When It All Goes to Hell

Someone pulls up the dashboard. The numbers are unambiguously bad, and have been for long enough that it’s not noise. A channel that was working three months ago is broken, or breaking, and everyone in the room is looking at each other.

What happens next is usually wrong.

The instinct, at every level of the organization, is to do something. Immediately. Visibly. The impulse to act is understandable: you’re paid to drive results, the results are going the wrong direction, action feels like the appropriate response. The problem is that the wrong action, taken quickly, doesn’t just fail to fix the problem, it makes the problem harder to diagnose, and sometimes makes it worse.

I’ve watched this play out more times than I’d like to count, across a number of channels over the years. The organizations that handled decline well were the ones who built the discipline to diagnose before they acted. This guide is about how to do that, at every level of the org.

Before Anyone Does Anything: Know What Type of Decline You’re Dealing With

Not all channel decline is the same, and the correct response depends entirely on what’s causing it. Different core causes require different approaches.

Platform or algorithm change

The channel’s operating environment shifted. Google updated its ranking algorithm. Meta changed its auction dynamics. Apple’s privacy changes eroded your targeting precision. The external system moved, and your results moved with it. You didn’t do anything wrong. The external environment of the channel changed.

Competitive dynamics

Someone else got better, or cheaper, or more aggressive. They’re outbidding you on paid terms, producing better content, or offering a more compelling product. Competitive dynamics aren’t always something that happens to you. Sometimes you created the competitive problem, through a pricing change, a product adjustment, or a repositioning that left your channel performance exposed. If you raised prices last quarter and your conversion rate dropped, that’s a competitive dynamics problem, and the solution might be commercial and more broad and strategic rather than tactical.

Internal execution failure

Budget was cut, creative went stale, a technical issue degraded landing page performance. The channel is fine, it’s the execution that’s broken.

Genuine demand shift

Customers stopped wanting what you’re selling at the rate they used to want it. This is the hardest one to accept and the one most organizations are slowest to name, because it implies the problem is upstream of the channel and may require existential self-evaluation.

Why the distinction matters

Treating a platform algorithm change like an execution failure wastes months on tactical fixes for a structural problem. Treating an execution failure like a demand shift leads to conversations about product strategy when the real answer is that someone adjusted bidding parameters. Before anyone takes any significant action, the first job is to figure out which of these you’re actually dealing with.

That diagnosis requires two to four weeks of clean data analysis, depending on the channel and the volume of the decline. It requires someone asking the uncomfortable questions: what else changed around the time performance dropped? What do competitors look like right now? What does the data look upstream of the metric that’s falling?

Here’s how each part of the organization should behave while it’s underway, and after.

If You’re the Individual Contributor

You’re closest to the data, which makes you the most valuable person in the room and also the most likely to make the situation worse by reacting too fast.

The characteristic failure at this level is pulling every lever simultaneously. When the numbers are bad and you feel responsible, the temptation is to adjust bids, refresh creative, restructure campaigns, and request a budget increase all at once. The problem is that you’ve now changed four variables and can’t isolate what, if anything, worked. The diagnostic process is over before it started.

Here’s what to do instead:

• Document what you’re seeing with specificity. Not “traffic is down” but “organic sessions dropped 34% week-over-week beginning March 12th, with the steepest decline in non-branded queries in the product category pages.” Precise problem statements lead to precise diagnoses.
• Pull the timeline. What else happened around the date the decline started? Platform updates, site changes, campaign adjustments, budget shifts, external news events affecting your category. Build a chronology before you build a theory.
• Make one change at a time, if you make any changes at all. If you believe you’ve identified the cause, fix that one thing, document what you did and when, and observe. Resist the urge to do more while waiting for the result.
• Check your peers. Is this happening to competitors? Tools like SEMrush, SimilarWeb, and industry forums can tell you quickly whether you’re alone or whether the whole category moved. If it’s category-wide, you’re probably dealing with a platform change or demand shift, not an execution problem.
• Communicate up on a defined cadence, not reactively. Pick a reporting interval (weekly is usually right for most channels during a decline) and stick to it. Send a consistent format: what the data shows, what you’ve investigated, what your current hypothesis is, what you’re doing about it, and what you need. Don’t send panicked updates every time you look at the dashboard.

