What Should Be in a Monthly SEO Report?
Most reports are produced to demonstrate effort rather than to inform a decision, which is why they are long, automated and unread. A report should answer what changed, why and what happens next. Almost none of them answer the third.
Most Reports Exist To Prove Effort
Length and chart count get mistaken for value, by both sides. A supplier producing a substantial document feels they have delivered something. A client receiving one feels they have received something. Neither impression survives anybody reading it properly.
Why long is easier. Considerably less work.
A long automated report takes minutes to produce once configured. A short considered one requires somebody to read the data, decide what matters and write down what they think. The harder job produces the shorter document.
What length is standing in for. Reassurance.
A thick document says work happened. It is a proxy for effort in exactly the way this whole series warns against. Nobody ever names it as one.
Who it fails. The person paying.
An owner wants to know whether this is producing work. Twenty pages of charts do not answer that, so the document gets skimmed and eventually stops being opened at all.
What that leads to. Silence, then cancellation.
Nobody complains about a report they have stopped reading. The arrangement continues until somebody asks what they are paying for. By then the answer has been unavailable for months.
What this page sets out. The document.
What belongs in it, what does not and why. How to explain it in conversation is a separate matter, covered in the next guide.
Three Questions A Report Must Answer
What changed. Why it changed. What happens next. A document failing any of those three is a data export rather than a report, whatever it is called and however well it is presented.
What changed. The easy one.
Which figures moved, in which direction, against a stated comparison. Almost every report manages this, because it is the part the tools produce by themselves.
Why it changed. The one that needs a person.
An interpretation of the movement. Was it the work, the season, a competitor or a measurement change. No tool produces this and it is where a report starts being worth something.
What happens next. The one that gets left out.
What will be done about it, then by when. Block three covers why this goes missing so consistently and what its absence tells you.
How to apply the test. Read the last page.
Open any report at the end and look for a plan. If the document simply stops after describing the past, it has answered two questions out of three.
What the three produce together. A decision.
The reader knows where things stand, why, plus what is being done. That is the point of the exercise and it is achievable in a small number of pages.
Why the order matters. It is a chain.
The plan should follow from the interpretation, which should follow from the movement. A plan disconnected from what the figures showed is a list of activity rather than a response.
What Happens Next Is The One Left Out
Most reports describe the past and stop. The client's actual question is about the coming month. A report without a plan in it is the clearest single sign that reporting has replaced working.
Why it goes missing. It requires a commitment.
Describing what happened is safe. Stating what will be done next creates something to be measured against, which is exactly why it is valuable and exactly why it gets omitted.
What its absence indicates. Two possibilities.
Either nothing is planned. Or the plan exists and is not being shared. Both are worth asking about and the first is considerably more common than anybody admits.
What a plan looks like. Specific and short.
Named pages, named changes, a rough order. Not a strategy document and not a list of areas of focus, which is what usually appears when a plan is requested.
The version that does not count. Continue as before.
A plan saying that the current approach will carry on is not a plan. It is a description of inertia, appearing whenever nobody has actually looked at the figures.
What it does for the client. Something to check.
Next month's report can be read against this month's plan. That single connection turns a series of documents into an account of an ongoing job.
What it does for the supplier. The same thing.
It forces a monthly decision about what to do, which is the discipline that keeps work happening. A supplier who resists that is telling you something.
The question to ask. One sentence.
Ask what is planned for next month and why. The quality of the answer is a better indicator of an arrangement than any figure in the document.
What Belongs In It
Five things. A report containing only these will be shorter and more useful than almost anything currently being sent to businesses in this country.
Visibility and traffic, with a stated comparison. The context.
Whether you appeared for more of what matters and whether more of the right people arrived, each labelled with what it is measured against and where the figure came from.
What was actually done. The work.
A plain list of changes made to the site. Block six covers why this matters more than most suppliers realise.
What it was expected to achieve. The reasoning.
Each piece of work with a purpose attached. Without it, nothing can be evaluated next month by anybody, including the person who did it.
Enquiries or outcomes where tracked. The answer.
Whether the business received more of what it wanted. Where this cannot be tracked, saying so plainly is better than filling the space with something else.
The plan. The part that goes missing.
What happens next month and why. Block three, belonging at the end where a reader reaches it having seen the reasoning.
What ties them together. The commentary.
Written interpretation running through all five. Without it these are five sets of facts sitting next to each other.
What Does Not
Everything below appears in reports constantly and none of it helps anybody decide anything. Removing all five typically halves the length of a document without losing a single useful sentence.
Third party authority scores. Not a search engine measure.
An estimate produced by a company that sells software. Our guide covers what they are and why they do not belong at the top of anything.
Ranking counts with no context. A number with no meaning.
How many searches you appear for somewhere, with nothing about whether any of them produce work. It rises as coverage grows regardless of quality.
Raw keyword lists. Bulk without information.
Pages of search terms with positions beside them. Nobody reads them, they exist to occupy space and they make a report look substantial.
Screenshots of tool interfaces. Evidence of access.
A picture of somebody's software proves a login rather than a finding. If a screen contains something important, describe what it shows.
Anything that only ever rises. The comfortable one.
Any cumulative total goes up whatever happens, which makes it useless as a measure and reassuring to include. Our metrics guide covers the family.
The test for anything else. Would it change a decision.
If a figure moved substantially and nothing about the plan would change, it does not belong in the document.
Show The Work, Not Just The Numbers
A client paying monthly wants to know what was done. Most reports omit it entirely, which is remarkable given it is the thing being purchased. Listing it is what makes a monthly fee legible rather than mysterious.
