Eleven hours per client per month, spent assembling numbers nobody disputes.

Where the hours go

• Pulling data from three or four platforms per client, each with a different export format
• Reconciling numbers that disagree, which they always do, and deciding which one to present
• Formatting into whatever template that particular client prefers
• Writing commentary that is largely the same each month with different numbers in it
• Handling the follow-up questions the report generated, which a better report would have answered

Note that only the last two involve judgment. The first three are mechanical, and they are where most of the time goes.

Why reporting automation alone does not fix it

Every agency has tried a reporting tool. Most still spend significant time on reporting. The reason is that automation solves the assembly problem and leaves the two harder problems untouched.

Problem 1: the numbers still disagree

Meta reports one revenue figure, Google reports another, the store reports a third, and their sum exceeds actual revenue. An automated dashboard presents all three faster. It does not decide what to tell the client, which is the part that takes judgment and time.

Problem 2: a dashboard is not a report

Clients do not want a data feed. They want to know whether things are going well, what changed, what you did about it, and what happens next. A live dashboard link answers none of those, which is why clients keep asking for the deck anyway.

A report structure that reduces follow-up questions

1. The headline: one sentence on whether the month was good, bad or flat, stated plainly
2. The three numbers that matter for this client, with the comparison period, and no others
3. What changed and why — the actual causes, including the ones outside your control
4. What we did, with the reasoning and the result of each action
5. What we are doing next month, and what we need from the client to do it
6. Everything else in an appendix nobody is required to read

Most agency reports invert this: twenty pages of data followed by two lines of commentary. The commentary is the product. The data is the evidence for it.

The operational side: running many accounts without mixing them up

The other tax on agency time is account separation. Every agency has a story about a change applied to the wrong client's account, and the mitigations are usually procedural — a naming convention, a checklist, a rule about only having one tab open.

Procedural controls fail under time pressure, which is precisely when the mistake happens. Structural separation does not.

• Separate workspaces per client, with data that cannot cross between them
• Per-client permissions, so a junior can prepare work in one account without being able to apply changes in another
• An audit trail of who approved what, which is also what you need when a client asks why something changed in March
• The ability to reverse a change on a supported action, without reconstructing the previous settings from memory
• One login across all of it, because a password manager full of client credentials is its own risk

What to do with the reclaimed time

The commercial argument for fixing reporting is not that reporting is expensive. It is that the time goes into work the client cannot tell apart from work any other agency does.

An agency spending eleven hours per client per month on assembly and two on analysis is selling assembly. Reversing that ratio is the difference between an agency that competes on price and one that competes on judgment — and it is the only version of the business that survives clients getting better at reading their own dashboards.