What you should not do: promise a timeline for recovery before you’ve identified the cause. “I’ll have this fixed by end of month” sounds good in the moment and creates a credibility problem when month-end arrives and the situation is the same or worse.

If You’re the Channel Manager

Your job during a channel decline is not to fix the channel. Your job is to protect the team’s ability to diagnose and fix it, which mostly means managing the organizational pressure from above so it doesn’t disrupt the work happening below.

The characteristic failure at this level is performative responsiveness. When a senior leader asks “what are we doing about this,” the pressure to announce action is real. Announcing changes before there’s a diagnosis looks like leadership but It’s theatrical, and the team knows it.

Here’s what to do instead:

• Establish a diagnostic timeline and communicate it clearly upward. “We’re investigating and will have a clear picture by [specific date]” is a complete and professional answer. It sets an expectation you can meet, and it signals that you’re running a process rather than reacting.
• Give the channel manager air cover. When leadership wants to get involved in tactical decisions during a decline, your job is to manage that access. Input from above is fine. Directives that bypass the diagnostic process are not. You’re the buffer.
• Address the budget question explicitly and early. If the channel is declining significantly, someone is going to ask whether you should keep spending at the current level. Have an answer ready: under what conditions does it make sense to maintain budget (if you believe the decline is temporary and diagnosis is in progress), reduce it (if you have strong evidence the channel is broken and you’re still investigating), or reallocate it (if you have evidence the channel is structurally non-viable). Frame any budget reduction as a deliberate choice, not a concession.
• Watch for the false recovery. Declining channels often appear to stabilize before the underlying problem reasserts itself, sometimes because of natural seasonality, sometimes because a small tactical fix produced a short-term bump. Resist declaring victory until you’ve seen a sustained recovery over a period meaningful for your channel’s typical volatility. Announcing recovery and then having to walk it back is worse than taking longer to declare success.
• Keep a written log of the diagnostic process. What was investigated, what was ruled out, what was tried, and what the results were. This is useful if the decline continues and escalates, and it’s evidence that the team is running a professional process rather than guessing.

What you should not do: announce a plan of action before the diagnostic process has produced a finding. “We’re restructuring the campaign architecture and bringing in a new agency to evaluate” sounds decisive. If it’s not grounded in a diagnosis, it’s activity disguised as leadership, and the team will spend the next month executing a change that may have nothing to do with the problem.

If You’re the Director/VP of Marketing

Channel decline usually surfaces to you through one of two paths: someone flags it proactively as part of a normal reporting cadence, or the numbers are bad enough that a peer or executive noticed before you did. The second one means the normal reporting cadence isn’t working, and that’s a separate problem to address once the immediate situation is resolved.

The characteristic failure at this level is scope creep. When a channel is in decline, there’s an instinct to expand the response into a broader strategy review, a vendor assessment, a team restructure. Sometimes those are warranted. More often, they’re a way of doing something visible and significant without doing the harder work of diagnosing and fixing the actual problem. The strategy review consumes the team’s capacity at exactly the moment the team needs to be focused on operational diagnosis.

Here’s what to do instead:

• Establish clear decision rights before the team starts the diagnostic process. Who is empowered to make tactical changes to the channel without additional approval? What changes require your sign-off? What changes require budget authority above yours? Ambiguity here creates delays and diffuses accountability.
• Set a timeline with explicit escalation criteria. By what date do you expect a diagnosis? If the diagnosis reveals an execution problem, what’s the expected recovery window? At what point does the decline constitute a material business risk that needs to go to the CEO or board? Having these criteria defined in advance removes the judgment call under pressure.
• Frame the conversation upward carefully. There’s a version of talking to the CEO about a declining channel that maintains your credibility, and a version that doesn’t. The version that works: “We have a performance problem in [channel], we’ve identified it, here’s what we believe is causing it, here’s what we’re doing, here’s when we expect to have more certainty.” The version that doesn’t: vague reassurances that things are being looked at, or excessive detail about tactical activity that obscures whether anyone has a clear theory of the problem.
• Consider whether the channel’s target metrics are still the right ones. A declining channel sometimes reveals a measurement problem rather than a performance problem. If you’ve been optimizing for a proxy metric that’s no longer connected to business outcomes, this is the moment to fix that, not after recovery.
• Separate the channel question from the team question. If the decline is attributable to execution failure, there may be a personnel conversation at the end of the diagnostic process. That conversation should happen after the diagnosis, not during it. Conflating the two undermines the team’s ability to work clearly.

