An SEO report should explain what changed, what was observed and which decision follows. Attractive charts alone do not establish whether the work answers the business's actual questions.
Ask what the percentage is based on
A reported increase of 50 percent needs a starting figure. Three enquiries instead of two produces the same relative change as 300 instead of 200, but carries different evidential weight. Ask for absolute values, the period and the data source.
Check whether the compared periods are genuinely comparable. Different numbers of working days, seasonal demand and changed measurement methods can distort a comparison. The explanation should identify these influences without using them as a blanket excuse for every disappointing development.
Separate searches containing the business name from other searches where the available data allows it. An increase in branded searches might, for example, coincide with another advertising activity. That does not make the visits worthless, but it changes how the result should be interpreted.
Ask to see completed work
Request specific pages and changes. “Content optimized” is harder to review than a list of edited service pages with an explanation of what was added. Open some examples yourself and verify their factual accuracy.
Distinguish technical completion from hoped-for impact. A redirect can be tested immediately. More relevant visitors are a later outcome influenced by other circumstances too. A report should not mix those two levels.
Also ask which task was deliberately postponed and why. This reveals whether priorities reflect limited time and budget or whether the report simply lists as many activities as possible.
End the discussion with a decision
Choose a small number of next steps with owners and dates. If a service page attracts visits but unsuitable enquiries, a clearer explanation of its scope may be the first useful change. More traffic would not automatically solve that problem.
Keep raw data exports and definitions of important metrics. They make later comparisons understandable even when the contact person or reporting tool changes.
A report is allowed to contain uncertainty. With a small dataset, “not yet conclusive” is often the correct assessment. In that case, ask for a specific question to observe and a date to review it rather than an apparently certain success story.
Unpack a numerical example together
Suppose a report lists 200 additional visits and five enquiries. First ask whether those enquiries could actually be linked to the visits or merely occurred during the same period. Both observations may help, but they should not be presented as identical measurements.
Then examine relevance. Were the enquiries within the service area and about services the business provides? A large number of unsuitable contacts can create work. Agree on simple criteria the business can apply without extensive person-level analysis.
Put costs in context too. A monthly fee may include both ongoing support and one-off work. Do not casually compare a month containing a technical rebuild with an ordinary maintenance month. Ask which tasks recur and which are complete.
If a finding remains unclear, record it as an open question with an owner. A useful report need not resolve every uncertainty immediately. It should prevent the same unresolved statement from reappearing each month as though it were an established fact.
