the monthly report, properly.

most monthly reports are long because length signals effort. they are also unread for the same reason. the useful version is short enough to be read by someone with four other priorities and specific enough to change a decision — which means it contains an argument, not just numbers.

here is what should be in it, what should be underneath it, and what should not be there at all.

there is a commercial reason to care beyond tidiness. the report is how a marketing budget gets defended internally. a founder who can forward one page to a finance director and win the argument will keep funding the channel; one who has to interpret a forty-slide deck will quietly reduce it.

— 01the six things it must contain.

spend, including fees, so the total cost of the channel is visible. qualified enquiries, using a definition agreed with sales rather than whatever the platform counted. cost per qualified enquiry, with last month next to it.

what moved and why — three or four sentences of causation, not correlation. this is the part that requires actual thought and the part clients value most. what is blocked, named and dated, usually approvals or dev work on the client side. and what happens next month, specifically enough to be held to.

that is the whole report. everything else — creative breakdowns, platform detail, audience data, graphs — belongs in an appendix or a dashboard for anyone who wants to go deeper. the top page is for decisions.

the discipline that makes it work is writing the summary first. if you cannot state what happened and what you are changing in four sentences, the analysis is not finished, and no amount of charts will cover the gap.

— 02the four things that should not be there.

reach and impressions as headlines. they are diagnostics. leading with them signals that the enquiry number is not being tracked or is not good.

engagement rate as a success metric. it is a ratio that falls as reach grows, so a good month can look like a decline. useful for comparing creative, useless as a verdict.

platform-reported revenue presented as fact. attribution is generous, particularly on paid social, and reporting it without a reconciliation against your own sales data overstates results in a way that eventually breaks trust.

activity counts. forty posts, twelve hours of optimisation, six meetings. nobody buys because you did things. activity belongs in a scope document, not a performance report.

the tell for all four is that they only ever move in one direction. a report designed to reassure has no bad numbers in it, which makes it useless as a management document — you cannot make a decision from a page where everything is up.

— 03cadence and format.

monthly, on the first working day, in writing, with a call only if there is a decision to make. a report that arrives on the fifteenth is describing history nobody can act on, and a report that only exists as a talked-through deck cannot be forwarded to a finance director.

quarterly, do the longer version: what we believed three months ago, what we now believe, what we are stopping. that document is where strategy actually gets revised, and it is worth an hour of everyone's time. it should also name what did not work, in plain terms. a quarterly review with no abandoned ideas in it is describing a team that never tested anything.

weekly reporting is usually a symptom of anxiety rather than a management tool, and at most volumes it reports noise. the exception is a launch or a live campaign, where a short weekly note on delivery and creative performance genuinely helps.

whatever the cadence, keep the format identical every month. comparability is most of the value — if the metrics or the layout change each cycle, nobody can see a trend, and trends are the only thing a monthly report is actually good for.

— 04the test of a good report.

can a reader who missed the call tell what happened, why, and what is next? can it be forwarded to someone in finance without a translation? and did it tell you anything that changed a decision? those three questions are worth asking of your own reporting once a quarter, whoever produces it.

if the answer to the third question is no for three months running, the report is theatre — and the honest response is to change what is measured rather than to make the document prettier. that is a conversation worth having with whoever prepares it, including us.

one last thing worth insisting on: the person who did the work should write the summary. reports written by an account manager who was not in the ad account read as description. reports written by the person who made the decisions read as argument, and the difference is obvious within two sentences.

if this is the problem: reports clients actually read.