What you should not do: use a channel decline as the occasion for a broader strategy review, vendor assessment, or team restructure unless the diagnosis specifically supports it. Those conversations may be warranted eventually. Launching them during the diagnostic period pulls the team’s focus at exactly the wrong moment and signals that you’ve already drawn conclusions the data hasn’t supported yet.

If You’re C-Suite/Senior Executive

Your involvement in a channel decline should be minimal and specific. The instinct to get close to the problem is understandable, but executive involvement in operational diagnostics usually adds pressure without adding information.

The characteristic failure at this level is confirmation bias. A declining channel becomes evidence for something you already believed: that the team is under-resourced, that the strategy was wrong, that you need new leadership, that the company over-invested in this channel. Those things might be true. But a channel decline is not a reliable signal for any of them, and using it as one leads to decisions made for the wrong reasons.

Here’s what to do instead:

• Ask one question and wait for the answer: “When will you have enough information to make a recommendation?” Then hold that date, and don’t ask again before it arrives.
• Resist the urge to offer hypotheses. If you share your theory of what caused the decline, the team will feel pressure to investigate your theory whether or not it’s the most likely explanation. The problem is that an executive hypothesis doesn’t land as a hypothesis. It lands as a directive, one that the team will investigate what you suggested first, even if they have stronger leads, because ignoring the CEO’s theory carries more career risk than ignoring the data.
• Provide resources if asked, not preemptively. If the marketing team says they need additional data access, analytical support, or budget flexibility to investigate, that’s a legitimate ask. Proactively assigning people to the investigation or scheduling executive working sessions is usually a way of demonstrating attention that makes the actual work harder.
• Set a clear accountability window. If the channel is still declining by a specific date with no credible diagnosis, that’s a meaningful data point. Name that date in advance. It creates a reasonable deadline without manufacturing urgency during the investigation period.

What you should not do: use the decline as evidence for a conclusion you already held. If you believed before the decline that the marketing team needed new leadership, or that the company was over-invested in this channel, the decline will feel like confirmation. It might be. But treating a channel performance problem as a referendum on strategy or personnel before the diagnostic process has run leads to decisions made for the wrong reasons on the wrong timeline.

The Communication Problem: Silence and Noise

Most organizations handle the communication dimension of a channel decline badly. The two failure modes are silence and noise, and both are damaging.

Silence looks like the team going dark while they investigate, providing no updates until they have an answer, which comes too late and by which point leadership has filled the information vacuum with their own theories. Noise looks like daily updates, reactive Slack messages every time a data point moves, and status reports that contain a lot of activity language (“we’ve looked at X, we’ve tested Y, we’ve investigated Z”) without any actual progress toward an answer.

The communication cadence that works is weekly written updates during the diagnostic period, in a consistent format. What the data shows. What has been investigated. What the current hypothesis is. What is being done about it. What is needed. That format is boring, and that’s the point. It signals that there’s a process running, it gives leadership something concrete to respond to, and it removes the temptation for reactive escalation when someone checks the dashboard on a bad day.

When the diagnosis is complete, the update changes format: here’s what we found, here’s what we’re doing about it, here’s the expected timeline for recovery, here’s what we’ll do if we don’t see improvement by that date.

Closing Thought

Channel decline is a normal part of operating marketing programs across channels subject to competition, platform changes, and market dynamics. The organizations that handle it badly aren’t the ones with worse luck, they’re the ones that treat a diagnostic problem as an action problem, and spend months working hard in the wrong direction as a result.

The diagnostic muscle, knowing who owns what, how information should travel, when to act and when to wait, is far easier to build before you need it than in the middle of the emergency.