MVA / FIELD GUIDES / Marketing reporting
Good lead counts. Bad source data. Both can be true.
You can have a useful inquiry count and still lack the information to compare marketing channels fairly. Here is how I keep those two questions separate.
A useful count can come with an unfinished source question.
Your team needs to know how many inquiries arrived. It also needs to know how marketing contributed. Those questions depend on different evidence.
I check what counts as an inquiry, when the event fires, and whether it repeats. Where a comparable business record exists, I use it to check the analytics total. I review source information and the rules that assign channel credit separately.
THE OUTCOME
100Recorded inquiries
The analytics total matches the comparable accepted-request total in this example.
A SEPARATE SOURCE REVIEW
- A source is supported by the evidence
- 66
- Source still unresolved
- 34
Channel comparisons need that context.
Invented values, not a client result. This is a source-review example, not a GA4 attribution report. Matching totals do not validate every journey.
Similar source labels can answer different questions.
In GA4, First user source describes how an identified user was first acquired. Session source describes the source assigned to a visit. Source in key-event reporting shows channel credit for important measured outcomes, such as an inquiry event you have marked as a key event.
Changing GA4’s reporting attribution model changes credit in relevant key-event reports. It does not change First user source or Session source. Before comparing channel totals, name the actual field each report uses.
Then check how source information reaches the site: campaign links, redirects, and connected advertising platforms are useful places to look. Missing information can have several explanations. The label alone does not identify the cause.
Give the chart a short, specific note.
Keep the supported finding and the unresolved question together. The total may be useful for understanding inquiry volume while the channel breakdown still needs investigation.
An attribution report distributes credit under a chosen model. It does not, by itself, prove that a campaign caused an inquiry. If an estimate would help the decision, show its assumptions and keep it separate from observed results.
Match the next step to the decision.
A team checking inquiry volume may already have a usable number. A team reallocating its marketing budget needs to understand whether the channel comparison is strong enough for that decision.
I explain what the report supports, what remains uncertain, and which investigation would help. That gives the team something more useful than a corrected chart with no explanation.