Smart Insights builds every table and report on top of a data model, which is the set of data the table draws from. This article explains what a data model is, how to read one, the three business areas the models are grouped into, and the field concepts that are easy to misread. Use it as a map before you build a custom report.
Before You Start
Smart Insights is a paid add-on. Contact your Account Manager to enable it.
Reporting permissions are granted by area (Accounting, Operations, Sales, Marketing). You can build only on the models for the areas your role can access.
Open Smart Insights from the left navigation sidebar, then click Create Insight to see the templates that sit on top of these data models.
What Is a Data Model?
A data model is the source of data a table or report is built on. When you click Create Insight and pick a template (like Opportunities Table or Job Payment Table), that template is wired to one data model behind the scenes. The data model decides which columns are available and what each row represents.
The most important thing to understand about any data model is its grain: what a single row stands for. For example:
Opportunities model: one row per opportunity (one deal or quote).
Jobs model: one row per job (the actual work performed; one deal can have several jobs).
Job Payment model: one row per payment.
Note: Choosing the wrong grain is the most common cause of a report that does not add up. If you sum job dollars on a model that has one row per payment, the totals will not match.
The Three Business Areas
Smart Insights groups its 20 data models into three business areas. Your reporting permissions decide which you can use.
Sales covers the pipeline: leads, opportunities, jobs, charges, and follow-up activity. Models: Opportunities, Leads, Jobs, Charges, Sales Activity, Follow-Up Activities. See Sales Data Models in Smart Insights.
Operations covers the work: crew assignments, materials, inventory, claims, storage, trucks, tasks, and employees. Models: Jobs Crew, Job Materials, Inventory Items, Claim Details, Claim Tasks, Storage Accounts, Storage Containers, Truck Utilization by Job, Tasks, Employees. See Operations Data Models in Smart Insights.
Accounting covers the money: job payments, storage payments, expenses, storage charges, invoices, and payroll. Models: Job Payment, Storage Payment, Expenses, Storage Charges, Storage Account Invoices, Payroll. See Accounting Data Models in Smart Insights.
Overview (composite) models: Each area also has an Overview template (Sales Overview, Operations Overview, Accounting Overview) that pre-joins several models so you do not have to connect them yourself. These are useful for dashboards that span more than one model. They expose more columns and can repeat a column name (like Company ID) that appears in more than one underlying model.
Key Concepts That Apply Across Models
These few ideas come up in almost every model. Understanding them once will save you from the most common reporting mistakes.
1. Estimated vs. Actual vs. Billed Figures
Many money and count columns come in more than one version, and they are not interchangeable:
Estimated: the figure the estimate calculated before the job runs.
Actual: the figure after the job is finished (meaningful only once the job is complete).
Billed: on jobs, what was actually charged to the customer.
Dollar totals come in Estimated and Actual versions (for example, Estimated Final Totals vs Actual Final Totals). A fourth label, Quoted, appears only on the Jobs model and only for the hourly rate, crew count, and truck count: it is the value quoted to the customer, which can differ from the system's estimate when someone adjusts the quote. Pick the version that matches your question. To decide whether a deal should show its actual or estimated figure, use its Status (Completed or Closed means actuals exist). Avoid relying on a flag like Has Any Job Been Finalized, which can turn true before actual figures exist.
2. Code Fields (Status, Type, Category)
Some columns store a category. Smart Insights shows you the label, but knowing the full set helps you filter. The most useful:
Opportunity Status: New Lead, Lead In Progress, Opportunity, Booked, Completed, Closed, Cancelled, Lost, Bad Lead.
Service Type: Moving, Packing, Moving And Packing, Load Only, Unload Only, Commercial, Storage In Bound, Storage Out Bound, Inner House, Junk Removal, Labor Only. If your company named its own service types, the custom name shows instead.
Move type (Opportunity Type): Local, Intrastate, Interstate.
3. Dates and Time Zones
Timestamps whose names end in "Utc" are stored in UTC (Coordinated Universal Time), the standard time zone Smart Insights uses to store data. Convert to your local time before you group by day, week, or month, or a late-evening record can land on the wrong day. Some date fields are blank on purpose: a job's Invoiced date is filled in only if your company uses the formal invoicing step, and a storage account's Move-Out date is blank until the account is closed. For a move or job's date, use Service Date or Job Date rather than these conditional dates.
4. Name Fields vs. IDs
Some columns show a person's name (Sales Person Name, Estimator, Move Coordinator, Assigned To) rather than a stable ID. They read well, but if two employees share a name, matching on the name can multiply rows. Use the Employees model's User ID when you need a dependable match.
5. Payments, Deposits, and Refunds
Payments are recorded against the opportunity, not a specific job, so on the Job Payment model the Job ID and Job Number are often blank. Join payments on Opportunity ID to capture every payment, including deposits. Refunds can post as a negative Amount, and each payment also carries an Amount Refunded column, so check both before you total a customer's payments. Full detail is in Accounting Data Models in Smart Insights.
6. Moves Only
Most models exclude "projects" (SmartMoving's grouping for large or long-distance parent jobs) and show move-related records only. If your company does not use projects, this has no effect.
FAQs
What is a data model?
A data model is the set of data a table or report is built on. It decides which columns you can add and what one row represents.
How do I know what one row means?
Check the grain. Each model in the area articles lists it at the top as "one row per ___."
Which model has the column I need?
Skim the area article for your topic (Sales, Operations, or Accounting). Each one lists its models, their key columns, and the rest of their columns.
Why do two models show different totals for the same thing?
Usually because of different grain or the estimated-versus-actual difference. A model with one row per payment totals differently than one with one row per job.