A 41% miss on qualified leads is not usually what loses the account. Reporting bad campaign results ends relationships for a different reason, and it has far more to do with the order of the report than with the size of the number in it.
Take a hypothetical September. A lead gen client, €42,000 of media split across Meta Ads, Google Ads and DV360, a target of 1,200 qualified leads at a €35 cost per acquisition. The month closes at 710 leads and a €59 CPA. Pacing was clean, 98% of budget spent by the 30th. Nothing technically broke. The money went out and the leads did not come back at the rate anyone promised.
Accounts do get lost after months like that, but rarely because of the number itself. They get lost in the forty minutes between the client opening the report and the account manager’s phone ringing, and what happens in those forty minutes is decided almost entirely by the order in which things were presented and by whether the client discovered the drop alone. It is a sequencing problem before it is a communication problem.
The instinct to soften the number is the first mistake
Watch what an account team does to that €59 CPA before the report goes out. Someone suggests comparing against the full quarter instead of against the plan, because the quarter average is €44 and looks less alarming. Someone else moves the CPA tile to the second row and puts click-through rate on top, since CTR actually improved. A third person writes a commentary line that opens with “while CPA was impacted by increased auction pressure”.
Each of those moves is defensible on its own. Together they produce a report where the worst number in the month is the hardest one to find, and that is exactly what the client will notice. Marketing managers are not naive readers. Most of them have their own lead count sitting in their CRM, and they have already had a conversation with their own boss about why the pipeline looks thin. Your softened report arrives into a room where the bad news has already landed.
The cost is not that the client sees through the softening. The cost is that from that month onward, they read every future report looking for what you moved. That is a permanent tax on every report you send, and it is the quiet mechanism behind most client reporting retention problems. You rarely get fired for the miss. You get fired three months later for the feeling that the reports are managed.
Separate what happened from why, before you write a line of commentary
The second failure is blending. An underperforming campaign report usually contains sentences like “CPA increased to €59 due to rising competition and creative fatigue”. That is a guess wearing the clothes of a fact, and if the real cause turns out to be a tracking change on the client’s thank-you page, you have spent your credibility on the wrong story.
Write two lists before you write any client report commentary. The first contains only things you can point at in the data:
- 710 leads against a plan of 1,200, with €41,200 of €42,000 delivered.
- Meta prospecting delivered 38% of leads at an €81 CPA. Meta retargeting delivered at €24 on a much smaller budget.
- Google Ads search volume on the two main non-brand terms fell 22% versus August.
- Average frequency on the DV360 display line reached 9.4 by the final week.
The second list contains candidate explanations, each with an honest confidence level and, more importantly, what would confirm it. Search demand down 22% is a fact that supports a seasonality story. Frequency at 9.4 supports creative fatigue, but only if performance decayed over the month rather than being weak from day one, which is a thing you can check in an hour. A change in how the client’s sales team qualifies a lead is the explanation nobody in the agency can verify alone, and it is the one worth asking about directly rather than writing around.
Keeping the lists separate also protects you in the call. When a client challenges your explanation and you have already labelled it as the most likely of three, you are in a discussion. When you stated it as fact and it collapses, you are in a different conversation entirely.
Reporting bad campaign results in the order the client reads them
Here is the sequence that actually decides the outcome in the hypothetical September.
A client opens the report link on a phone, somewhere between meetings. They see the first screen. Whatever is on that screen is the month, as far as their reaction is concerned, and anything that needs a scroll or a second document arrives after their opinion has already formed. If your explanation lives in the email body and the number lives in the report, the client gets the number first in practice, because they click the link before they finish reading the email.
So build the first screen to carry both. The headline tile shows 710 leads against 1,200 and €59 against €35, stated plainly, with the miss visible rather than implied. Directly underneath it, in two or three sentences, sits what you know: prospecting carried the cost, non-brand search demand fell, here is the part you are still verifying. Then the breakdown by line item, so the client can see where the €41,200 actually went. Then what changes next month, with dates and owners.
That order is the opposite of the instinct, which is to build up context and arrive at the bad number having earned it. Clients do not read reports in build-up order. They read the top, decide how they feel, then look for support. The general case for this is worth reading in why section order beats design, and if you are still deciding which number deserves the top tile at all, the metrics every client report should show covers the choice.
This is one of the few places where the reporting tool genuinely changes the outcome rather than the effort. A PDF forces the split: chart in the attachment, explanation in the covering email, and the client reads them in whichever order their inbox decides. In Screebl, the report is a live page built from a reusable template, so a text block under the KPI tile that dropped is part of the report itself. The client sees €59 and reads why in the same glance, with no second document to find. If you have written a campaign brief for that report, the optional assistant on the shared link can answer the follow-up questions from that report while you are still in a meeting, which is usually where the first wave of questions goes anyway.
When you do not know the cause yet, say that, with a date
Telling a client bad news is manageable. Telling them nothing while you investigate is what turns a miss into a trust problem, because silence in reporting week reads as a search for an excuse.
The defensible version is specific about its own uncertainty. Something like: the prospecting CPA is the cause of the miss, we have three candidate explanations, we are checking lead-level data against your CRM on Tuesday and will confirm which one holds by Thursday. The client now has a date, an owner and a next action instead of a verdict. Most clients accept that far more easily than agencies expect, as long as the date is real and you hit it.
What does not survive is going quiet until the monthly call. By then the client has built their own explanation, usually a less generous one than yours, and the call becomes a negotiation about their theory rather than a review of yours. If your reports regularly trigger a round of questions anyway, clients reply to reports for structural reasons worth fixing before a bad month arrives.
The report sent on time protects the account better than the report that looks best
The last trap in a bad month is delay. The report slips from the 3rd to the 5th while someone rebuilds a chart and someone else waits for a quote from the media buyer. The client has already noticed, because they knew the month was weak and expected the number on the 3rd.
A late report in a good month is an inconvenience. A late report in a bad month is evidence, and the client will fill the gap with a story about what you were doing in it. Keep the delivery date fixed and let the content carry the uncertainty: “here is the month, here is what we know, here is what we confirm by Thursday” lands better on the 3rd than a complete account of the cause on the 6th. If your calendar makes the first week of the month structurally impossible, that is a cadence problem rather than a writing problem, and choosing a reporting cadence is the thing to fix. Clients who get a weekly line of sight almost never experience a monthly report as an ambush, which is the single biggest reason bad months stay survivable.
Then run the review as a review. Open the live report in the meeting, walk the client down the same sequence they already read, and make the decisions in front of the data instead of in front of a deck built to explain it away. A live dashboard changes that meeting more than better slides do.
The hypothetical September still ends with 710 leads. Whether it also ends with a renewal depends mostly on choices you make in the two hours before you send the link. If getting the explanation and the number onto the same screen is currently the hard part of your month, the plans are a reasonable place to start.
