through product experience, content, partnerships, distribution, and the repeated cues that make a company easier to recognize and easier to choose.

What gets spent down is not just brand in the abstract. It is the set of commercial advantages that broad demand-building supports over time. Predisposition. Base demand. Pricing power. Retention quality. Future conversion efficiency. The ability for lower-funnel spend to keep scaling.

This does not happen automatically every time performance spend rises. It happens when a business keeps asking lower-funnel activity to do more of the total growth job while broad memory-building work is being starved. Then the same capture system needs more pressure to produce the same output, and the commercial side effects start to show up elsewhere.

They show up in practical ways. More lower-funnel pressure usually means narrowing the audience pool, leaning harder on the same hand-raisers, and paying more to reach people who are already expensive to influence. More offer-led conversion can increase price sensitivity and weaken full-price behavior. More dependence on immediate response can reduce the flow of buyers arriving already predisposed to choose you. Over time, lower-funnel activity has to do more work because less work was done earlier.

Product marketing often feels this before anyone else names it. Launch investments get cut mid-cycle because pipeline attribution does not show the connection yet. Competitive positioning work, analyst briefings, customer narrative programs shape how a category thinks about a vendor before any RFP gets written. They lose budget because they cannot prove themselves in a weekly channel readout.

Performance data can show where demand was captured. It is much weaker at showing what made that demand available to capture in the first place. Nikki Lindgren, presenting Analytic Partners' work on this question, puts the distinction clearly. The visible metrics show where demand was captured. They do not show what materially increased the probability of that demand showing up for you.

A lower-funnel campaign can still convert well while the system underneath it is weakening. A business can still hit the number while becoming more dependent on incentives, more exposed to price sensitivity, and more expensive to grow next quarter than it was last quarter.

## **The dashboard tells the truth. Just not the whole truth.**

The dashboard is telling the truth.

Just not the whole truth.

It can tell you what converted. It can tell you what got credit. It can tell you what moved this week. It can tell you what looked efficient inside the current window.

It is much worse at telling you what is being spent down underneath.

Base sales erosion is hard to see in a weekly dashboard. Future demand starvation is hard to see in channel reporting. Weaker predisposition does not show up neatly in a campaign readout. Rising dependence on incentives can look like good execution right up until margins and retention tell a different story.

The dashboard shows response more clearly than contribution. It is built to observe response inside the reporting window. It is much weaker at isolating incremental contribution over longer horizons.

Attribution is useful for operational decisions. It is a poor governor of capital allocation if you ask it to answer questions about causality it was never built to answer. Nikki Lindgren's distinction applies here too. Attribution tells you what was present at the point of response. Incrementality asks what would not have happened without the activity. Those are different questions. Too many organizations use the first to answer the second.

Les Binet and Will Davis, presenting at the IPA Effectiveness Conference in 2025, put a number on this pattern. Since the pandemic, ROI has risen by 4 percent. Net marketing profit is down 11 percent in the same period. The efficiency metric improves. The business is bleeding out while the heart rate monitor says it is fine. They called it the death spiral: campaigns and the profits they generate shrinking together while the dashboard stays mostly green.

WARC's recent work describes a similar dynamic as a doom loop. Slowing growth encourages optimization against the most visible metrics. Budget shifts toward performance formats. Payback plateaus. The cycle repeats. In many organizations, nothing in that system triggers a correction until the damage is already meaningful.

## **The clock is wrong**

The long bet keeps losing because it is being judged on the wrong timeline.

Fast bets show signal now. Slow bets show signal later. When every line in the budget is evaluated on the same weekly or monthly cadence, the long-confidence bet loses most conversations by default. The reporting cycle is wrong for the job.

This is where a lot of smart companies go off course. They are not choosing between a good bet and a bad bet. They are choosing between a fast-visible bet and a slow-visible bet. One is easier to defend this quarter. The other does more to protect the next six. That difference gets flattened in most budget conversations.

Consider what this means for any campaign under review. A campaign that runs in January gets 100% of its cost booked in January. It may generate only 30% of its total commercial impact by the end of the month. The same activity that shows a $2 in-period ROAS can resolve to $4 by July and $5 by December once carryover effects settle. That distinction does not appear in a weekly channel readout. It shows up in the delta between what the dashboard said then and what the business grew to later. No one would plant an apple tree on Monday and dig it up on Friday because there is no fruit yet. Most organizations do the equivalent with their slowest-clock investments every quarter.

Weekly dashboards are built to help marketers make faster operating decisions. Over time, they have become capital-allocation tools by accident. Once that happens, long-payback investments keep losing to fast-payback ones, even when the slower investment is what protects future efficiency.

The evidence is not ambiguous. Stronger brand equity makes performance work harder. Overconcentration in performance reduces total returns, even if the magnitude varies by category and execution quality. WARC's *The Multiplier Effect* puts numbers on it: overinvesting in performance produces a penalty in the range of 20% to 50%, and shifting to a more balanced portfolio lifts revenue returns by 25% to 100%, with a median uplift of 90%. The cost of moving further in the wrong direction is a 40% decline in median revenue ROI.

The clock is wrong.

The practical answer is not to put every line of spend on a longer clock. It is to separate operating metrics from capital-allocation metrics, and stop letting the first govern the second.

Activation can have some longer effects, and broad demand-building can produce some short-term lift. The distinction is not a clean split by channel. It is the dominant job each investment is being asked to do. When that gets lost, the business starts starving the work that would have made later conversion easier, cheaper, and less dependent on force.

## **How the spiral ends**

I do not think the death spiral starts with bad decisions. I think it starts when a good team keeps asking a short-term system to solve a long-term problem.

That is why each decision inside the spiral still looks reasonable. The data supports it. The dashboard stays mostly green. The earnings hold for a while. And by the time the underlying pattern is visible, the business has usually spent longer than it realizes trading future strength for present efficiency.

How it ends is simpler.

One version is slow liquidation. The company keeps feeding the visible system harder and harder. Performance keeps getting the budget because it still has the clearest spreadsheet. Broad demand-building keeps losing because it takes longer to prove. The business mistakes asset depletion for discipline until the decline is too visible to explain away.

The other version is interruption. The business widens the lens before the damage compounds too far. It changes the measurement architecture. It separates attribution from incrementality. It stops asking a weekly dashboard to govern long-term capital allocation. It protects a baseline of broad, memory-building investment and then optimizes capture on top of it, instead of starving the former to flatter the latter. The work being done in the interruption version is building the conditions that make future capture cheaper: the recognition, the familiarity, the predisposition that means buyers arrive with you already on the shortlist before the auction starts.

That does not mean abandoning activation.

It means giving activation its proper job and protecting the work that makes activation cheaper later.

The death spiral does not begin with a bad call. It begins when a business keeps making good calls inside the wrong system.

---

**The Monday Version**

Pull your lower-funnel channel data for the last four quarters. If performance looks stable but the budget, targeting intensity, and offer pressure required to sustain it keeps rising, you may be at the plateau. If you run win/loss interviews, look for how often buyers say they started with a shortlist that did not include you. Both are telling you the same thing. The question to bring into the next planning conversation is not which channel needs more support. It is whether the current capture system is exhausting the accessible pool faster than the business is building into the one behind it.

---

Opening Folders, JD