How the Monthly Report Works

There is one test that separates a report you can rely on from a report you are being asked to believe, and it takes ten minutes to run. Pick any figure on the page and try to re-derive it yourself, from an account registered in your own name, without asking anybody. If you can do that for every line, the report is an audit trail and its accuracy is your problem to verify rather than somebody's promise. If you cannot do it for even one line, that line is a claim, and a document made of claims is a newsletter with a retainer attached. Everything below is about building a report that survives the test.

Why is verifiability the test rather than accuracy?

Because you cannot assess accuracy from the outside, and you can assess verifiability instantly. A number you can check is one you never have to trust, which removes the entire question of whether somebody is being straight with you.

This matters more in a small business than in a large one, because you have no analyst to audit anything and no leverage to demand a rework. The document arrives, you read it, and either it stands on sources you hold or it does not.

It also removes an awkward conversation. Asking a supplier to prove a number sounds adversarial. Asking that every number name its source is a format request, and any competent shop agrees to it without friction.

The wider question of which numbers deserve a place at all is settled in the reporting piece. This one is about whether the ones you get can be checked.

Kingfisher Inn and Guide Services, TX

What does "an account in your own name" mean in practice?

That the advertising account, the search reporting, the analytics and the business listing are all registered to you, with the supplier granted access, rather than created in their account with you as a guest.

The distinction is invisible while the relationship is good and decisive when it ends. Access granted to you can be revoked. Access granted by you cannot.

Google's own guidance contains a specific version of this worth applying at the very start. When a shop offers an audit, it advises granting read access to your search reporting rather than write access at that stage.

Extend the same instinct across every account. It costs one email at the outset and it is the difference between a report you can audit and a report you can only receive, which is the wider audit set out in the ownership piece.

Which sources are genuinely free to check?

More than most guides realise. Search Console is free and reports which searches surfaced your site, the number of times it appeared, and how many people clicked through. On its own that covers most of the search half of any report.

The advertising side is verified from the platform's own billing screen, which is the only number in the whole document that cannot be argued about. Spend is spend.

The booking side is verified from your own calendar and your own bank, which nobody else can touch. If a report claims trips that are not in your calendar, that is not a discrepancy to discuss, it is a finished conversation.

Google also states plainly that appearing in organic results costs nothing and that advertising has no effect on organic placement, which is worth knowing before anybody presents the two as one blended achievement.

What is the difference between a measurement and a metric somebody built?

A measurement exists whether or not anybody reports it. A constructed metric exists because a supplier defined it, and it is only checkable if the formula is published alongside.

Both can be legitimate. A blended score that combines four real inputs is a reasonable summary, provided you can see the four inputs and the arithmetic. What is not legitimate is a score with no stated derivation.

The tell is whether the number has a unit. Dollars, clicks, days and trips are measurements. Points, scores and indices are constructions, and constructions need their recipe printed next to them.

Ask one question of any score on your report: what would make this go up. If the answer is specific and you can verify it independently, keep it. If the answer is vague, it is decoration.

Why does the field-versus-lab distinction matter here?

Because the same trap appears in performance reporting, and the platform that publishes the thresholds is explicit about it. Lab measurement is essential during development and is not a substitute for field measurement.

The guidance is explicit that a page behaves differently depending on the handset, the signal, what else the device is busy with, and what the visitor does once it loads. A simulated load contains none of that. Field figures come from people; lab figures come from software imitating them.

A report quoting a speed score from a simulated run is quoting the weakest available evidence, and it is the number most commonly screenshotted into monthly summaries because it is easy to produce and moves satisfyingly.

The same principle generalises across the whole document. Prefer what actually happened over what a tool modelled, and where only a model is available, say so on the page.

What should the report actually contain?

Five lines, each with its source named, and nothing else. What was spent. How many people arrived. How many made contact. How many booked. What that was worth.

Every one of those is a measurement with a unit and an account behind it. Together they describe the whole funnel, and any question worth asking can be asked of them.

What earns a sixth line is a leading indicator agreed in advance, for work that cannot show bookings yet. Search impressions for a foundation programme, for instance, agreed at the outset as the thing that should move first.

Everything past that is commentary, which is fine as commentary and must not be presented as result. The problem is not that reports are long; it is that length hides which lines are load-bearing.

What does a healthy report do that a decorative one does not?

It states what changed, what is being changed next, and what is expected from that change. Then the following month it says whether the expectation held.

That last part is the whole mechanism, and it is what almost no report does. A prediction recorded in advance and checked afterwards is the only structure that produces learning rather than narration.

It also protects the supplier, which is worth saying because this reads as adversarial otherwise. A shop that predicted a slow first quarter and was right has demonstrated competence more convincingly than one reporting a good month it did not forecast.

Ask for that structure explicitly at the start. Most shops will accommodate it and the ones that resist are telling you something before you have spent anything, as covered in the twelve questions piece.

Why do reports drift toward activity?

Because activity is always available and outcome is not. When bookings have not moved, a page still has to be produced, and the material that exists is the work performed rather than the result achieved.

There is measured context for the pressure this creates. The CMO Survey reports that under profit pressure most marketers focus on developing stronger performance tracking as the primary way to demonstrate value, alongside a shift toward short-term impact.

