A Monthly Marketing Data Reconciliation Routine for Small Businesses

Executive summary

At month end, the website report may show 120 sessions from a campaign, the inbox may contain 18 enquiries, and the sales sheet may show five purchases. Those numbers are not automatically contradictory. They measure different things, use different dates, and may include repeated visits or enquiries that never become sales.

Reconciliation means documenting why the totals differ and checking for avoidable gaps. It does not mean editing figures until every dashboard agrees. Allow 45–75 minutes per month for a small business with a few channels; a larger operation may need more time.

Write down the definitions first

Before comparing numbers, record what each system counts. A session is not a person; a form submission is not necessarily a qualified lead; an enquiry is not a sale. The Traffic acquisition report describes session-level source information, while User acquisition uses user-scoped dimensions. Google notes that these reports can show different values because their scopes differ, so do not compare them as though they were identical measures.

For each metric, define the unit, date basis, source system, inclusion rule and owner. For example, “new enquiries” might mean distinct enquiry records created during the month, excluding spam and duplicate submissions. “Completed sales” might mean paid orders recorded by the finance system, regardless of when the first enquiry arrived.

Use a reconciliation table

Measure	System of record	What to compare	Common explanation for a difference
Tagged sessions	Analytics	Campaign/source totals	Repeated visits, missing tags, attribution rules
Form submissions	Website or inbox	Valid submissions by date	Spam, failed notifications, duplicate forms
Phone or walk-in enquiries	Enquiry log	Recorded enquiries	Staff omissions, unknown source
Qualified opportunities	CRM or sales tracker	Status and creation date	Qualification timing, incomplete updates
Paid sales	Billing or finance system	Completed transactions	Purchase date differs from enquiry date

Do not add these rows together. They represent stages of a journey, not separate customers. A single person can appear in several systems.

Reconcile from the business outcome backwards

Start with the number that matters operationally, such as completed sales or qualified enquiries. Check that the source system is complete and that the date filter matches the reporting period. Then inspect whether each relevant record has an enquiry reference or campaign information that can connect it to earlier activity.

Work backwards only as far as the evidence permits. If the sale has no campaign identifier and the customer does not remember the source, record it as unattributed rather than guessing. An unattributed sale is a data limitation, not proof that marketing had no role.

For website forms, submit an approved test once after a meaningful site change and confirm that the record reaches the expected inbox. Do not create test submissions in production every month if doing so pollutes reports; agree how test records are marked and excluded.

Check dates and lag

A person may click an ad on the final day of a month, enquire the next month and purchase later still. Comparing clicks in October with sales completed in October can therefore understate campaigns with a longer buying cycle. Conversely, sales in October may come from enquiries generated earlier.

Choose a reporting approach and label it. A monthly operations report can group enquiries by creation date and sales by transaction date. A campaign cohort report can follow enquiries from a defined period to later outcomes. These are both useful, but they answer different questions.

Where offline conversions are imported into advertising systems, timing and upload rules can affect reporting completeness. Google’s current offline conversion guidance describes requirements and delays for imports; use the official setup documentation for the integration actually configured rather than assuming that a spreadsheet is automatically connected.

Investigate discrepancies with a short exception log

Record the difference, its likely cause, evidence, owner and next action. Examples include a missing notification, duplicate lead records, inconsistent campaign spelling, a sales status that was never updated, or a redirect that removed tracking parameters.

Prioritise issues by impact. A missing five-figure sale record is more important than a small naming inconsistency in a one-off post. Do not spend hours trying to resolve a difference that cannot be resolved from available evidence; document the limitation and improve the next period’s process.

Make the monthly report decision-focused

The final report should state what is known, what remains uncertain, and which action follows. For example: “Website forms fell, but call enquiries were not consistently logged; do not conclude total demand declined until call recording improves.” That is more honest and useful than claiming that a particular channel failed.

Use consistent date ranges, note any tracking changes, and avoid comparing partial months with full months. If a measurement setup changed, mark the break in the time series so readers do not mistake a change in collection for a change in customer behaviour.

Monthly checklist

• Confirm date range, timezone and metric definitions.
• Check analytics tags and campaign labels for obvious anomalies.
• Compare valid form submissions with received enquiries.
• Review duplicates, spam and missing status updates.
• Compare qualified opportunities and completed outcomes.
• Log unexplained differences and assign owners.
• Record tracking changes and limits in the report.

Conclusion and next steps

Reconcile to improve the process, not to manufacture agreement. Keep each system’s definition visible, use an exception log, and report uncertainty directly. A trustworthy report may contain unmatched totals; its value comes from explaining the differences and guiding the next improvement.

Official references

• GA4 Traffic acquisition report — consult the current documentation for product-specific behaviour.
• User acquisition vs Traffic acquisition reports — consult the current documentation for product-specific behaviour.
• Google Ads offline conversion import guidelines — consult the current documentation for product-specific behaviour.