Section order gets treated as a design decision, something to settle once the numbers are in and the logo is placed. It isn’t. Client report structure is a reading decision, and the reader is a marketing manager who will open the link on a phone, scroll for maybe forty seconds before the call, and form an opinion about the month from whatever happened to be at the top.
Most reports lose that forty seconds to impressions. Not because anyone decided impressions were the most important thing, but because impressions were the first metric column in the export, and the template got built in the order the data arrived. Platform order, column order, the order the media plan was written in: these are all defaults that came from somewhere other than how a client reads.
The fix is not a better chart. It’s deciding, once, what the first screen has to answer, and then building the template around that answer instead of around the sheet.
Mistake 1: opening with volume because volume comes first
A campaign export usually starts with the dimensions, then impressions, then clicks, then spend, then whatever conversion column the platform provides. Build KPI tiles left to right in that order and you have told the client that 4.2 million impressions is the headline of their month.
Almost no client cares. Reach and impressions matter as context for something else: whether the plan delivered, whether frequency got out of hand, whether a creative burned out. On their own they are a statement that money was spent, which the client already knew.
The practical test I use on a blank template: if the client read only the top row and closed the tab, would they have the number they would repeat to their own boss? For a lead gen account that’s cost per acquisition and volume of qualified leads. For a brand campaign it might be completed views and VTR against the target. For retail media it could be sales attributed against spend. Whatever it is, it belongs in the first tile, and everything else on the first screen exists to explain or qualify it.
That also means fewer tiles. A row of eight KPIs is a row with no hierarchy. Four is usually enough, and one of them should be the one that decides whether the contract renews.
Mistake 2: grouping by platform when the client is thinking in outcomes
A very common report section order goes: overview, then DV360, then Meta, then Google Ads, then YouTube, then a summary. It feels tidy. It matches how the media was bought, how the invoices arrive, and how the account team is organised internally.
It does not match the question the client is holding, which is usually some version of “did this work, and what do we do next month”. Platform sections answer “what happened inside each buying tool”, which is a different question and a much later one.
There is a second problem with leading by platform. It forces the client to do the addition. If Meta delivered a 2.10 euro CPA and DV360 delivered 6.80 euro, the client has to hold both numbers in their head and work out the blended picture themselves, which is exactly the moment someone decides the expensive channel should be cut without knowing it was doing the upper funnel work.
Keep the channel breakdown. Move it below the point where the report has already given the blended outcome, the reason behind it and the recommendation. By the time a reader reaches a platform table, they should be looking for evidence, not for the headline. Which metrics belong in each of those channel sections is its own question, and reporting by campaign type goes deeper on it than I will here.
Mistake 3: burying pacing three scrolls down
Pacing is the number most likely to generate an email, and it is usually the last thing on the page, tucked into a spend table near the bottom because that’s where budget data lives in the export.
Consider a campaign with a 20,000 euro monthly budget that has delivered 11,300 euro by day 22. Everything else in the report can be excellent and that single fact will still dominate the call, because underdelivery means either a refund conversation or a rushed catch-up week that tanks efficiency. Overdelivery is worse in a different way.
Pacing belongs on the first screen, next to the outcome metric, expressed against the plan rather than as a raw spend total. “11,300 euro of 20,000 euro, 73% of flight elapsed” tells the client something. “Spend: 11,300 euro” tells them nothing they can act on.
A client report structure that follows the reading order
The sequence that has held up for me, across both performance and brand accounts:
- Outcome. The one metric that defines success for this campaign, with the comparison that makes it legible: last period, the target, or the benchmark you agreed at kickoff.
- Cause. Two or three sentences plus the chart that supports them. What moved, and why. This is the section no template can generate for you, and the one clients quote back.
- Spend and pacing. Delivery against plan, blended efficiency, anything that changes the invoice conversation.
- What happens next. The actions you have already taken and the ones that need a decision. Put the decisions in a list so nobody can claim they were buried.
- Evidence. Channel tables, creative performance, geography, daily curves, frequency. The detail that supports everything above, for the reader who wants it.
Outcome, cause, spend, action, evidence. The first four fit on one screen if you are disciplined about tile count. Choosing what goes in that outcome slot is a harder question than it looks, and the metrics worth putting in a client report is a useful starting point when you are deciding between three candidates.
Where campaign type changes the order, and where it shouldn’t
The outcome metric changes constantly. A PDOOH flight reports on delivered plays and audience impressions, a BVOD buy on completion rate, a lower-funnel search account on CPA and conversion volume. That’s fine, and it is the main reason agencies end up with two or three templates rather than one.
What should not change is the sequence itself. Outcome before cause, cause before spend, evidence last, no matter what the channel is. A client who reports to a CFO and a client who reports to a brand director read in the same shape; they just care about different numbers in the first slot.
One genuine exception: an account in its first month has no comparison to lead with, so the first screen is better spent on delivery against plan and a short read on early signals. Say so explicitly rather than showing a change column full of dashes.
The reason this is worth an hour on a Sunday is that you only do it properly once. Screebl templates lock the layout once reports are built from them, so a section order you set deliberately in September is the order every client report inherits in October, without anyone re-deciding it at 22:00. That’s the whole argument for building the layout once and reusing it: the decisions worth making slowly get made slowly, and then they stop being decisions. If you want to see which plan fits the number of clients you report on, the pricing page is the quickest answer.