Scope that honestly, because those respondents lead marketing functions at companies far larger than a guide business. What it establishes is that reporting harder is the standard institutional response to being asked to justify a budget, which is a dynamic rather than a guide-specific finding.

Recognise it when it appears in your own inbox. A report that grew three pages in a quiet quarter has responded to pressure by producing more page, and the five lines are still the five lines.

How long should the read take?

Five minutes, and if it takes longer the document is wrong rather than you. A one-page summary with the five lines and three comparisons is the whole product; anything supporting it belongs behind that page.

The three comparisons that matter are against last month, against the same month last year, and against what was predicted. The middle one is the only fair test for a seasonal business and the one most reports omit.

Year-on-year comparison is not a nicety here, it is the only way to separate your decision from the weather. A quiet August against a busy July tells you nothing; a quiet August against last August is a signal. It also surfaces the weekday pattern that aggregate figures hide, which is the whole subject of the Tuesday problem piece.

If your report does not carry the year-ago column, ask for it. It requires no new data collection, only that somebody looks it up once.

Michael O'Brien Flats, FL

What should you do with the report once you have read it?

Make one decision, even when the decision is to continue unchanged. A report that produces no decision has cost you five minutes and told you nothing you will act on.

Continue is a legitimate outcome and should be recorded as one. Writing down that you chose to hold, and why, is what stops the same conversation recurring every month without resolution.

The decisions available are narrow and that is a feature. Continue, change one thing, pause, or stop. Anything more elaborate is usually a way of avoiding one of those four.

Keep the pages. Twelve reports in a folder is a year of evidence you cannot reconstruct afterwards, and it is what makes the following November's decision an informed one rather than a fresh argument.

Can you run this without a supplier?

Yes, and the do-it-yourself version is the same five lines in a spreadsheet. Nothing about the format requires anybody else, which is precisely why it is a fair standard to hold a supplier to.

Set aside twenty minutes at the end of each month. Open the billing screen, open Search Console, count the inquiries in your inbox, count the trips in your calendar, and write five numbers in a row.

Twelve rows later you have a year of data nobody had to be paid to produce, and you have also learned exactly how much work a monthly report actually is, which is useful context when one is being sold to you. That same arithmetic, applied to the fee rather than the format, is in the solo operator piece.

If you already do this, a supplier's report should tell you something your own row does not. If it tells you less, that is a finding. The measurement sequence to start from is in the diagnostic piece.

What belongs in the first report, before anything has happened?

The starting position, recorded once. Whatever the five lines read on the day work begins, written down before anybody touches anything.

Without that row, every later number is uninterpretable. A report claiming growth needs something to have grown from, and reconstructing a baseline three months in is guesswork dressed as history.

It is also the cheapest possible protection and it costs one afternoon at the start. Open each source, write the number, date the page, and file it.

Ask for it as the first deliverable rather than assuming it will appear. A supplier who begins work without recording where you started has removed the only means of judging them, whether or not that was the intention.

How should the report change as the work changes?

The five lines stay fixed and the leading indicator moves. That separation is what keeps a year of pages comparable while still reflecting what is actually being done.

During foundation work the leading indicator might be search impressions, because bookings cannot move yet and something has to be observable. Once advertising starts, the indicator becomes cost per inquiry, which is available within weeks.

What must not happen is the five core lines being redefined mid-year. A denominator that changes quietly is how a flat year becomes a growth story, and it is rarely deliberate; somebody simply starts counting inquiries differently.

Fix the definitions in writing at the outset. What counts as an inquiry, what counts as a booking, and whether revenue is gross or net. Twenty minutes of agreement prevents a year of ambiguity.

What do experienced guides do differently?

They spot-check one number every month rather than auditing everything. And they read the year-ago column first.

Spot-checking one line is enough because it is unpredictable. Pick a different figure each month, open the source, and confirm it. A supplier who knows any line might be checked behaves as though all of them will be.

Reading the year-ago column first changes what you notice, because it strips out the seasonal shape before your impression forms. Month-on-month movement in a seasonal business is mostly calendar, and reading it first anchors you to noise.

Experienced operators also keep their own row alongside the supplier's page. Where the two disagree, the conversation is short and specific, and where they agree, trust accumulates for free.

What are the common mistakes?

Accepting numbers from accounts you cannot open. Reading month-on-month in a seasonal business. Treating a long report as a thorough one. And never recording a prediction, so nothing can ever be wrong.

The account mistake is the structural one and it is set at the beginning of a relationship rather than discovered later. Everything created in somebody else's login is a number you will be asked to take on faith.

The no-prediction mistake is the one that makes a report unfalsifiable. Without a stated expectation, every outcome can be narrated as progress, and the document becomes an account of effort.

The length mistake is the easiest to correct. Ask for one page. A supplier who cannot summarise a month onto one page either has too little to say or is hoping you will not look closely, and the full inventory of how these relationships sour is in the agency exits piece.

What surprises people?

That the whole standard is one question. That most of the sources are free. And that producing the report yourself takes twenty minutes a month.

The twenty-minute figure is the one that reframes the purchase. You are not buying the report, which is trivially cheap to make. You are buying the work it describes, and the report is only the receipt.

Once that is clear, the format question settles itself. A receipt should be short, itemised, and checkable against your own records, and if it is none of those things you are looking at an invoice with adjectives. The choosing a marketer hub holds the rest of it.