Start with the one question
"What would this report have looked like if the month had gone badly?"
If the answer is that the same charts would have risen slightly less, the report cannot fail — and a report that cannot fail cannot inform a decision.
Ask it directly. The response, more than the document, tells you what you are dealing with.
Check it against your own experience
You have information no report does: roughly how busy the phone was, roughly how many enquiries arrived, roughly what closed.
If the report says forty conversions and you remember eleven enquiries, something is being counted that is not an enquiry. That single comparison catches the most common reporting failure — tracking firing on page load rather than on submission.
The five things to look for
Are calls counted? For a service business, a report covering only forms describes the smaller half of your leads.
Is branded search separated? People searching your business name were already coming. Counting them as marketing success overstates results.
What comparison is used? Month against previous month hides longer declines. Ask for twelve months and for the same month last year.
Do rankings come with search volume? Without it, first place for a term nobody searches looks identical to first place for one that produces work.
Is anything ever negative? Real months contain setbacks. A report that has never contained one has been edited.
What to ask for instead
One page: enquiries by source including calls, what closed, cost per booked job by source, what was done, what is next.
That request is entirely reasonable. If the answer is that the data does not exist, that is the real finding — and it is fixable within a month by adding call tracking and recording which enquiries became jobs.
The uncomfortable possibility
Sometimes vague reporting reflects missing data rather than an agency hiding. If nobody set up call tracking and nobody records outcomes, there genuinely is nothing better to report.
Worth establishing which you have before drawing conclusions, because the two need different responses.
The conversation is the real report
The most useful reporting is five minutes on the phone: here is what happened, here is what I think it means, here is what I want to do next, here is what I got wrong.
A provider comfortable with the last part is worth more than one producing forty polished pages, and you can assess that without understanding a single metric.
When to escalate
If after several months nobody can tell you what the work produced in enquiries by source, that is a legitimate reason to have a direct conversation — and if it does not improve, to look elsewhere.
Getting a second opinion
Bring your last report to a call on 832-338-2926. Twenty minutes usually establishes what it does and does not tell you.
Frequently asked questions
What is the single best question?
What would this report have looked like if the month had gone badly? If the same charts would simply have risen slightly less, it cannot fail and cannot inform a decision.
How do I check a report without understanding analytics?
Compare it against what you remember. If it reports forty conversions and you recall eleven enquiries, something is being counted that is not an enquiry.
What five things should I look for?
Whether calls are counted, whether branded search is separated, what comparison period is used, whether rankings come with search volume, and whether anything negative is ever reported.
What if the data does not exist?
That is a fixable problem rather than a dishonest one — call tracking and a record of which enquiries closed can both be in place within a month.