The Anatomy of Your CRM
Reporting on the HubSpot Reporting exam starts with the CRM data model: every number on a dashboard is a count, sum, or average of records and properties on an object.
Objects, records, associations
HubSpot stores contacts, companies, deals, tickets, products, and custom objects as separate record types. A report either aggregates one object (deals, for example) or joins across associations (deals associated to contacts and companies). The primary object decides what a count means: 100 deals is not 100 contacts.
Properties determine what can be measured
- Number properties can be summed and averaged; text properties cannot be aggregated.
- Date properties power trends and date filters. Create date and close date answer 'deals this quarter' differently.
- Dropdown and radio properties group cleanly; free-text fields produce messy breakdowns.
- Empty values are excluded from averages, so a blank deal amount shrinks the apparent average.
Pipelines and lifecycle
Deal and ticket stages are properties; stage order is what makes a funnel a funnel. Contact lifecycle stage is a single property that is overwritten as a contact advances, so it shows current state rather than history. Lifecycle stage date properties and create dates preserve history.
Worked example
Question: how much closed-won revenue came from marketing-sourced deals last quarter? Primary data source: deals. Filters: stage equals Closed Won, close date equals last quarter. Join contacts and filter on the contact's original source. Metric: sum of amount. Reasoning: the amount lives on the deal, so deals must be primary; the source lives on the contact, so a join is required. One deal with three associated contacts can be counted three times unless you count distinct deals.
Traps
- Reporting contacts when the KPI is deals.
- Assuming lifecycle stage is historical.
- Forgetting that blank numeric values drop out of averages.
- Merged or deleted records quietly reducing historical counts.