What the omission causes. Suspicion.
An owner who cannot see what happened assumes less happened than did. Suppliers doing genuinely good work are regularly doubted purely because nobody wrote it down.
What the list should contain. Plain description.
Pages changed, what changed on them, what was published, what was fixed. Ordinary language, no interpretation, no attempt to make routine work sound impressive.
Why plain is better. It reads as competence.
Straightforward description of ordinary tasks is more convincing than technical vocabulary. Anybody dressing up small work sounds exactly like somebody dressing up small work.
What it protects. The supplier.
A recorded month of genuine work is a defence against the assumption that nothing happened. It is more useful to the person doing the work than to the client.
The uncomfortable version. A thin month.
Occasionally the list is short. Showing it anyway, with an explanation, is better than concealing it behind charts. The alternative is a document that never admits anything.
What to do with it. Read it against the plan.
Last month's plan and this month's work should resemble each other. Where they do not, that is worth a conversation.
Say When Something Did Not Work
A report where everything always went well is not believed by anybody sensible. Recording a change that did not deliver is precisely what makes the rest of the document trustworthy. Almost nobody does it.
Why it builds trust. It proves somebody looked.
A supplier reporting a failure has evidently examined the results rather than assembling favourable figures. That single admission tells a client more about the arrangement than a good month does.
What perpetual success signals. Selection.
If nothing ever underperforms, either the reporting is choosing what to show or nobody is checking. Neither is comfortable and both are common.
How to record it. Without drama.
This was tried, this was expected, it did not happen, here is what we think and here is what we are doing instead. Five clauses and no apology required.
What makes it possible. Stated expectations.
You can only report that something underperformed if you said in advance what it was meant to achieve. That is why block four insists on recording the purpose.
What it prevents later. The awkward conversation.
A client discovering a bad quarter themselves is far more damaging than one told about a poor month at the time. Small admissions prevent large confrontations.
What a client should do with it. Not punish it.
Treating an admitted failure as a fault trains a supplier to stop admitting them. The response worth giving is asking what happens next.
Length And Cadence
Shorter is better and we give no page count, because the right length is whatever it takes to answer three questions. Monthly is the norm and it is frequently too frequent for the comparison to mean anything, which is an awkward thing to say about a monthly report.
Why shorter wins. It gets read.
A document somebody finishes has communicated something. A longer one containing more accurate information communicates nothing if it is skimmed and filed.
The length test. Did you finish it.
Ask yourself whether you read the whole thing. If not, the report has already failed regardless of what is inside it.
Why monthly is awkward. Too short a window.
A month contains a great deal of ordinary variation and not much signal. The comparison problems are covered in our comparison guide.
What to do about that. Report monthly, judge less often.
Keep the rhythm because businesses need it, while making the year comparison the headline so nobody judges a quarter's work on a month's noise.
Contact is not reporting. A distinction worth making.
We commit to contact every three weeks, which is about the work rather than the numbers. A conversation and a document do different jobs.
The dashboard question. Also different.
A live screen answers whether anything broke. Our dashboard guide covers why it is not a report.
Automated Reports Are Not Reports
A scheduled export with no commentary is the thing that gave reporting its bad name. The figures can perfectly reasonably be automated. The interpretation of them is the entire product. That part cannot be.
What automation legitimately does. The assembly.
Gathering figures, arranging them and applying consistent comparisons. Genuinely useful, genuinely time saving and entirely uncontroversial.
Where it stops. Meaning.
No system can say why something moved, whether it matters or what to do about it. Those require somebody who knows the business. They are what a client is paying for.
How to spot a pure export. No sentences.
Look for written interpretation between the charts. If a document contains no paragraphs, nobody has looked at it before it was sent.
The second tell. Identical structure.
The same sections in the same order every month regardless of what happened. A considered report changes shape depending on what actually mattered.
Why it persists. It is cheap.
Automated reporting costs almost nothing to send and looks like a service. That combination keeps it alive across the industry.
What to ask for. A paragraph.
Ask for a written summary of what the month meant. Our guides on choosing an agency cover what a good answer sounds like.
Reporting Is Not Fixing
A beautifully presented decline is still a decline. Everything on this page is about making a report useful. A useful report is still only a description. If nothing changes on the website, the quality of the document is beside the point.
How the substitution happens. Slowly and reasonably.
Reporting is visible and schedulable. Fixing is uncertain and harder to demonstrate. Over enough months the visible thing quietly takes the place of the uncertain one without anybody deciding it should.
What makes this page relevant to it. Two of the five.
The work list and the plan are the only parts of a report that describe action. Their absence is why a report can be excellent and mean nothing.
The pattern to watch for. Improving reports.
A document getting better while the site stays untouched is the clearest version of the problem, because the improvement is being made to the wrong thing.
What to ask when you suspect it. Two questions.
What changed on the site this month. What will change next month. A supplier doing the work answers both without hesitating.
Where the position lives. The pillar.
Our measurement guide holds it. Everything sits on the performance guide.
Open it at the
end and look
for a plan.
Most reports describe the past and stop, while the client's actual question is about the coming month. A plan creates something to be measured against, which is exactly why it is valuable and exactly why it gets omitted. A plan saying the current approach will continue is not a plan either. It is a description of inertia, appearing whenever nobody has actually looked at the figures.
What goes in ours:
A report where everything always went well is not believed by anybody sensible. Recording a change that did not deliver is what makes the rest of it trustworthy.
Every guide.
One question.
What each number actually means, how to read the data without being misled, what belongs in a report and what does not, then how to connect any of it to the enquiries the business is actually paying for.
